Page 71 – Practical Gastro

There is a robust body of evidence for the etiology and management of adult gastroparesis, but limited in the pediatric population. Pediatric gastroparesis is usually overlooked and can remain untreated for a long period of time. The aim of this review is to provide the most up to date evidence on the spectrum of pediatric gastroparesis, emphasize the differences from the adult setting as well as extensively address management approaches and treatment recommendations.

Gastroparesis is characterized by delay in gastric emptying in the absence of mechanical obstruction. The etiology and management of gastroparesis have been well studied in adults, but limited in the pediatric population. Most common identifiable etiologies of pediatric gastroparesis include: post-viral illness, drug side effects, post-surgical complications, diabetes mellitus, and mitochondrial disease. The most common symptoms are usually age-dependent. Nausea and abdominal pain are more common in older children and adolescents, while vomiting is more common in younger children. The gold standard for diagnosing gastroparesis remains gastric emptying scintigraphy, although normal values in children are limited. Treatment includes dietary modifications, pharmacotherapy, and gastric electrical stimulation, maintenance of nutrition, attention to glucose control, and psychological aspects.

Eduardo D. Rosas-Blum¹* Essam M. Imseis²* Richard W. McCallum³ 1Paul L. Foster School of Medicine, Texas Tech University Health Science center at El Paso. Department of Pediatric, Division of Gastroenterology, Hepatology, and Nutrition 2University of Texas Health Science Center at Houston, Department of Pediatric, Division of Gastroenterology, Hepatology, and Nutrition ³Paul L. Foster School of Medicine, Texas Tech University Health Science Center at El Paso, Department of Internal Medicine, Division of Gastroenterology, Hepatology, and Nutrition *These authors contributed equally to this work.

INTRODUCTION

Gastroparesis is characterized by delay in gastric emptying associated with upper gastrointestinal symptoms without mechanical obstruction.¹ The care for these patients is often complex, difficult, and frustrating.² In adults, the most common etiologies are secondary to complications from diabetes or surgical interventions³ as well as “idiopathic”. The gold standard for the diagnosis of gastroparesis is a scintigraphic gastric emptying study.⁴

There is a robust body of evidence for the etiology and management of adult gastroparesis, but limited in the pediatric population. Pediatric gastroparesis is usually overlooked and can remain untreated for a long period of time.^5,6 In children, the most common identifiable causes for gastroparesis are secondary to viral illness or complications of a surgical intervention.⁷ The aim of this review is to provide the most up to date evidence on the spectrum of pediatric gastroparesis, emphasize the differences from the adult setting as well as extensively address management approaches and treatment recommendations.

Etiology and Pathophysiology

The normal gastrointestinal function is complex and depends on coordination between the smooth muscles, enteric, and central nervous systems. The most common gastrointestinal dysfunctions associated with adult gastroparesis include: impaired gastric accommodation, postprandial antral hypomotility, pyloric dysfunction, duodenal dysmotility, dysfunction of the autonomic nervous system, and visceral hypersensitivity.⁸ However, in infants and children there are developmental aspects which are superimposed on these recognized abnormalities.

Delayed gastric emptying occurs very frequently in premature infants^9-11 (< 28 weeks gestation) as the normal gastric emptying gradually matures with age. At 32 weeks gestation, the gastric emptying patterns are similar to older infants, children, and even adults.1,¹⁰ In normal term infants, expressed breast milk leads to faster gastric emptying compared to formula. Also, larger volume feedings are associated with a slower gastric emptying rate.¹⁰

In two large pediatric studies examining the etiology of gastroparesis, no recognized cause was found in up to 70% of cases (idiopathic). The identifiable causes included: viral gastroenteritis (18%), drug side effects (18%), post-surgical complications (12.5%), mitochondrial disease (8%), and diabetes mellitus (2%-4%).¹ Recognized pathogens associated with post-infectious gastroparesis include: parvovirus- like agents, Lyme disease, and rotavirus.¹² In the vast majority of post-viral gastroparesis, the emptying delay tends to improve or resolve spontaneously over days to several months.^12,13 At the time of presentation, identifying a virus may be problematic but a preceding gastroenteritis-like illness can be elicited from the family. Therefore, this raises the question whether so called “idiopathic” gastroparesis represents a latent post- gastroenteritis neuronal injury; this type of gastroparesis also spontaneously resolves. In adults, there have been many reports of histological abnormalities associated with gastroparesis which include: depleted or reduced numbers of interstitial cells of Cajal,^14,15 degeneration of the myenteric plexus combined with loss of ICC¹⁶ myopathic gastroparesis,¹⁷ stomach muscular layer eosinophilia,¹⁸ and lymphocytic myenteric ganglionitis.¹⁹ There have been no similar reports in pediatric gastroparesis based on our research of the literature.

Post-surgical gastroparesis has been described as a complication for antireflux surgery.²⁰ The main cause is an accidental injury to the vagus nerve which is an infrequent complication. Infrequently, patients with autoimmune diseases (i.e. systemic scleroderma) can present with delayed gastric emptying.^7,21 Mitochondrial disorders are often associated with intestinal dysmotility disorders. Screening for mitochondrial disorders involved serum levels of lactate or pyruvate, while confirmation can be accomplished by sequencing of the mitochondrial DNA or/and muscle or liver biopsy. In a small study looking at patients with mitochondrial disorders, delayed gastric emptying had poor response to prokinetic therapy.²² Patients with hypertrophic pyloric stenosis have been extensively followed up. To date, there is no evidence of increased gastroparesis following the surgical correction (pyloromyotomy).^23,24

Pseudo-obstruction is a rare but well described entity (familial or acquired) characterized by deficiency in the smooth muscles or nerves of the gastrointestinal tract. Gastroparesis is present in some of these patients as part of the diffuse involvement of the gastrointestinal tract, and particular attention needs to be paid to assessing the gastric emptying in children being considered for a subtotal colectomy for refractory constipation, in addition to histologically assess the resected colon smooth muscle histology.

Pediatricians often encounter infants with regurgitation or “reflux”. Gastroesophageal reflux (GER) of infancy is a normal physiologic event that requires no therapy and generally improves over time. Empiric treatment of GER for these infants is generally not recommended if this GER is without complications, but a subset of infants with GER may also have significantly delayed gastric emptying for which prokinetic therapy may be helpful.²⁵

Other important distinction from gastroparesis is cyclic vomiting syndrome (CVS), which is characterized by intense and stereotypical episodes of emesis. Patients with CVS usually are asymptomatic between episodes, without complains of abdominal pain or postpandrial distress, and have normal gastric emptying.

Clinical Symptoms of Gastroparesis

In children, presenting symptoms of gastroparesis appear to be different from those noted in adults.^5,26 In infants with gastroparesis, vomiting appears to be the most prevalent presenting symptom while both vomiting and abdominal pain are more commonly noted in young children between 1 and 10 years of age. Like adults, adolescents report more abdominal pain and nausea than younger children, but the incidence of nausea in adolescents still remains less than that which is noted in adults with gastroparesis. Nausea is quite common in adults with gastroparesis ? noted in more than 80% ? and is more common in diabetic gastroparesis than idiopathic gastroparesis. Vomiting is also more commonly noted in adults versus adolescents and is noted in 60% and 90% of idiopathic and diabetic gastroparesis in adults.²⁷ Other symptoms present in children with gastroparesis include early satiety and weight loss which are present in approximately 25% of children of either gender. Bloating is less common and reported in fewer than 10% of children. One apparent reason for this age-related symptom difference in children may be the inability of infants and young children to express and describe symptoms of abdominal pain and nausea.

Gender differences also appear to be more pronounced in adults with gastroparesis. In adults, gastroparesis is predominantly noted in females which comprise approximately 80% of patients.²⁸ When analyzed as a single group, children with gastroparesis appear to be nearly equally divided between male and female gender. However, male-female incidence appears to change with age at diagnosis. Gastroparesis appears to be slightly more common in male children (<12 years of age) who comprise over 61- 72% of patients in this age group.5, 26 Adolescents with gastroparesis tend to be overwhelmingly female and comprise approximately 3/4 of patients in this age group. This is consistent with current theories related to explaining female susceptibility to gastrointestinal motility disorders in adults through the hormonal changes experienced after puberty.

The presence of comorbid conditions has been noted in children with gastroparesis. Unlike adults, non- psychiatric comorbidities are commonly noted in children diagnosed with gastroparesis with 38% of children suffering from some other major neurologic disorder? seizure disorder, cerebral palsy, developmental delay, or prematurity. Psychiatric symptoms have been noted in adults with gastroparesis with higher depression and anxiety scores on psychological testing and rates of depression exceeding 60%.^3,29 In one study in children, only 28% of children with gastroparesis were noted to suffer from psychiatric disorders which included attention-deficit hyperactivity disorder, depression, anxiety, bipolar disorder, or other behavioral problems.⁵

Based in our literature review, an association between gastroparesis and pervasive developmental disorders, attention deficit and hyperactivity disorder (ADHD), or Down’s syndrome in children is not evident. Rumination syndrome is an entity that needs to be clearly differentiated from nausea and vomiting of gastroparesis, and gastric emptying in rumination is generally normal. Finally, there is an overlap between adolescents with eating disorders (anorexia, bulimia) and gastroparesis, and in this setting, identifying the diagnosis and instituting appropriate treatment is a challenging problem.

Diagnosis

Gastroparesis is a condition of delayed gastric emptying without evidence of mechanical obstruction and is typically associated with symptoms of nausea, vomiting, abdominal pain, early satiety, or bloating. It is imperative that mechanical obstruction (i.e. hypertrophic pyloric stenosis, intestinal webs, malrotation, duodenal atresia, anular pancreas, etc.) be considered and ruled out when necessary, particularly in young children or children with severe or significant symptoms.

Gastric emptying may be recognized as early as the 12th week of gestation possibly coinciding with increased amniotic fluid volume and the development of the suck reflex in utero. The percentage of fetuses that demonstrate normal gastric emptying also appears to increase in frequency in late gestation.³⁰ Gastric emptying appears to play a critical role in the developmental process and gastroparesis can be suspected if there is an abnormality in the normal process of growth or developmental milestones.

The gold standard for diagnosing gastroparesis remains gastric emptying scintigraphy. In this test, the solid or liquid contents of a test meal are radiolabelled so the amount of radiolabelled food remaining in the stomach at specified time intervals can be used to compute the rate of gastric emptying. A recent consensus statement in adults favors a 4 hour gastric emptying scintigraphy scan over a shorter 2 hour scan since the sensitivity of the test appears to improve with a 4 hour duration.³¹ While consensus exists regarding adult normative values, normal values in children are limited. A small study in infants and children revealed a gastric emptying of 32-64% one hour after ingestion of radiolabelled milk and 44-58% in children receiving radiolabelled milk feedings.³² In children 5-10 years of age, the time to empty half of a child-friendly Rice Krispie™cake technetium 99m-radiolabelled meal was 107.2 minutes. Most pediatric centers use consensus- defined adult normal values published using a standardized meal consisting of egg-whites, toast, jelly, and water or as appropriate for age. Using this adult protocol, gastric emptying is defined as normal if less than 90%, 60%, 30%, and 10% of the test meal remains in the stomach 1, 2, 3, and 4 hours following ingestion. In children as in adults, vomiting or an inability to ingest the test meal and its radiolabelled tracer must be noted since either of these may affect baseline and residual tracer counts noted during scintigraphy.

Wireless capsule motility testing (SmartPill™) is another modality that has been used to detect gastroparesis. This device is an orally ingested 26mm X 13mm non-digestible pill taken following ingestion of a standard test meal. b5 measure luminal pressure, pH and temperature throughout the entire GI tract and is thereby able to quantitate rates of gastric emptying, small bowel transit and colonic transit. While this device is currently FDA approved for use in adults, it is not approved for use in children. There are inherent challenges in this age group as the smart pill is difficult to swallow in younger children and normal values are not well-established. Nevertheless, there is limited data supporting its use in children. In a small study of 22 symptomatic children age 8-17 years old, wireless motility capsule testing was well tolerated with no adverse events and had 100% sensitivity and 50% specificity in detecting gastroparesis when compared to a 2 hour scintigraphic gastric emptying study.³³ Its major asset is being able to measure small bowel and colon transit thus providing a total profile of gut transit to assist in therapeutic decisions without use of radiation. In this limited pediatric study, the capsule was also found to have greater sensitivity in detecting abnormalities of small bowel motility when compared with antroduodenal manometry, perhaps because the capsule asseses the entire small bowel actively.

Breath testing is another means of assessing gastric emptying in children. In breath testing, orally ingested food is enriched with naturally occurring 13-carbon. A number of 13-carbon enriched substrates are utilized such as radiolabelled 13C-octanoic acid for solids and 13C-sodium acetate for liquids. After ingestion, these 13-carbon substrates are rapidly metabolized in the liver upon leaving the stomach and following their absorption in the duodenum. After oxidation in the liver, these isotopes are excreted from the blood into the exhaled breath. The rate of gastric emptying is the rate limiting step in excretion of this compound in exhaled breath.³⁴ As a result, quantitative measurement of this 13-carbon dioxide can help in determining the rate of gastric emptying. Multiple studies have validated its use in adults. Small studies in children of various ages seem to support a role for breath testing in analysis of gastric emptying in children. For example, use of radiolabelled 13C-octanoic acid breath testing did reveal good correlation with scintigraphy in assessment of gastric emptying of solids in 25 children 5 to 10 years of age.³⁵ Use of breath testing in small studies of both premature infants and term infants also demonstrates a potential role for 13-carbon breath testing with reproducibility and correlation with other methods including scintigraphic testing.^36,37 The major benefit is that the patient does not have radiation exposure and also it can be repeated multiple times to assess the response to treatment. This method is now FDA approved for use in adults only. In gastroparesis, where there is accompanying small bowel bacterial overgrowth (SIBO) in more than 50% of patients, there is a concern that this SIBO can interfere with metabolism and absorption of 13C-octanoic acid thus changes the calculations required for assessment of gastric emptying by breath testing. Hence its future role in adults with gastroparesis remains unclear; for now is it not approved in children.

Treatment

The management of gastroparesis can be complex (Figure 1). Initial efforts should correct fluid and electrolyte abnormalities since correction of these derangements can assist with management of gastroparesis.^38,39 Hyperglycemia is often noted in exacerbations of diabetic gastroparesis. Hyperglycemia can delay gastric emptying and contribute to symptoms of gastroparesis in both idiopathic and diabetic gastroparesis.⁴⁰ It is accepted that acute hyperglycemic states (serum glucose >250mg/dl) will delay gastric emptying and is a major contributor of nausea and vomiting in the newly diagnosed diabetic where an infection (often urinary) may be the trigger. Correcting this hyperglycemia may therefore also enhance gastric emptying. Chronic gastroparesis related to diabetes takes up to 5-10 years to evolve.

Dietary Management

Dietary modification is a key first step in the management of gastroparesis. In children as in adults, small frequent meals may be helpful in managing gastroparesis. Low fat, low fiber foods can also be helpful in managing gastroparesis since fat and fiber can retard gastric emptying and since fiber may be associated with an increased risk of bezoar formation in individuals with gastroparesis. Blenderized foods and ingestion of liquids during meals may also be helpful since gastric emptying of solids may be slower than liquids. A study in adults with diabetic gastroparesis revealed that a low particle size diet consisting of “foods” that were mashable with a fork” resulted in improvement in symptoms as well as more significant improvement in rates of gastric emptying when compared to a control group taking a low fat, low fiber diet.⁴¹

Dietary management of infants can be difficult due to their dietary restrictions, but there may be some strategies that may be helpful. In infants who are predominantly fed commercial formula, a small study of 6 infants revealed a significant difference and more rapid gastric emptying when those infants were fed infant formula containing high medium-chain triglyceride versus high long-chain triglyceride formula and when those infants were fed glucose-polymer containing formulas versus lactose-containing formulas.⁴² In separate studies, gastric emptying in infants fed breast milk appeared to be more rapid when compared to infants fed commercial formula.³⁷ There are also some studies that revealed an increased rate of gastric emptying in infants and non-infants when fed a formula containing whey protein or whey hydrosylate, although data regarding protein content in infant formula and rates of gastric emptying are conflicting.^43-45

Pharmacological Management

Since nausea and vomiting are common symptoms in children with gastroparesis, antiemetics may be helpful. Studies supporting their use appear to be limited, and recommendations for antiemetic therapy appear to be largely based on anecdotal experience and the adult literature. Furthermore, while use of antiemetic agents may result in improvement in symptoms in patients with gastroparesis, they do not appear to have beneficial effects on gastric emptying.³⁹ since their main source of action is centrally with the chemorector trigger zone at the floor of the 4th ventricle. Ondansetron, a 5HT-3 antagonist and antiemetic, can be helpful in management of vomiting associated with gastroparesis or other pediatric gastrointestinal disorders, such as cyclic vomiting syndrome, and doses of 4-8 mg every 8 hours are generally recommended in children and adolescents and can be administered orally, intravenously, or rectally if necessary.⁴⁶ Promethazine can be used in children but only with caution. The FDA issued a 2004 black box warning that recommended use of promethazine only in children greater than 2 years of age and only with the lowest effective dose since there have been several reports of respiratory depression and death with use. Furthermore, a more recent FDA warning recommended that injectable promethazine be administered only via the deep and intramuscular route and not into the skin or artery since there may be a risk of gangrene with these routes of administration. Oral and rectal routes of promethazine administration are also treatment options. Other over-the-counter antiemetic with some antiemetic effect include: meclizine, ginger, peppermint drops, and some homeopathic remedies.

With continued symptoms of gastroparesis unresponsive to conservative measures, use of prokinetics may be helpful. Only a few agents exist that can accelerate gastric emptying in children, and there are problems that can arise with use of most of these agents. Since erythromycin is a potent stimulator of gastric contractions, it can be utilized as an “off label” use of the agent. The main mechanism of action is to occupy the motilin receptor in the stomach and mimic the action of motilin. Lower doses of erythromycin are recommended in a range of 1-3mg/kg every 6 to 8 hours, since this can delay tachyphylaxis which occurs with chronic erythromycin use and usually occurs within 4 weeks of starting the medication.⁴⁷ Symptoms of nausea and vomiting can actually be provoked with higher doses of erythromycin most commonly used for infections (10 mg/kg/dose). Prolongation of the QTc and interaction with other inhibitors of cytochrome P-450 3A have been reported to result in cardiac arrhythmias and even death in individual case reports in adults. Azithromycin may theoretically be used as an alternative since it has fewer drug interactions, less incidence of QTc interval prolongation, a longer half- life, and fewer gastrointestinal adverse effects.^48,49 The dose of azithromycin is twice that for erythromycin. Data supporting its use is lacking at this time. Ingestion of oral azithromycin and erythromycin should be used with caution in infants if the exposure occurs in the first 2 weeks of life as the possibility of increasing the risk of developing infantile hypertrophic pyloric stenosis has been reported.⁵⁰

Metoclopramide is an FDA approved drug for gastroparesis in adults but not in children, although it is commonly used in management of children with gastroparesis. While there is a large amount of experience using metoclopramide in adults, the published evidence supporting its use in children is limited. In a small study of 6 post-surgical infants with gastroparesis, metoclopramide more than doubled the rate of gastric emptying but no beneficial effect of metoclopramide was noted in premature infants. There remains some concern over adverse events accompanying long- term use of metoclopramide including a potential risk of tardive dyskinesia and Parkinson-like syndrome. Akathisia, anxiety, hyperactivity, tremor, and sleepiness can develop in the first few days to months after treatment is initiated. These symptoms are completely reversible after decreasing the dose or stopping the medication. A recent 2009 FDA black box warning recommended metoclopramide only for short term use less than 12 weeks. A recent 2015 Canadian federal health warning recommended that metoclopramide should not be used in children less than 1 year of age since they appear to be at higher risk of extrapyramidal symptoms The Canadian recommendation further states that metoclopramide not be used in children greater than 1 year of age unless treatment is clearly necessary. The Canadian report states that the Canadian health department (HealthCanada) has identified only 8 reports of extrapyramidal symptoms suspected of being associated with metoclopramide in children receiving the recommended daily dose. However, the Health Canada warning cites a recent review of the European data that found cases of EPS in children less than 18 years of age treated with metoclopramide with most cases occurring when recommended doses were used. As a result, caution should be emphasized when considering this agent for children, and follow up is essential for monitoring for side effects. The tardive dyskinesia adverse events may be irreversible in some cases.

Less commonly used medications may be necessary when symptoms persist. Domperidone can enhance gastric emptying and may be considered for use in children. An FDA administered IND must be obtained prior to administering this agent. Domperidone is useful where chronic prokinetic therapy is being contemplated or when metoclopramide and erythromycin have resulted in side effects or are not effective. It is a dopamine 2 receptor inhibitor, as is metoclopramide, but it does not cross the blood-brain barrier decreasing the risk for extrapyramidal side effects. Its effects are antiemetic acting centrally and as a prokinetic acting peripherally. Dosing is similar to metoclopramide in adults, starting at 10 mg four times a day but its benefit is lack of significant side effects and dosing can be increase up to 80 mg a day in adolescents. An electrocardiogram needs to be followed to address the rare reports of prolong Q-T intervals.

Baclofen is a γ-Aminobutyric acid (GABA)?B
receptor agonist which increases lower esophageal
sphincter pressure and decreases the transient lower
esophageal sphincter relaxations. In children, Baclofen
significantly improved the gastric emptying compared
to placebo in a trial involving 30 children.⁵¹ In adults,
Baclofen is commonly used for refractory GERD but
there are no randomized trials for its use in gastroparesis.
The baclofen dose in adults is 10 mg four times a day,
while in children is 0.5 mg/kg/dose to a max of 40 mg
a day.

Other agents can also be helpful for management of associated abdominal pain accompanying gastroparesis. Low-dose tricyclic antidepressants or cyproheptadine are also potentially helpful since they have been utilized in other functional abdominal pain disorders such as irritable bowel syndrome and cyclic vomiting syndrome.

Interventional: Surgical and Endoscopic Options

Placement of a gastrostomy tube for venting or jejunostomy for feeding may be helpful in severe cases. Parenteral nutrition may be necessary in severe and refractory cases to help in maintaining nutrition. However, a jejunostomy tube is the recommendation due to severe TPN complications and costs.

Gastric electrical stimulation (GES) has emerged as a reasonable alternative in those with refractory symptoms of gastroparesis or where oral medications are not tolerated or are ineffective; this occurs in approximately 25% of adult patients with gastroparesis. With few good pharmacologic options and with the concerns over adverse drug effects, this approach has recently garnered more support. The history of gastric electrical stimulation and pacing date back to the 1960’s.⁵² In 2000, the FDA approved the Enterra system for humanitarian use in which a surgically implanted pacer delivers high frequency electrical pulses to electrodes placed at the junction of the antrum and body. A recent meta-analysis in adults revealed that gastric electrical stimulation appeared to result in significant improvement in symptoms of gastroparesis in individuals with diabetic and non- diabetic gastroparesis. Improvement in gastric emptying is not a goal.⁵³ Guidelines published by the American College of Gastroenterology in 2013 recommend this approach only for compassionate use in adults or children with refractory symptoms, particularly nausea and vomiting. Some uncontrolled data does exist supporting the use of gastric electrical stimulation in children. In a study of 9 children aged 8-17 years, all 9 children reported symptom improvement and quality of life improvement following gastric electrical stimulation during a follow up of 8 to 42 months.⁵⁴ In another study in 16 children age 4-19 years, there was significant improvement noted in all children with improvements in severity of both nausea and vomiting.⁵⁵ As noted in adults with idiopathic gastroparesis, there was not clear improvement in gastric emptying noted in those children who received gastric electrical stimulation which is consistent with the mechanism of action of GES which is to affect central control of nausea and vomiting via vagal afferents. It is important to remember that abdominal pain is not a target for GES since the main goal is to improve nausea and vomiting.

Surgical intervention may be necessary in cases of severe, symptomatic gastroparesis that appears to persist despite aggressive nutritional and pharmacologic intervention. Use of enteral feeding tubes may be necessary to maintain or improve nutrition, hydration, or metabolic derangements. Placement of these tubes may also allow venting for patients with excessive gastric distention or enteral secretions with simultaneous feedings as in instances where gastrojejunal feeding tubes with gastric and jejunal access are placed. They also permit the administration of medications via the enteral tube which improves the absorption of prokinetics and antiemetics, as well as “other agents” (i.e. antiseizure, pain medication, etc). Current recommendations by the American College of Gastroenterology favor a trial of nasoenteric postpyloric feedings prior to jejunostomy feeding tube placement in individuals with weight loss of more than 10% or refractory symptoms of gastroparesis.³⁹ In 2 large pediatric series, surgical placement of either a gastrostomy or jejunostomy tube was required to aid with management in a small percentage (4%, 19/469) children.^5,26 In one of these series, all five children requiring surgical feeding tube placement were noted to have CNS comorbidities and developmental delays complicating their management.⁵ A large study in adults with gastroparesis reported improvement following jejunostomy tube placement with 39% reporting improved nausea and vomiting, 52% fewer hospitalizations, 56% with improved nutrition, and 81% with improvement in overall health status.⁵⁶) Enteral feeding devices offer good reversible and, in some cases, temporary measures for treating children with severe, complicated gastroparesis who are unresponsive to more dietary or pharmacologic conservative intervention. It should be emphasized that when the patient has reached the stage of requiring a feeding tube, it will be advisable to consider a GES placement for symptomatic control of the nausea and emesis. Also, a jejunal feeding tube approach is preferable from a percutaneous endoscopic Gastro- Jejunal tube placement as it facilitates maintenance and most importantly, smooth muscle biopsy can be obtained at the time of the surgical jejunostomy to assess the ICC and neuronal status.

Intrapyloric injection of botulinum toxin has also been investigated as a potential treatment for gastroparesis due to encouraging results in multiple small series of patients with gastroparesis. A small open-label retrospective study of intrapyloric botulinum injection in children with gastroparesis revealed that approximately two-thirds of children reported improvement in a variety of symptoms and 40% of responders requiring only one injection. The uncontrolled and open-label nature of this study are obvious limitations, and similar encouraging findings have also been noted in uncontrolled open-label adult studies. Two double-blind, placebo-controlled studies examining the effects of botulinum injection in adults with gastroparesis reveal no improvement in symptoms compared with placebo.^57,58 Based on these placebo- controlled trials, botulinum use is not recommended for children or adults with gastroparesis.39 While the use of intrapyloric injection of botulinum toxin is not approved or endorsed, there is interest in its future role since it could be a first-step where a good response to botulinum injection could predict if a pyloroplasty can be beneficial in the long term.

Pyloroplasty has also been used in the management of gastroparesis and has the advantage of less radical alteration of the gastric anatomy compared with gastrectomy.⁵⁹ In one large adult series of 50 patients with gastroparesis, 83% reported symptom improvement following laparascopic pyloroplasty, and there was also significant improvement in gastric emptying noted on scintigraphy with median preoperative T1/2 of 180+/- 73 minutes and postoperative T1/2 of 60+/-23 minutes (p < 0.001). 68% of these patients had previous foregut procedures and/or cholecystectomy and 64% underwent concomitant procedures, such as paraesophageal hernia repair and gastrostomy takedown, at the time of their pyloroplasty. More recently in adults, pyloroplasty is being combined with GES placement with excellent results since gastric emptying can be normalized, a goal not achieved by GES alone.⁶⁰

Partial or subtotal gastrectomy is also a potential treatment for severe, refractory gastroparesis, but larger studies supporting its use are largely confined to adult populations. In one series of seven adults patients with vomiting due to gastroparesis, subtotal gastrectomy with removal of 70% of the stomach and creation of a Roux-en-Y loop of jejunum to prevent reflux gastritis resulted in substantial subjective improvement in all but one of seven patients.⁶¹ A larger series of 62 adult patients with postvagotomy gastoparesis who underwent near-total complete gastrectomy with a Roux-en-Y reconstruction revealed symptom relief in 43% of patients. A high percentage of postoperative complications were also noted in this larger study in 40% of patients and included narcotic withdrawal syndrome (18%), ileus (10%), wound infection (5%), intestinal obstruction (2%), and anastomotic leak (5%). There was also significant reduction in nausea (93% to 50%), vomiting (79% to 30%), and postprandial pain (58% to 30%) following surgery, but there were not significant differences in chronic pain, diarrhea, and dumping syndrome in this study. Gastrectomy should therefore be reserved only for those with severe symptoms unresponsive to other interventions. Data from McCallum et al. indicates that about 3-5% adult patients who failed GES for gastroparesis will require a total gastrectomy. If the patient has a previous gastric surgery (Billroth I or II) or GIS tumor resection, a GES is not recommended as the best approach but rather they should undergo a subtotal gastrectomy.

TAKE HOME CLINICAL PEARLS FOR THE PRACTITIONER

Pediatric gastroparesis is not similar to the adult
gastroparesis. The diagnosis and treatment for
gastroparesis is well established in adult but there
is limited evidence-base literature in the pediatric
gastroparesis.
1. 2. The most common etiologies in pediatric gastroparesis are idiopathic and post-viral. Most gastroparesis in children tends to resolve spontaneously without any treatment implying that many “idiopathic” gastroparesis could be subclinical gastrointestinal infection cases. Also, treating the underlying pathology (infection, hyperglycemia, etc.) will improve the gastric emptying.
2. 3. Symptoms for pediatric gastroparesis include: nausea, vomiting, and abdominal pain. Gastric emptying scintigraphy remains the gold standard to diagnose gastroparesis in children despite the lack of normal values in pediatric patients.
3. 4. The treatment for pediatric gastroparesis should start with dietary modifications. Pharmacological therapy is the second and third line of therapy for persisting pediatric gastroparesis. A GES is the choice of the treatment for those cases refractory to medical therapy.
4. 5. Future research in the etiology and treatment for persistent gastroparesis is very much needed in the pediatric population, particularly focusing on histological changes in neurons, ICC, and smooth muscle of the gastric muscularis propia.

Foreward

May 2015 • Volume XXXIX, Issue 5

Richard W. McCallum

Richard W. McCallum, MD FACP, FRACP (AUST), FACG, AGAF, Professor of Medicine and Founding Chair, Department of Internal Medicine, Director, Center for Neurogastroenterology and GI Motility, El Paso, TX

Hello colleagues. I certainly hope the first six contributions in this series, The Clinical Spectrum of Gastrointestinal Motility and Functional Bowel Disorders for the Practitioner, have enhanced your knowledge base, improved the care provided by us and our staff in our practice and given us confidence that we are gaining the credentials to warrant the term “Card Carrying Motility and Functional Gut Expert”. Armed with this new knowledge let’s expand our horizons even further.

Next, we will reach out to our Pediatric Gastroenterology colleagues, who sometimes can feel a little neglected since they are not being specifically catered to by adult GI articles. We address this deficiency with two articles: The first, “Pediatric Biliary Dyskinesia and Sphincter of Oddi Dysfunction” and the second, “Update in Pediatric Gastroparesis”. These cover a large spectrum of GI issues in children and we will continue to re-visit pediatric GI in future articles.

We then turn our attention to “late breaking news”. Since it is an entertainment term, like the Brian Williams saga in the television world, much news can be overstated. Such is the case with Domperidone and the “Doom and Gloom” over less than well-founded and documented statements on cardiac toxicity that I see bandied around. We review this agent in great detail and conclude that it remains the “champagne” of the antiemetic/prokinetic world. We strongly endorse its efficacy and safety and recommend much broader use in your practices.

We then turn our attention to a series of three articles covering the topics of “Gas and Bloating” — “Integrative Medicine” and wait for it…”Herbal Medicine” where we will provide clinical pearls. Sometimes referred to as the “scourge” of our practice, gas and bloating of undetermined origin has a rational algorithm of diagnosis and treatment, the former relying on the judicious use of glucose and fructose breath testing and the latter on knowledge of treating small bowel bacterial overgrowth, utilizing low fructose, FODMAP diets, and food allergy concepts. Here we are under the expert guidance of second year GI Fellow, Dr. Juan Castro.

We then turn to integrative and holistic medicine with one of my GI Fellows, Dr. Christine Yu, who found time in her second year of GI Fellowship to do online training and certification in this field. Balance is the key – apparently to everything – but definitely in regard to brain / gut balance. The challenge is how to integrate “old world” wisdom with new world technology and actually surprise our patients with “old school” genuine patient involvement. Combined with this article I have asked one of my colleagues, a unique individual with a Ph.D in herbal medicine, Dr. Armando Gonzales – Stuart, to teach us how herbal medicines are not “bad news” based on concerns for liver and other toxicity, but actually have a host of great things to offer our patients.

Here at Texas Tech El Paso, on the border with Mexico we appreciate the rich Hispanic heritage and culture. The common use of food supplements in Hispanic families reminded me of the benefits of this approach when balanced with the patients’ own unique personal medical diagnoses and ways to complement our prescription medications for symptoms. Now we can all learn about this aspect of over the counter use for relief of symptoms in our patients.

The Paul L Foster School of Medicine is flourishing here at Texas Tech, El Paso. We graduate our second medical school class in May, at the same time as I am presenting you our second installment of this series.

A final word. I plan to see that great rock band The Eagles in El Paso. Remember the line from their song Hotel California: “you can check out any time you like, but you can never leave”. So you are now committed. Please stay in the classroom of GI Motility and Functional Bowel Disorders with the University of Practical Gastroenterology.

A CASE REPORT

Acute Liver Failure Due to Gemcitabine

William Ferges,Jeanine Chiaffarano,

S. Devi Rampertab,Nakechand Pooran

Gemcitabine induces apoptosis during DNA replication and has been used as a chemotherapeutic agent in various types of cancer. This drug is generally considered to be well-tolerated with mild common side effects including gastrointestinal disturbance, flu-like symptoms and rash. Myelosuppression and abnormalities of liver enzymes, usually mild, transient and reversible, are frequently reported. Gemcitabine has rarely been identified as a cause of acute liver failure (only four cases of cholestatic hepatotoxicity have been reported). This case involves a patient with acute cholestatic liver failure after adjuvant treatment with gemcitabine for pancreatic adenocarcinoma.

William Ferges, M.D., Jeanine Chiaffarano, D.O., S. Devi Rampertab, M.D., Nakechand Pooran, M.D. Rutgers – Robert Wood Johnson University Hospital, New Brunswick, NJ

CASE REPORT

A 48 year-old woman presented to the emergency room with several weeks of intermittent epigastric abdominal pain. She was found to have obstructive jaundice with a mass in the head of the pancreas. Endoscopic ultrasound revealed localized disease and she underwent an uncomplicated Whipple’s resection. Pathology confirmed a moderately differentiated T3N0 adenocarcinoma with local invasion into the duodenal wall and peripancreatic soft tissues but without lymph node involvement or encasement of the celiac axis or the superior mesenteric artery. Following the surgery, the patient’s liver enzymes began to normalize with only mild elevations in the alkaline phosphatase and aspartate aminotransferase (AST) (see Table 1). The patient was started on six cycles of adjuvant gemcitabine.

After initiating gemcitabine, she developed abnormal liver enzymes characterized by mild elevations in her bilirubin and transaminases (see Table 1). Due to a concern for a biliary anastomotic stricture, she was evaluated by balloon-assisted endoscopic retrograde cholangiopancreatography (ERCP) and was found to have a normal biliary tree with no evidence of stricture. Chronic hepatitis workup (A, B and C serologies, antinuclear antibodies and smooth muscle antibody) was negative. Computed tomography (CT) scan of the abdomen and pelvis with contrast revealed no signs of disease or liver lesions. She denied alcohol, illicit drug use, and herbal medication exposure. She was continued on her chemotherapy.

After she completed her sixth cycle of gemcitabine, the patient developed worsening cholestasis (see Table 1) and was subsequently hospitalized with liver failure as evidenced by a coagulopathy and development of encephalopathy. Transjugular liver biopsy revealed a hepatic injury in a cholestatic pattern with moderate pericellular fibrosis and steatosis (see Figure 1). This was consistent with drug-induced liver injury felt to be secondary to gemcitabine. The patient’s hepatic encephalopathy was medically managed with lactulose and rifaximin. After a long discussion with the patient and her family regarding her poor overall prognosis, she was discharged with home hospice.

DISCUSSION

Hepatotoxicity due to drugs is a common cause of acute liver failure. It has been estimated that more than half of the cases of acute liver failure referred to tertiary care centers in the United States are due to drug toxicity.¹ In most cases of drug toxicity, the drug is suspected to be the sole cause of liver injury. Drugs are typically divided into two categories: (1) dose-dependent hepatotoxins (predictable toxicity related to the blood level of the medication, the dosage and the duration of intake), an example of which is acetaminophen, and (2) idiosyncratic hepatotoxins whose toxicity is unpredictable and thought to be based upon genetic predisposition, age or environmental factors.²

Idiosyncratic reactions often occur without warning, unrelated to the dosage of the medication. Such hepatotoxins can cause a wide range of histologic changes and can be highly variable in the latent period before the onset of an injury. There are a multitude of different proposed mechanisms of idiosyncratic reactions including disruption of the cell membrane, activation of immune-mediated responses, disruption of bile excretion, activation of apoptosis pathways and inhibition of mitochondrial function.^3,4,5 While drug induced hepatotoxicity is common, the effects are often reversible with withdrawal of the offending agent and supportive care. Only rarely do patients develop acute liver failure.

Gemcitabine is a fluorine-substituted nucleoside analog, which has been widely utilized as a chemotherapeutic agent. This antimetabolite inhibits DNA synthesis during replication resulting in apoptosis of tumor cells. It is typically well-tolerated with only mild side effects including mild gastrointestinal disturbance, flu-like symptoms and rashes. Myelosuppression is one of the more commonly noted side effects of the medication. Elevations in liver enzymes are also commonly reported but are typically mild, transient and reversible. Rarely, gemcitabine can elicit severe idiosyncratic reactions resulting in acute liver failure that may be irreversible.

There have been only four previous reports of acute cholestatic liver failure due to gemcitabine, three of which were fatal.^6,7,8 Of note, one of the fatal cases was suspected to be due to a combination of both gemcitabine and vinorelbine.⁸ A fourth case has been more recently reported by Stellman et al. in which a patient with acute liver failure attributed to gemcitabine had a complete recovery after the medication was withdrawn.⁹ Similar to our patient, the clinical presentation for each of these cases was notable for marked cholestasis. Two of these cases, however, had a mixed hepatocellular and cholestatic pattern of injury. The exact mechanism of injury caused by gemcitabine has not been clearly identified. Another report suggested an alternate mechanism of acute liver failure due to gemcitabine through veno-occlusive disease.¹⁰ (See Table 2)

The time course of the acute liver failure and the histologic findings seen on liver biopsy clearly implicate gemcitabine as the cause of liver failure in our patient. Our patient did not develop acute liver failure until completion of her sixth cycle of gemcitabine, which emphasizes the documented latent period prior to development of acute liver failure. This shines a light on the importance of closely monitoring liver enzymes in patients during and after treatment with gemcitabine.

SUMMARY

Our patient is one of only a few cases of gemcitabine- induced cholestatic liver failure reported in the literature. Drug-induced hepatotoxicity is one of the most common causes of acute liver failure, and must be considered in all treated patients with liver enzyme abnormalities. Although gemcitabine hepatotoxicity usually causes mild elevations in liver enzymes that normalize when the medication is discontinued, cases of fatal cholestatic liver failure have been reported. Careful monitoring of liver enzymes should be performed on all patients treated with gemcitabine given the risk of acute liver failure. Additionally, patients with underlying liver disease may be at greater risk of the potential hepatotoxic effect of gemcitabine and should be monitored more closely.

A CASE REPORT

Colonic Polypoid Ganglioneuroma

Saire Mendoza M.,Parajuli D.

Saire Mendoza M.¹ Parajuli D.² ¹Section of Gastroenterology and Hepatology, Ochsner Medical Center ²Division of Gastroenterology, Hepatology and Nutrition, University of Louisville

INTRODUCTION

Intestinal ganglioneuromas can be found throughout the gastrointestinal tract and have three different presentations. They have been reported to be associated with multiple endocrine neoplasia type IIB (MEN-IIB), even before the GI endocrine neoplasms are diagnosed,¹ and occasionally with Von Recklinghausen’s disease (VRD).^2,3 These benign tumors are the result of submucosal (nerve) plexus hyperplasia in the intestinal wall,⁴ and consist of ganglion cells, nerve fibers and supporting cells. We report a case of a colonic ganglioneuroma found incidentally on screening colonoscopy. Management of these lesions, the risk of gastrointestinal malignancy and further assessment for occult non-gastrointestinal malignancies will be reviewed.

CASE

An otherwise healthy 69 year-old African American male underwent screening colonoscopy. A 1 cm subepithelial lesion, which was not characteristically adenomatous under both white light and narrow band imaging, was seen at the recto sigmoid junction (Figure 1). This lesion was biopsied and tattooed for future location. The initial sample did not yield a pathologic diagnosis thus rectal endoscopic ultrasound was performed. The lesion was noted to be hypoechoic and arising from the submucosa layer (Figure 2). Bite-on-bite cold forceps biopsies were taken and sent for histologic examination. The final pathologic findings were reported as a polypoid ganglioneuroma (Figure 3).

DISCUSSION

Intestinal ganglioneuromatous disease (GN) is divided into three groups.

1. Polypoid ganglioneuromas are small (<2cm), sessile or pedunculated polyps that are endoscopically difficult to distinguish from juvenile, hyperplastic or adenomatous polyps. Microscopically, they are divided into three patterns with disarranged crypt architecture with spindle and nerve ganglion cells within the mucosa being the most common.
2. Ganglioneuromatous polyposis (GP) consists of numerous, often 20 to 40, sessile or pedunculated lesions. There is less demarcation among the contents that make up the ganglioneuromas in this subgroup.
3. Diffuse ganglioneuromatosis (DG) presents as disseminated, nodular, intramural or transmural lesions ranging from 1 to 17cm in size and can lead to segmental bowel strictures; histologically these lesions involve the myenteric plexus. Most of DG cases are associated with MEN-IIB or VRD^3,5,6; and may radiographically resemble Crohn’s disease on barium studies and computed tomography scans ⁷

The largest case series of ganglioneuromatosis was reported by Shekitka et al.³ in 1994 where 43 patients with intestinal ganglioneuromatous disease were followed. While some patients presented with abdominal pain, rectal bleeding or megacolon, many were asymptomatic. However, cases with watery diarrhea secondary to vasoactive intestinal polypeptide secretion and diffuse ganglioneuromatosis involving the colon and rectum have been reported,^8,9 as well as upper GI bleeding from a ganglioneuroma of the duodenum¹⁰ and polypoid GN causing colonic intussusception.¹¹

The prevalence of diffuse ganglioneuromatosis in patients with MEN-IIB syndrome is nearly 100%, and often gastrointestinal involvement predates the development of medullary thyroid cancer.¹ Diarrhea or constipation from diffuse alimentary tract involvement, was reported in a case series of 16 patients with MEN-IIB.¹ DG was found before medullary thyroid carcinoma in 12 of these16 cases. Therefore, the authors concluded that diffuse ganglioneuromatosis heralds the development of endocrine neoplasms, especially medullary thyroid carcinoma, which is the most common and most feared component of the MEN-IIB syndrome.

Around 25% of patients with Von Recklinghausen’s disease have shown some gastrointestinal involvement, mainly neurofibromas and occasionally ganglioneuromatosis.^2,5 In one series, the mean interval between diagnosis of Von Recklinghausen’s disease and the onset of gastrointestinal symptoms related to gastrointestinal neurofibromatosis was approximately 37 years.² Also, the simultaneous finding of diffuse ganglioneuromatosis and associated adenocarcinoma of the colon in patients with VRD⁵ seems to be more related to the high incidence of adenocarcinoma in patients with Von Recklinghausen’s disease¹² rather than secondary to ganglioneuromatosis itself.

Polypoid ganglioneuromatosis and ganglioneuromatosis polyposis do not appear to be associated with systemic disorders such as MEN-IIB or VRD.^3,6,13,14 Shekita et al.³ followed 16 patients with Polypoid GN and, after eight years, 13 were still alive and did not have evidence of MEN-IIB or VRD. Similar findings were obtained for patients with ganglioneuromatosis polyposis, however, two of the three patients had multiple skin tags and cutaneous tumors which were thought to be lipomas. Chan and colleagues⁶ also reported these cutaneous findings in patients with ganglioneuromatosis polyposis.

Intestinal ganglioneuromatosis has also been found in association with adenomatous, hyperplastic and juvenile type polyps.^13,15 However, adenomas do not appear to be the prevailing feature in the setting of ganglioneuromatosis.

CONCLUSION

Three patterns of intestinal glanglioneuromatous disease have been described: polypoid ganglioneuromas, ganglioneuromatous polyposis (GP) and diffuse ganglioneuromatosis. Most cases of diffuse ganglioneuromatosis have been associated with MEN- IIB. The clinical presentation depends on size and location of ganglioneuromas. In diffuse gangliomatosis, gastrointestinal symptoms usually precede the development of thyroid medullary neoplasms and should prompt further diagnostic work up. There is not enough literature to suggest that intestinal ganglioneuromatosis has malignant potential.¹⁷ As such, an aggressive surgical approach in an asymptomatic patient without increased risk of carcinoma may be unnecessary.⁷ In the cases where GN is found along with hyperplastic or adenomatous polyps, current standard colonoscopy surveillance recommendations after polypectomy¹⁶ should be followed.

Abbreviation

VRD: Von Recklinghausen’s disease

Acknowledgement

We thank Dr. Silvia D. Potenziani Pradella from the Department of Pathology and Laboratory Medicine at University of Louisville for contributing images of polypoid ganglioneuromatosis of the colon.

UNUSUAL CAUSES OF ABDOMINAL PAIN, #8

Unusual Causes of Abdominal Pain

Omair Atiq,Lan Peng,Christian Mayorga

Omair Atiq M.D.,¹ Lan Peng M.D.,² Christian Mayorga M.D.,¹ 1Division of Digestive and Liver Diseases, ²Department of Pathology, University of Texas, Southwestern Medical Center, Dallas, TX

CASE

A 48 year old woman presented to the emergency department with chief complaints of abdominal pain, nausea and vomiting for five days. Her abdominal pain was located in the mid epigastrium and right upper quadrant (RUQ). It was constant and burning in nature. The vomiting was approximately three times per day and consisted of watery material and food particles, and she occasionally noticed bright red blood. She also noted having one loose bowel movement which was dark in color. She otherwise denied any fevers, chills, dysphagia, odynophagia or hematochezia.

Her medical history was significant for HIV/AIDS on highly active anti-retroviral therapy (last CD4 count: 51 cells/mcL, viral load: 47377 copies/ml), oral candidiasis, disseminated varicella zoster and anemia of chronic disease. On presentation, her vitals were stable. Her abdomen was soft, non-distended, with moderate tenderness on palpation over her RUQ.

Initial laboratory values revealed WBCs: 3.92×10(9)/L, Hb: 9.0 gm/dL, platelets: 225×10(9)/L. Computed tomography (CT) scan of the abdomen showed concentric wall thickening involving the third part of duodenum and proximal jejunum. Her symptoms did not respond to intravenous antiemetics and analgesics. An esophagogastroduodenoscopy (EGD) showed a normal esophagus and stomach. There was patchy, mildly erythematous mucosa in the duodenal bulb. Upon entering the second portion of duodenum, the mucosa appeared markedly abnormal. There was severe villous blunting, mucosal atrophy, diffuse erythema and scattered erosions (Figures 1 and 2). These changes persisted into the third position of duodenum and beyond. Biopsies were taken for histology and microbiology (Figures 3 and 4).

ANSWER AND DISCUSSION
Enterocolitis Caused by Strongyloides Stercoralis (Threadworm)

The pathology report showed duodenal mucosa with inflammation and significant architecture distortion including villous loss (Figure. 3). Both strongyloides adult (black arrow) and larvae (yellow arrow) are seen within crypts. Mucosal ulcer and eosinophilic and neutrophilic infiltration are present (Figure 4).

The patient was diagnosed with enterocolitis caused by Strongyloides stercoralis. She received a course of ivermectin. Her abdominal pain completely resolved within a week, and she remained asymptomatic at two months after hospital discharge.

Strongyloides stercoralis (threadworm), an unsegmented helminth of class Nematoda, was first described in 1876 when it was identified in the feces
of French colonial troops suffering from diarrhea in Cochin-China.¹ Strongyloides is prevalent in the tropical and sub-tropical regions of the world and is endemic in the southeastern United States. Immunosuppressed patients are at greatest risk because impaired cellular
and humoral immunity alter parasite proliferation, resulting in increased parasitic burden and possible dissemination to other organs. Studies have reported the incidence of strongyloidiasis in patients with HIV to be near 2.5 %.² The most common gastrointestinal (GI) symptoms result from involvement of the upper GI tract and include nausea, vomiting, diarrhea, weight loss, and abdominal pain. Additional manifestations of GI strongyloidiasis include malabsorption, GI hemorrhage, colonic pseudopolyposis, granulomatous hepatitis, biliary obstruction, and eosinophilic ascites. Endoscopic findings include ulcerations, gastritis or duodenitis as evident in our patient. Diagnosis can be made with stool
examination, enzyme linked immunosorbent assay or a duodenal or jejunal biopsy (~90% diagnostic yield). Ivermectin (200 mcg/kg/day) is the first-line agent for treatment, and a course of five to seven days has been
suggested in immunosuppressed patients with systemic disease. Lifelong suppressive therapy may be indicated for both gastrointestinal and pulmonary infections in patients with relapses.³ Intestinal strongyloidiasis is not an uncommon infection in immunosuppressed patients and should be in the differential diagnosis for patients presenting with nausea, vomiting and abdominal pain.

Endoscopy should be considered if symptoms are not resolving with conservative management.

GASTROINTESTINAL MOTILITY AND FUNCTIONAL BOWEL DISORDERS, SERIES #7

Pediatric Biliary Dyskinesia and Sphincter of Oddi Dysfunction

Aldo Maspons,

Richard W. McCallum

Adult gastroenterologists consider biliary dyskinesia a functional disorder and new research has opened the door for Type 3 sphincter of Oddi dysfunction to be considered as a functional disorder. Pediatric gastroenterologists should follow the lead of our adult colleagues and consider the reclassification of biliary dyskinesia and sphincter of Oddi dysfunction type 3 as functional disorders.

Biliary dyskinesia is a gall bladder disorder in the setting of a normal biliary tree. Adult gastroenterologists consider it a functional disorder, but this is not the case in pediatrics, as it is amongst the most common reasons for laparoscopic cholecystectomies. Sphincter of Oddi dysfunction is a motility disorder in the contraction of the sphincter leading to obstruction of the bile or pancreatic juice.1 New research has opened the door for Type 3 sphincter of Oddi dysfunction to be considered as a functional disorder in adult gastroenterology. Pediatric gastroenterologists should follow the lead of our adult colleagues and consider the reclassification of biliary dyskinesia and sphincter of Oddi dysfunction type 3 as functional disorders.

Aldo Maspons, MD¹ Richard W. McCallum, MD² ¹Division of Pediatrics, El Paso Children’s Hospital/Texas Tech ²Department of Medicine, Division of Gastroenterology, Texas Tech University Health Sciences Center, Paul L. Foster School of Medicine, El Paso, TX

INTRODUCTION

Pediatric Biliary Dyskinesia

Biliary Dyskinesia (BD) is a gall bladder disorder in the setting of an anatomically normal biliary tree. The clinical manifestations include episodic colicky right upper quadrant pain or epigastric pain triggered by food, associated with nausea and vomiting in the absence of cholilithiasis. The pain is thought to be caused by contraction of the sphincter of Oddi at the time of gall bladder relaxation. The most commonly affected tend to be female patients and/or overweight patients.^2,3

Pediatrics often takes its cues from the adult world in terms of diagnosis and treatments so as to mitigate any potential risks for our society’s most vulnerable citizens. Biliary dyskensia (BD) is an example of a disorder first described in adults and treated operatively with subsequent practice adoption in pediatrics.⁴ BD is considered a functional disorder according to Rome III,⁵ but it is not recognized as a childhood functional disorder.⁶ Despite BD being classified as an adult functional disorder, laparoscopic cholecystectomy has been relied on as the final therapeutic solution.⁷

Diagnostically, a Hepatobiliary Iminodiacetic Acid (HIDA) scan along with an injection of cholecystokinin (CCK) is used to assess gall bladder function by evaluating bile excretion, accumulation, and ejection from the gall bladder. Gallbladder emptying less than 35% along with typical symptoms established by the Rome III Criteria establish the diagnosis of BD.⁵ One study found problems with solely using a HIDA scan to make a diagnosis of BD. Some children who had abnormal ejection fractions on their first scans, later had normal ejection fractions.⁸ Caution must be taken regarding opiate pain medication in the background at the time of the abnormal HIDA scan. This is one example of the problems encountered with extrapolating adult data to fit the needs of pediatric patients. The clinical response to cholecystectomy is subsequently used to confirm the diagnosis of BD if there is no recurrence of pain for 12 months.⁵ HIDA scans and Ejection Fractions should be considered as only one aspect in the decision making process.

The diagnosis of BD has been on the rise and this is seen in the steady rise of laporoscopic cholecystecomy. Mehta et al. describe the trend of cholecystectomies at one of the countries largest children’s hospitals in which from the years 1980-1996, there were no documented cases of BD, but from the years 2005-2008, there were 64 cases, making it the third most common indication for a cholecystectomy, behind complicated obstructive disease, and symptomatic cholelithiasis.²

Their study, much like those of others, demonstrated that histologically, the gall bladder had features of chronic cholecystitis.^8-10 Friesen et al. demonstrated that patients with BD had an increase in mast cell density, which parallels other functional disorders.⁸ When comparing the differences between patients who underwent a cholecystectomy secondary to stones vs. BD, patients with BD had more mast cells in the lamina propria. Mast cell degranulation and the high levels of mast cells within the lamina propria, however, did not correlate with the ejection fraction.

What has been lacking in BD research is randomized control trials with larger groups for analysis. The research data for the diagnosis and treatment is based on retrospective studies with small number sizes. Time of response to surgery has also been limited and mostly to only a few months after surgery. Few studies have looked into long-term follow-up. One such study looked at 2.8 years after surgery³ and demonstrated that 44% of their cohort was symptom free after laporscopic cholecstectomy. Nelson, et al., evaluated 55 patients and compared cholecystectomy to children who did not undergo surgery, but instead were observed in the setting of BD.⁸ At the two-year follow up, 55% of the cholecystecomy group had complete resolution of pain and 55% of the observation group also had complete resolution of pain. Twenty-six percent of the cholecystectomy group and 20% of the observation group had partial resolution of pain. Two years after the original two-year follow-up, there were no differences in the rate of response to pain. This study used long term data to suggest that a non-operative approach may be just as effective as an operative one, which is consistent with the long -term outcome of functional disorders.

The relationship between pediatric functional disorders and gall bladder disease was explored by Chumpitazi et al.¹¹ They found that children with BD may also have concomitant gastroparesis, suggesting a neuroenteric abnormality affecting both biliary and gastric motility. Srinath et al. argue that BD is a safe diagnosis in children in that there are minimal associated complications. This should emphasize, according to them, the need to explore therapies used in treating BD as a functional disorder and there should be less reliance on the surgical routes of care for this disorder where there are complications from such interventions.¹²

Currently, there are no established biliary functional disorders in pediatrics (Table 1).¹³ More research for conservative management has begun to set the stage for Biliary Dyskinesia in Pediatrics to be managed like other functional disorders in Pediatrics-conservatively, medically, perhaps with combination therapy involving psychological aspects to treatment.¹²

Pediatric Sphincter of Oddi Dysfunction

The sphincter of oddi (SO) is located at the duodenal junction of the biliary and pancreatic ducts.⁵ Sphincter of oddi dysfunction (SOD) is a motility disorder manifested as an abnormality in the contraction of the sphincter of oddi (SO)¹⁴ that leads to an obstruction of bile or pancreatic juice.¹ Patients typically have biliary obstructive pain not relieved with cholecystectomy. There are three distinct types of SOD. Type 1 consists of patients with a dilated bile duct and abnormal liver labs. Type 2 consists of patients who have either a dilated bile duct or abnormal liver labs, but not both. Type 3 have neither a dilated duct nor abnormal labs.¹⁵

SOD can present with symptoms that include epigastric pain or right upper quadrant pain that radiates to the back or right shoulder in episodic fashion. It can present as idiopathic pancreatitis, biliary type pain with an intact gallbladder and no gallstone, and present after a cholecystectomy. In addition, SOD has been studied as a possible cause of recurrent abdominal pain in children.¹⁶ Currently in the adult literature, SOD is regarded as a functional disorder,⁵ but not so in the pediatric literature.⁶ There is limited data in pediatrics in regards to the number of studies performed. All studies are retrospective chart reviews and the number of patients is small. Brown et al. and Lemmel et al. found 3 of 38 children and 7 of 29 children with SOD as the cause of recurrent pancreatitis, respectively.^17,18 Varadarajulu et al. evaluated 6 patients, three type II, and three type III over a three year time period.¹ Cheng et al. as a part of a large ERCP series evaluated, six patients with SOD type II, 35 patients with SOD type III over a ten-year period. Chronic abdominal pain and recurrent pancreatitis were the two most common presentations. Post ERCP rates of pancreatitis in these patients ranged from 21-30%.¹⁹ Misra et al. performed a retrospective chart review with telephone follow up.¹⁶ Twelve patients, 9 of which had SOD Type III, underwent SO manomentry and sphinctertomy for treatment. Of the twelve patients, three did not respond. Seven patients at the five-year follow up continued to be asymptomatic. Phone follow up, however, did not include a standardized pain scale assessment.

Sphincter of Oddi Dysfunction is first suspected with the above mentioned clinical presentation. To confirm the diagnosis, the ampulla of Vater is cannulated with an endoscopic rertrograde cholangiopancreatography (ERCP). Sphincter of Oddi manometry is then performed with a low compliance infusion pump system (Figure 1). Basal sphincter of Oddi manometry pressure greater than 40 mm Hg or greater are considered abnormal and based on adult data.¹⁹ Treatment of the SOD consists of a sphincterotomy that is preformed on the biliary and/or pancreatic sphincter, depending on which one is above the adult based normal value. There are no pediatric manometrical based pressure values and therefore, adult values are used. Adult data demonstrates that SOD II is more likely to have manometric demonstrable SOD 55% of the time vs. 28 % of the time in SOD III.²⁰ These procedures are not without risk. Post ERCP pancreatitis rates in pediatrics has occurred in up to 13% of cases.²¹

Recently, Cotton et al. performed the first multicenter, sham-controlled, randomized control trial with a large patient volume involving 214 patients.¹⁵ The results of this study have shifted the dynamic of SOD III. Cotton et al. demonstrated that sphincterotomy was not more effective than a sham ERCP in post cholecystecomy patients with SOD type III.¹⁵ Futhermore, manometrical findings, biliary or pancreatic pressures, were not associated with sphincterotomy outcomes. Given these results, treatment with sphincterotomy for patients with SOD III should be called into question because it was demonstrated that sphincterotomy, which is not without risk, is a procedure no better than placebo.

This recently published trial is being interpreted by adult gastroenterologists as moving SOD into a functional bowel disorder category such as irritable bowel syndrome. Hence, the stage invites SOD III in children to be considered a functional disease with therapy that at the very least should no longer involve an invasive approach. Further studies are needed to determine the pathophysiology of pain and a medical route of therapy. Like other pediatric functional disorders, and much like what is suggested for the treatment of biliary dyskinesia, a therapeutic combination that includes a non-invasive approach should be considered in the management of this disease. This combination of treatment should include psychological techniques to deal with anxiety and visceral sensitivity that are so often a part of pediatric functional abdominal pain.²² In the future we speculate that brain-gut modifying agents may be worthy of treatment trials, e.g. Tricyclics. In addition, treatment with cholinergic therapy addressing colon motility will have a role.

Diet could also be considered another form of
therapy. There already is precedence for this kind of
therapy in irritable bowel syndrome (IBS) with the
FODMAPS diet. Much like the role of the intestinal
microflora in IBS, perhaps it is also playing a role
in SOD. Changing the intestinal microflora is one of
many treatment possibilities for a disease whose current
treatment should be further explored.

INFLAMMATORY BOWEL DISEASE: A PRACTICAL APPROACH, SERIES #93

Adapting Lugasi’s Core Conditions for Optimizing Transition of Pediatric to Adult Care in IBD

Lisa B. Malter

The incidence of pediatric-onset inflammatory bowel disease is rising. As this growing population nears adulthood, they are faced with transition of care (TOC) to adult gastroenterologists. This critical transition period has been studied in pediatric patients with chronic medical conditions and several core conditions for successful transition have been identified. In this paper we apply Lugasi’s conditions for successful transition of care specifically to patients with inflammatory bowel disease and suggest a blueprint for a successful TOC process.

Joseph Kingsbery1 Martin J. Wolff¹,² Nancy McGreal³ Lisa B. Malter^1,2 ¹New York University School of Medicine, Department of Internal Medicine, ²New York University School of Medicine, Division of Gastroenterology, New York, NY; ³Duke University Medical Center, Divisions of Adult and Pediatric Gastroenterology, Durham, NC

INTRODUCTION

Transition of care, recently defined as the lengthy process of preparing the adolescent for a life as an adult and receiving care from adult health care providers, is a pivotal point in an adolescent’s life.¹ It is essential that this process result in uninterrupted care, but frequently there are barriers to a successful transition. Patients and their families are often hesitant to leave their pediatric caregivers, and do not fully understand the meaning of – and reason for – transitioning to adult-centered care (ACC). Despite the increasing emphasis placed on this process, transitioning care often remains a stressful time for pediatric patients and their families.^2,3,4 To address these concerns, in 2010 Lugasi et al. performed a metasummary of 46 qualitative and descriptive quantitative studies involving adolescents with chronic medical illness before and/or after their transfer to ACC. This comprehensive metasummary which included adolescents with cystic fibrosis, congenital heart disease, sickle cell disease, solid organ transplant and diabetes identified 5 core conditions for a successful transition process. In this article we have adapted these “core” conditions to the adolescent population with inflammatory bowel disease (IBD).

Core Conditions

I. Meaning of transition
II. Expectations
III. Level of patient’s knowledge/skills
IV. Planning/Timing
V. Environment

I. Meaning of Transition

Transitioning from child-centered care (CCC) to ACC is an understandably stressful time for pediatric patients and their families. Often when the subject of transitioning care is broached, the initial fear of patients and parents is that of abandonment.^2,3,4,5,6,7 It is important to elicit the main concerns that patients and families have about transitioning to ACC, gauge their understanding of the reason for transitioning care, and to highlight the benefits of transitioning care (Table I).

For parents transition from CCC to ACC reduces their authority in their young adult’s health care. Parents who have been involved in all treatment decisions must relinquish these responsibilities to their children and trust that they make well-informed long term decisions.³ This new onus is daunting to the patient, who must now assume responsibility for all aspects of their own healthcare, inform themselves of a new healthcare system and redefine their role in the self-management of their condition.⁸

This shift in responsibility may be anxiety- provoking for the parent and patient. The literature has identified certain aspects of transitioning care that pediatric patients are particularly interested in, which affords an opportunity to advocate for ACC.

Perhaps the most consistent theme among transition- aged patients is their preference for being treated like adults.^4,5,9 Transition aged patients described CCC as patronizing.⁴ They prefer the medical staff to address them as opposed to their parents. Despite their initial anxiety, adolescents appreciate taking responsibility for their own care and gain self-esteem and confidence from the transition.⁴ In the “patronizing” setting of CCC it is easy to forget the patient’s social environment and the decisions they are making without the assistance of their parents on a daily basis. In CCC, pediatric gastroenterologists do address the impact of tobacco use, drug use, and sex – as well as less hot-button topics such as sun exposure and stress – on their disease. Adult gastroenterologists reported that their newly transitioned patients were not sufficiently privy to these topics and transition-aged patients reported that they would like more information on these topics.^6,9

In counseling patients and families about transitioning to ACC, it is important to acknowledge and validate their concerns to ensure that transitioning care is not construed as abandonment by their pediatric care team. Instead, the pediatric provider should convey that transitioning care is an involved and lengthy process in which they will be educated and ushered toward ACC, as well as highlight the benefits of transitioning: engendering autonomy, responsibility and self-management skills.

II. Expectations

Pediatric patients and their families frequently have preconceptions about ACC (table II). This section describes the preconceptions that pediatric patients and their families have toward ACC and provides a general outline of the important differences between CCC and ACC that patients should be informed of in advance of initiating the transition process.

Lugasi et al. describe that pediatric patients and family members generally think of ACC as cold, impersonal, and business-like, whereas CCC is familiar and friendly. Escher et al. describes the perceived differences between pediatric and adult gastroenterologists similarly.^2,3 Though these differences are subjective and difficult to investigate, pediatric care for IBD tends to be provided at tertiary care centers with a multidisciplinary team in a family- focused arena, whereas adult care is often provided by a single doctor at a local hospital and is patient focused.¹¹ Newly transitioned patients in a flare of their disease usually will be admitted to a general medicine floor at a local hospital, as opposed to the familiar setting of a pediatric unit.

Pediatric practitioners focus on physical, developmental, and psychosocial milestones of their patients with IBD. The dynamic nature of IBD as well as medication side effects during critical stages of growth and development often result in failure to meet these milestones, and pediatric practitioners are keenly aware of this.¹³ Puberty is sometimes delayed due to malnutrition, and psychosocially patients suffer as they miss school during flares and commonly struggle with self-esteem and identity issues.¹³ Pediatric practitioners aim to alter the early disease course in their patients to minimize long-term sequelae. Adult practitioners also aim to minimize disease flares, but the focus of care begins to shift toward malignancy surveillance, new drugs/treatments, and long-term effects of IBD and IBD treatment, as well as family planning.³

Patients and families develop expectations of endoscopy, radiation exposure and treatment of IBD based on the care they have received in CCC, though they should be made aware of potential changes to expect. Frequent endoscopic surveillance is an important aspect of ACC. Endoscopy is a means of monitoring for malignancy, evaluate response to therapy and predict future risk of recurrence. Endoscopies are more frequent in ACC and performed under moderate sedation, whereas in CCC endoscopy is usually performed under deep sedation or monitored anesthesia care. This change alone may cause anxiety in a transition-aged patient, and therefore it is important for patients to be aware of this difference.

Children are considerably more sensitive to radiation than adults. A study from Columbia University in 2008 showed that earlier exposure to radiation was associated with a higher risk of developing cancer later in life. A study from Great Britain in 2012 showed a positive association between radiation dose from childhood CT scans and the subsequent development of leukemia.^16,17 At the University of Michigan in 2013, pediatric patients with IBD had undergone 1.08 CT scans in their lifetime.¹⁸ In contrast, radiation exposure amongst all-comers with IBD in Ireland between 1999 and 2009 showed a trend toward more radiation exposure on over a ten-year period in adult compared with pediatric populations.¹⁹ These trends have been reproduced in studies of adult patients with IBD in Alberta, Canada,²⁰ as well as a study of lifetime radiation exposure in adult IBD patients in the UK.²¹ There is significant variation in resource utilization depending on the local patterns of practice, but there is an overall trend toward significantly more diagnostic radiation in adult patients with IBD. Transition-aged patients must be made aware of the long-term risks associated with exposure to radiation during CCC and also of the trend toward more radiation in ACC.

Patients who are stable on certain drug regimens in CCC may be asked to change medications by their adult gastroenterologist, potentially adding uncertainty while transitioning care. Newly transitioned patients, when compared with their pre-transition counterparts, have poorer rates of medicine compliance, which may be an indication of discomfort and unfamiliarity with new treatment regimens.12 Adult gastroenterologists treating IBD typically present several therapeutic options to the patient in the hopes at arriving at a shared decision. In CCC the patient’s parent or guardian is much more likely to be the decision-making party regarding a particular therapeutic venture. Additionally, pre-conceptive counseling and more aggressive medical or surgical management of longstanding disease are more common in ACC. In ACC patients are expected to play a more active role in their management and often require re-education on the nature of their disease, the risks and benefits of their therapy, and counseling as to the possible consequences of medical non-compliance.

Pediatric patients and family members often have preconceived notions about ACC and the changes to expect. It is important to address these notions about the general environment of ACC, and to acknowledge that there are significant changes to be expected simply by transitioning from CCC to ACC. Speaking openly with patients and family members about these changes will help to reduce uncertainty during and after transition as well as to ensure that their expectations of transitioning care are realistic.

III. Level of Patient’s Knowledge/Skill

In a joint study published by the Journal of Pediatric Gastroenterology and Nutrition (JPGN) in 2008, adult gastroenterologists affiliated with the Crohn’s and Colitis Foundation of America (CCFA) were asked what they expected from young adult patients with IBD entering their care in terms of patient knowledge about their disease.⁶ Among the most important expectations were knowledge of the name, dose and major side effects of the medications at the time of transition, their medical condition, and an understanding of their disease course and prognosis. Adult gastroenterologists routinely reported that these expectations were often unmet upon patient transition into their care.

The JPGN and adult gastroenterologists affiliated with the CCFA recommend early and repetitive teaching of these basic concepts. There are checklists based on chronological age of their patients for age-appropriate knowledge of their disease.²³ Readiness assessments based on disease knowledge such as “IBD yourself” ,¹⁴ and “MyHealth Passport for IBD”¹⁸ have been devised but have not yet been validated for assessing readiness for transition. The Social-Ecological Model of Adolescent and Young Adult Readiness to Transition (SMART) is a more comprehensive readiness assessment tool that highlights the importance of patient, parent and provider components to a successful transition.^24,25

These assessment tools are limited and knowledge alone is not sufficient. Fishman et al. suggest that self- efficacy – an individual’s mental or cognitive state based on past or present experience determines his/her ability to organize and implement a pattern of behavior necessary for health promotion – which correlates better than knowledge alone.^14,15 Similarly, in a recent study, healthcare providers caring for adolescents with IBD identified the patient’s developmental maturity as one of the greatest predictors of transition success, and identified difficulties with abstract reasoning as predictive of a failed transition. Patients have difficulty understanding that absence of symptoms may not correlate to disease activity emphasizing the need to take medications despite feeling “well”. ²²

A patient’s knowledge of their disease and medications, as well as evaluation of a patient’s social understanding and self-advocacy would best predict a patient’s readiness for transition and determining the proper timing for transition.

IV. Planning/Timing

Transitioning to ACC must be broached early in the course of a young patient’s chronic illness. However, there is no consensus on the exact timing of transfer of care.⁹

The age at which transfer must occur has varied from country to country based upon age. In the UK and France, the legal age limit for pediatric medical care is 18 years of age, whereas in the USA it is 21 years. In the UK, patients with cystic fibrosis and their parents reported frustration at the inflexibility of the legislatively determined age of transition.¹ Since the legal age of transfer in the USA is 21, the transition timeline must not follow as “rigid” of a protocol as elsewhere.

The exact timing of transfer must be tailored to the emotional and cognitive competency level of the patient during a time of medical stability.⁶ Transfer should occur at a time of relative disease quiescence; active flares are contraindications to transitioning care. However, there is debate about the timing of transfer with regards to social stability.

Many patients that are “transition-aged” or otherwise deemed “ready” for transition are commonly 16 to 19 years old. Transferring medical care in the setting of other major life changes creates undue stress and be deleterious to the patient’s health.¹¹ On the other hand, pediatric and adult gastroenterologists in the UK agreed that the age of 18 was an ideal time for transfer since 18 marks the end of secondary school and entry into university education or employment.²⁶ Transfer therefore should occur not reflexively at a specified chronological age, but rather be assessed on a case-by- case basis (Figure 2).

V. Environment

Appropriate resources are integral to a successful transition environment. Currently there is no standardized transition of care protocol in the US since transition programs are highly variable. Transition preparation ranges from brief educational sessions to more extensive visits with adult and pediatric gastroenterologists and a dedicated nurse or social worker to facilitate transition.⁸ Most important in transitioning care is the joint visit with the pediatric and adult gastroenterologist.⁷ These joint visits engender a sense of continuity and security with their new provider. Patients and family members surveyed after a joint meeting with their former and future provider felt that the visit was useful in terms of establishing confidence in the new physician.⁷

As awareness of this critical transition period becomes more widespread, programs have helped ease the psychosocial stress that accompanies growing up with a chronic medical illness. At Duke University, the Adolescents Transitioning to Leadership and Success program (ATLAS) pairs college students living with a chronic disease with children also coping with an illness. The children and their mentors discuss issues pertinent to living with a chronic disease, how it may influence relationships with their family, their peers, as well as illness-related experiences.²⁷ Children are paired with an empathetic friend and are equipped with a strong support system. Doctors, nurses and social workers help assist them in navigating a complex medical system.

In many pediatric practices it is not feasible to orchestrate a multidisciplinary transition process incorporating transition nurses, psychologists/ psychiatrists and social workers. However, providing patients and families with information about available support programs and arranging joint meetings between pediatric and adult gastroenterologists positively impact the transition experience.^15,21,28

CONCLUSION

Transitioning from CCC to ACC is a critical event in a young person’s life. Pediatric practitioners have recognized the importance of this process but this process is not as yet optimized. The discussion about transitioning care must begin early in the pre-adolescent course of disease. The healthcare provider should validate and address the patient’s and family’s concerns about transition while highlighting the benefits of this transition. The pediatric care provider must educate the patient about their disease, the natural history of their illness and what to expect during and after the
transition process. Joint meetings among pediatric and adult gastroenterologists, family and patient builds trust between patient and providers. By satisfying the 5 core conditions of TOC set forth by Lugasi, a
potentially tumultuous time in a young patient’s life can be translated into an empowering event guiding their future health care.

LIVER DISORDERS, SERIES #2

An Overview of Hepatic Fibrogenesis

George Tan,

Duminda Suraweera,

Gaurav Singhvi

Hepatic fibrosis is a dynamic process that results from chronic liver injury. While the common mechanism involves hepatocyte injury, inflammation and eventual distortion of hepatic architecture, each etiology has a unique pathway. In this review we discuss major causes such as alcoholic liver disease, non-alcoholic fatty liver disease, chronic viral hepatitis and cholestatic liver disease. In addition we review current diagnostic modalities and provide general management principles.

George Tan MD, MBA, Kaiser Permanente, Daly City, Duminda Suraweera, MD, Olive View-UCLA Medical Center, Gaurav Singhvi, MD, UCLA, David Geffen School of Medicine

Hepatic fibrosis is a dynamic process that results from chronic liver injury. It kills over one million people a year worldwide and has many etiologies. While the common mechanism involves hepatocyte injury, inflammation and eventual distortion of hepatic architecture, each etiology has a unique pathway. In this review we discuss major causes such as alcoholic liver disease, non-alcoholic fatty liver disease, chronic viral hepatitis and cholestatic liver disease. In addition we review current diagnostic modalities and provide general management principles.

Hepatic fibrosis is a major public health problem that carries with it a high morbidity and mortality. It results from the wound-healing response to chronic liver injury and can result in cirrhosis and hepatocellular carcinoma (HCC).^1,2 Common causes include alcoholic liver disease, non-alcoholic liver disease (NAFLD), chronic viral hepatitis, and cholestatic liver disease. Other less common etiologies include autoimmune hepatitis, Wilson’s disease, hemochromatosis, and schistosomiasis (See Table 1).^2-7 Regardless of etiology, the end result of chronic hepatic injury is a fibrotic liver with hepatic stellate cells playing a pivotal role in the formation of hepatic fibrosis.^8-10

Fibrogenesis begins with hepatocyte injury and inflammation that activates hepatic stellate cells (HSC). These activated HSCs then transdifferentiate into myofibroblasts, which results in an increased extracellular matrix (ECM) deposition in the liver leading to fibrosis.^2,11-13 This is shown pictorially in Figure 1. Bone marrow derived fibrocytes and epithelial- mesenchymal transition (EMT) from hepatocytes and cholangiocytes also contribute to fibrosis. Several fibrogenic mediators get recruited in the inflammatory cascade including transforming growth factor (TGF- beta), platelet derived growth factor (PDGF), insulin- like growth factor I (IGF-I), endothelin-I (ET-I), and reactive oxygen species (ROS).^13,14 Repeated hepatic injury results in this proinflammatory microenvironment and leads to liver fibrosis, cirrhosis, and the development of HCC. Early intervention can lead to the reversal of hepatic fibrogenesis.¹⁴

Alcoholic Liver Disease

Alcoholic liver disease (ALD) is due to chronic and excessive alcohol consumption and is a leading cause of liver disease worldwide.¹⁵ In fact, ALD is the third highest risk factor for disease and disability globally with nearly 4% of the world’s deaths attributed to alcohol consumption.^16,17 Per the National Institute on Alcohol Abuse and Alcoholism, the 12th leading cause of death in the United States is cirrhosis with 48% of those deaths due to alcohol.¹⁸ Not only is there a high mortality associated with alcohol abuse, but it leads to increased social problems including violence, child neglect and abuse, and absenteeism in the workplace.¹⁶

By definition, ALD can occur when daily alcohol ingestion exceeds 20g in women or 30g in men. This number should not be taken as an absolute threshold as patients vary based on differences in genetic susceptibility and other risk factors.^19,20 Indeed, the spectrum of alcoholic liver disease is vast and includes simple steatosis or fatty liver, alcoholic hepatitis, end- stage cirrhosis, and HCC.²¹ Of note, nearly 100% of heavy drinkers have fatty liver, but only 10-20% of them advance to alcoholic hepatitis or obtain the final pathologic changes of ALD associated fibrosis and cirrhosis.^15,22,23 In ALD-associated fibrosis, the major cell type that contributes to fibrogenesis is the activated hepatic stellate cell. While the underlying mechanism of fibrosis in ALD is very similar to the mechanisms seen in other chronic liver diseases, methionine metabolism abnormalities, hepatocyte apoptosis, oxidative stress, and endotoxin lipopolysaccharides which activate Kupffer cells may play special roles in ALD fibrosis.^16,22 In addition, emerging mechanisms underlying hepatic fibrogenesis in ALD include lipogenesis, cannabinoid receptor activation, and IL-1 signaling.^11,16

Non-alcoholic Fatty Liver Disease (NAFLD)

As opposed to ALD, non-alcoholic fatty liver disease (NAFLD) occurs in the absence of chronic alcohol consumption (less than 20g of pure alcohol/day for women and less than 30g of pure alcohol/day for men) or other liver diseases and has emerged as the most common chronic liver disease in Western countries.^24-26 Obesity, unhealthy diet, sedentary lifestyle and genetic predisposition are all risk factors associated with the development of NAFLD. Higher rates of insulin resistance, diabetes mellitus, hypertension, dyslipidemia and the metabolic syndrome are associated with this disorder.²⁷ In fact, excessive food intake, especially high fructose corn syrup and saturated fats have been shown in numerous studies to contribute to the development of NAFLD.²⁸ NAFLD is thought to affect 30% of the general adult population and 70-80% of patients that are diabetic or obese.²⁹ Furthermore, the impact of NAFLD on society is significant since it is estimated that NAFLD increases healthcare costs by 26% and will be the leading cause of liver transplantation by 2020.²⁵

NAFLD encompasses two clinicopathological entities that range from simple steatosis to non-alcoholic steatohepatitis (NASH). Simple steatosis accounts for 80-90% of NAFLD cases and is characterized by an excessive amount of fat in the liver, and is mostly benign and non-progressive. NASH constitutes the remaining 10-20% of NAFLD cases and is characterized by steatosis coupled with inflammation and fibrosis, and can progress to cirrhosis and HCC.¹⁹ The development of NASH is often described by the “two-hit” mechanism with the “first hit” being the development of steatosis and the “second-hit” involving environmental factors such as oxidative stress and proinflammatory cytokines coupled with genetic factors leading to hepatic injury.^29-31 Once patients develop NASH, approximately one-third go on to develop hepatic fibrosis.³² In addition to obesity and sedentary lifestyle, NAFLD increases with age, central obesity, and has a strong genetic predisposition with several affiliated gene polymorphisms.³³ The renin- angiotensin system (RAS) seems to play an important role in the development of NASH, as does the bacterial endotoxin within the gut-liver axis.^34-36

Chronic Viral Hepatitis

Hepatitis B (HBV) and Hepatitis C (HCV) viruses are leading causes of chronic liver disease. It is estimated that over two billion people have been infected with HBV, of which over 300 million are chronic carriers.³⁷ On average only about 10% of patients with HBV progress to chronic disease.³⁸ Of the chronic HBV patients, about 20% will develop liver cirrhosis.³⁹ The risk of HCC is about 100 times greater than the general population.⁴⁰ HBV promotes liver fibrosis via expression of the hepatitis B virus X (HBx) protein. This particular protein increases the expression of type 1 collagen, TGF-beta and increases the cell proliferation rate.⁴¹ Studies have also shown that the HBx protein accelerates proliferation of HSC cells thereby facilitating liver fibrosis.⁴²

It is estimated that over 185 million people worldwide are infected with HCV. Eighty percent of those infected progress to chronic infection.⁴⁴ Furthermore 20% of patients with chronic HCV will develop cirrhosis within 25 years and 25% of these patients develop HCC or decompensated liver disease.⁴⁵ HCV is the primary cause of liver transplantation in the United States.⁴⁶ There are a total of 6 identified genotypes of HCV. In the United States, 97% of all infections are from genotype 1, genotype 2 and genotype 3.⁴⁷ The inflammatory cascade that leads to cirrhosis is likely initiated by HCV core and NS3 proteins.⁴⁸ The subsequent cytokine and chemokines generated lead to increased recruitment of inflammatory cells such as macrophages, dendritic cells, natural killer cells and cytotoxic T cells. HCV activated Kupffer cells release ROS and other proinflammatory mediators thus leading to the common hepatic fibrosis pathway.

Cholestatic Liver Disease

Cholestatic liver disease primarily results from an impairment of hepatobiliary production and excretion of bile. Cholangiocytes and hepatocytes proliferate in response to injury leading to biliary damage, periductular fibrosis and cirrhosis.⁴⁹ The two most common causes of chronic cholestatic liver disease are primary biliary cirrhosis (PBC) and primary sclerosing cholangitis (PSC). PBC is a progressive autoimmune condition with an incidence of approximately 100 cases per million people.⁵⁰ It primarily affects women in the fifth decade of life and is associated with an increased incidence of HCC.^51,52 The pathogenesis of PBC is an autoimmune mediated process and can be divided into several stages. The initial stage involves infiltration of the portal triad by lymphocytes, plasma cells and eosinophil granulocytes.⁵³ As PBC progresses fibrotic septa extend from the portal tracts and link them together. This is called “bridging fibrosis” and is the characteristic finding in PBC.⁵³ Eventually the hepatic architecture becomes distorted leading to cirrhosis and the formation of regenerative nodules. Diagnosis of PBC is made when 2 of the following 3 criteria are met: presence of anti-mitochondrial antibodies, elevation serum alkaline phosphatese > 1.5 times the upper limit of normal and consistent histologic findings on liver biopsy.⁵¹

Primary sclerosing cholangitis (PSC) is a progressive inflammation and fibrosis of the intra and extra hepatic bile ducts. It is estimated that 1 in 100,000 people will be affected. Most are males and the median age of diagnosis is 40.⁵⁴ Of note, 75% of patients with PSC have inflammatory bowel disease (IBD).⁵⁵ PSC leads to cholestasis, progressive hepatic fibrosis and decompensated cirrhosis over the course of 10-15 years.⁵⁶ PSC increases the risk for hepatobiliary and colorectal cancer.⁵⁷ The pathogenesis of PSC is poorly understood. It is believed to be a complex immune mediated disease. There is likely a genetic predisposition that is subsequently triggered by an environmental component.⁵⁸ Diagnosis is primarily made through liver tests and imaging. ⁵⁹ No auto-antibody has been found that is specific to PSC.⁶⁰ Cholangiography shows short, multifocal, annular strictures alternating with normal and slightly dilated intervening segments leading to the classic “beads-on-a-string” appearance.⁶¹

Diagnosis of Hepatic Fibrogenesis

Liver biopsy with histologic examination has been the gold standard for diagnosis and staging of hepatic fibrogenesis. However several non-invasive methods also are available to assist in diagnosis. Ultrasound (US) is the first modality as it is non-invasive, cost-effective and does not expose patients to radiation. Characteristic findings of cirrhosis on US include a coarse nodular appearance of hepatic parenchyma, hepatomegaly, ascites and caudate lobe atrophy.⁶² Computer tomography (CT) is also a commonly used modality in the evaluation and diagnosis of liver fibrogenesis. CT is believed to have sensitivity of 77.1% and specificity of 67.6%.⁶³ Magnetic resonance imaging (MRI) has been used to quantify fibrosis with a sensitivity of 85% and specificity of 100%.⁶⁴ More recently a new method called transient elastography (TE) has been developed. It relies on the principle of shear waves. A transducer emits a 50MHz pressure wave through the liver and the resulting shear wave is measured by US. The shear wave velocity is correlated with liver stiffness, which in turn estimates liver fibrosis. For the diagnosis of cirrhosis, TE has a sensitivity of 83% and a specificity of 89%.⁶⁵ Several serum biomarkers are also available in the non-invasive diagnosis of liver fibrogenesis. Often these biomarkers are described as direct, which reflect extracellular turnover, or indirect, which reflect overall liver function. AST-Platelet Ratio Index (APRI) is a common biomarker used in the estimation of fibrosis. A higher APRI value is indicative of worsening fibrosis. APRI score of 1.0 had a sensitivity and specificity of 76% and 72% respectively for the prediction of cirrhosis.⁶⁶ Fibrotest is a another biomarker panel that uses alpha-2 macroglobulin, haptoglobin, total bilirubin, apolipoprotein-A, GGT, age and gender to calculate score between 0.0 to 1.0, with 1.0 meaning significant fibrosis.⁶⁷ Direct biomarkers include hyaluronic acid (HA), amino terminal of serum procollagen III peptide (PIIINP), tissue inhibitors of metallopreinase-1 (TIMP- 1). HA is a glycosaminoglycan found in the extracellular matrix. It enters circulation during matrix turnover and is degraded in the liver through hepatic endothelial cells. High levels of HA can be due to increased matrix turnover or reduced clearance. PIINP is a marker of collagen turnover with increased levels correlated with tissue repair and fibrosis. PIINP has been found to accurately predict fibrosis in the setting of PBC, NAFLD and viral hepatitis.^68-71 TIMP-1 is an enzyme that inactivates chollagenase with levels found to be higher in patients with liver fibrogenesis.⁷² Enhanced liver fibrosis (ELF) test uses a combination of PIIINP, HA and TIMP-1. It has been found to have a sensitivity and specificity of 90% and 69% respectively in those with chronic liver disease.⁷³

Recommendations for the Primary Care Provider

From this overview on the basics of hepatic fibrogenesis, several recommendations can be made for the primary care physician when co-managing these patients (see table 2). First should be the removal of any liver injury- causing factors such as viral agents, alcohol, toxins and medications. This not only halts the progression of hepatic fibrosis, but it often leads to its regression.³ Alcohol abstinence is the most effective treatment for ALD, and this should be enforced at every encounter as it may completely reverse steatosis.^16,17 Viral hepatitis screening, vaccination, and treatment is paramount since even cirrhosis has been reversed in several patients, when HBV and HCV have been treated.⁹

When it comes to NAFLD, generalists can screen for risk factors of the metabolic syndrome by measuring waist circumference, obtaining body mass index (BMI), and screening for insulin resistance and dyslipidemia.³³ Obesity, especially central obesity, is a major risk factor for NASH and BMI is a good marker for predicting NAFLD.^74,75 Counseling on regular moderate physical activity for 3 to 5 days per week should be recommended.^27,76,77 In addition, high calorie diets that are rich in trans/saturated fat and high fructose-sweetened beverages should be avoided, while low calorie diets supplemented with monounsaturated fatty acids, omega-3 fatty acids, and probiotics should be encouraged.⁷⁸ 6-gingerol, a key component of ginger, and curcumin, a bioactive component in turmeric have both been shown to have anti-inflammatory and antioxidant properties that may be hepatoprotective.^79,80 Pharmacotherapy should include an angiotenstin receptor blocker (ARB) for hypertension, statin therapy, with close monitoring of liver tests, for dyslipidemia, and metformin or pioglitazone for diabetes.^24,25,34 Vitamin E and pentoxifylline may have a role for NASH, but need further study.²⁰

The advent of antiviral therapy has revolutionized the management of viral hepatitis. In managing HBV there is now strong evidence that antiretroviral therapy lowers disease progression and the incidence of HCC in patients with high serum HBV DNA levels and advanced liver disease.⁸¹ Recently several oral direct acting antiviral medications that target different stages of the HCV life cycle have become available. Treatment using these agents is genotype specific. Treatment of HCV in HIV coinfected patients can be a challenge as there are extensive drug interactions with HIV antiretrovirals.^82,83

Treatment of PBC and PSC are limited. For patients with PBC, treatment with ursodeoxycholic acid (ursodiol) has been shown to delay progression of hepatic fibrosis.⁸⁴ Other medications such as colchicine and methotrexate may be effective.^85,86 There are no medical therapies that alter the natural course of PSC. Liver transplant is the only definitive treatment for patients with advanced disease. Overall prognosis in PSC remains poor with a 12 year median time from diagnosis to death or liver transplant.87 Acute decompensation can occur in patients with PSC due to sudden obstruction of the hepatobiliary system. Relieving obstruction can improve outcomes.

CONCLUSION

Hepatic fibrogenesis is a dynamic process with a multitude of etiologies. In the United States there are over 100,000 hospitalization and 36,000 deaths from liver disease.88 Etiologies include ALD, NAFLD, viral hepatitis and cholestatic liver disease. Diagnostic modalities ranging from invasive liver biopsy to non- invasive imaging and serum markers provide physicians with an array of options for further evaluation of suspected liver disease. In addition to the etiology specific treatments available, general measures can be taken to prevent and arrest the progression of hepatic fibrogenesis.

NUTRITION ISSUES IN GASTROENTEROLOGY, SERIES #140

Guide to Front Line Drugs Used in the Treatment of Short Bowel Syndrome

Lingtak-Neander Chan,

John K. DiBaise,

Carol Rees Parrish

In patients with short bowel syndrome (SBS), malabsorption of drugs is an important consideration, particularly if the expected clinical response to a medication is not attained. Factors include length, location and health of the remaining bowel, the form of the medication administered and the site of action of the drug. Part 4B of this 5-part series will focus on conventional pharmacological agents used in the treatment of SBS.

As you may recall from Part IV-A of this series, a 64-year-old woman is hospitalized with hypotension, chronic diarrhea (6-10 per day in the last 4 weeks), fatigue and general malaise. Seven weeks ago, she was hospitalized with acute abdominal pain and severe nausea. Further work-up revealed acutemesenteric infarction due to a superior mesenteric artery thrombus and she underwent extensive resection of the ischemic bowel leaving her with about 100 cm of small bowel from the ligament of Treitz anastomosed to the proximal transverse colon. She was discharged from the hospital 2 weeks later on home parenteral nutrition (PN). Her weight is down 6 kg since hospital discharge 5 weeks ago. In addition to PN, she also takes loperamide 2 mg TID as needed (she has not been using it because “it doesn’t help”), cholestyramine 4g BID, mirtazapine 15 mg daily, levothyroxine 0.075 mg daily and oral glutamine 30g daily. Famotidine, 40 mg, has been added to her PN. Stool output, which is watery and without blood, ranges from 2500 to 3000 mL daily. She is not eating much as she is so depressed and uncomfortable and “it just comes right out anyway”. She is afraid to leave her house for fear of “having an accident”. Part 4B of this 5-part series on SBS focuses on conventional anti-motility agents used in the treatment of short bowel syndrome that will allow you to help this woman control her diarrhea and improve her quality of life.

Lingtak-Neander Chan, PharmD, BCNSP, Associate Professor of Pharmacy and Nutritional Sciences, University of Washington, Seattle, WA. John K. DiBaise, MD, Professor of Medicine, Mayo Clinic, Scottsdale, AZ. Carol Rees Parrish MS, RD, Nutrition Support Specialist, University of Virginia Health System Digestive Health Center of Excellence, Charlottesville, VA

INTRODUCTION
Chronic Diarrhea

While gastric hypersecretion is a transient complication in SBS, chronic diarrhea is often a long-term and sometimes quite disabling complication experienced by most patients and requires continued intervention and monitoring. A recent study involving patients with SBS found that chronic, uncontrolled diarrhea has a more negative impact on quality of life than home PN therapy.¹ Therefore, optimizing the antidiarrheal regimen is a critical component in the care for patients with SBS. Established antidiarrheal regimens in the management of SBS are summarized in Table 1.

Antimotility Agents

Since opioid receptors are widely distributed in the GI tract and play a key role in the regulation of GI tract motility, secretion, and sensation,² opioids should be considered the first-line antimotility therapy in the treatment of chronic diarrhea in patients with SBS. Additionally, opioid receptors also regulate chloride secretion and luminal water movement in the colon. Opioid agonists therefore also exert an antisecretory effect on the colon. High-level clinical evidence supporting their use is unavailable in the literature. This absence of evidence, however, is not evidence of lack of clinical utility. The lack of high-level published data can be explained by the fact that these compounds have been available for use for decades and their role in clinical management has generally been accepted. There is also a lack of more established alternative therapy in managing patient’s symptoms. Additionally, there is no financial incentive for drug manufacturers to conduct such clinical trials. Opioid derivatives used as antimotility agents can be loosely categorized as locally-acting agents with low systemic effects (e.g., loperamide, diphenoxylate) and systemic agents (e.g., codeine, morphine).

Loperamide (e.g., Imodium A-D)

Loperamide is one of the most widely used over-the- counter anti-diarrheal drugs. It is a synthetic opioid. With low systemic absorption and poor penetration across the blood-brain barrier, loperamide is a safe antidiarrheal agent with minimal systemic side effects. Its antimotility action is mediated by both binding to opioid receptors in the intestine as well as inhibiting calcium channels and calmodulin in intestinal smooth muscle.3 It also has an antisecretory effect on the colonic epithelial cells.⁴ Since the oral bioavailability is negligible, loperamide is usually well-tolerated; however, factors that alter the pharmacokinetics and pharmacodynamics of loperamide may potentially exacerbate the untoward effects.

In SBS, an initial dose of 4 mg (2 capsules or tablets, or 30 mL) every 6 to 8 hours is recommended. Loperamide should be taken 30 to 60 minutes prior to a meal and again at bedtime. While the maximal recommended daily dose is 16 mg in generally healthy individuals, in the SBS patient, a dose up to 8 mg every 6 hours (32 mg total) may be needed. In patients with inadequate response (continued watery diarrhea in spite of taking 8 to 16 mg per dose; total dose at 32 mg/day or higher), other pharmacotherapeutic approaches, such as narcotics, etc., should be considered.

Diphenoxylate (e.g., Lomotil®) and Difenoxin (Motofen®)

Diphenoxylate is also a synthetic opioid and likely exerts its antimotility effect by binding to the opioid receptors in the intestine. At low doses, diphenoxylate primarily has antimotility effects; however, at high doses (e.g., over 40 mg daily), it produces typical opioid systemic effects such as euphoria and sedation suggesting an increased systemic effect. Difenoxin is an active metabolite of diphenoxylate. It is rapidly and extensively absorbed after oral administration with peak effects occurring within 40 to 60 minutes in most patients. Although commercially available in the U.S., it is less commonly used than diphenoxylate.² To reduce the abuse potential for these drugs, they are formulated with the anticholinergic agent atropine. Excessive use of diphenoxylate/difenoxen-atropine combination can lead to symptoms ranging from unpleasant sensations similar to those with cholinergic discharge, to serious cardiovascular symptoms such as palpitation and tachycardia.

The typical dose for diphenoxylate is two tablets (or 10 mL) four times daily as needed; generally before meals and at bedtime. Although atropine is an anticholinergic agent and may decrease GI secretion, the amount in the combination is too low to cause clinically significant side effects other than acting as a deterrent for opioid abuse.

Given the extensive clinical experience, both loperamide and diphenoxylate can be considered as first-line antimotility agents in SBS. Based upon both the pharmacokinetic and side effect profile, loperamide is usually considered the preferred agent for initial therapy.

Paregoric

Paregoric, USP, also known as camphorated tincture of opium, is an oral liquid that contains 0.4 mg/mL of anhydrous morphine as its main active ingredient. Paregoric also contains glycerin, benzoic acid, and generally 45% alcohol. Although paregoric elixir has been used to manage pain, GI discomfort, and other ailments since the early eighteenth century, its role in the treatment of chronic diarrhea has diminished considerably because of its high potential for addiction and abuse. Drugs with more limited effect on the brain such as loperamide and diphenoxylate are preferred. According to one study, 4 mL of paregoric solution is approximately equipotent to one diphenoxylate/atropine tablet in controlling diarrhea.⁵

Opium tincture

Opium tincture, USP or deodorized tincture of opium typically contains 10 mg/mL of morphine. One of the most dangerous medication errors is the confusion between paregoric (especially if labelled as camphorated tincture of opium) and opium tincture. Given the same volume, there is a 25-fold difference in the amount of opiate extract between these products. If prescribed incorrectly, serious overdosing and fatalities can result. To minimize the risk of medication error, it is helpful to remember that opium tincture should be dispensed and administered with small droppers or oral syringes and each dose should not exceed 1-2 mL, every 6-8 hours when treating diarrhea. Because safer and possibly more effective alternatives such as loperamide are available, both paregoric and opium tincture have very limited role in the management of SBS. See Tables 1 and 2 for clinical considerations and cost when prescribing opium tincture.

Codeine and Morphine

Both codeine and morphine can be used as antimotility and antisecretory agents in patients with SBS as both compounds bind to the opioid receptors in the GI tract leading to decreased GI secretion and increased orocecal transit time. Although codeine has an antidiarrheal effect, about 10% of an administered dose is metabolized by the enzyme CYP2D6 to morphine, a more potent metabolite.^6,7 In patients with decreased CYP2D6 enzyme activity (i.e., poor metabolizers), either from genetics or drug interactions (e.g., CYP2D6 inhibitors such as citalopram, fluoxetine, paroxetine), the antidiarrheal effect of codeine could be reduced compared with patients having normal CYP2D6 activity.⁸ Conversely, DNA variations also occurs in a subgroup of patients resulting in more active CYP2D6 enzyme activities. These “ultra-rapid metabolizers” convert codeine to morphine faster and more completely than in other people and are more likely to have higher than normal amounts of morphine in their blood after taking codeine. High levels of morphine can result in respiratory depression and failure, which may be fatal.

Data from an international clinical pharmacogenetics databank show that 77-92% of the population are phenotypically “normal” CYP2D6 metabolizers. The allele frequency for poor-metabolizer is about 5-10% and ultra-rapid metabolizer between 1-2%.9 Therefore, codeine is generally a safe and effective treatment option for a majority of the patients as long as the dose is initiated cautiously and patient-specific side effects are carefully monitored (e.g., sedation, rash, nausea, respiratory depression). In at-risk patients, such as children, patients with adverse reactions or unpredictable response to codeine previously, or if there is concern about a particular patient’s CYP2D6 phenotype based on his/her clinical response/drug dosing to other CYP2D6 substrates (e.g., tricyclic antidepressants), CYP2D6 genotyping should be performed before drug initiation to optimize the patient’s codeine regimen and minimize serious adverse events.⁷ Genotyping can be performed using tissue samples obtained from either buccal swabs or whole blood samples. Before ordering the test, clinicians should discuss with the laboratory personnel in their institution to ensure that samples are properly collected so the test results can be accurately documented in the patient’s medical record. Pharmacogenetic testing is now covered by most insurance companies.

Since both codeine and morphine are well-absorbed, systemic side effects are the primary concerns when used as antimotility agents. Renal failure can increase the accumulation of morphine and its active metabolites and potentially increases the risk of experiencing systemic side effects; in this situation, start at one-half the usual dose.

Considerations when Using Opioid-based Antidiarrheal Agents

When a patient with SBS (or non-SBS high ostomy output) is showing inadequate response to aggressive dosing of loperamide or diphenoxylate, clinicians often find themselves in a clinical conundrum µ When should a systemic opioid be used? Should I add it to the other antidiarrheal or should I stop the other drug first? Does the bowel anatomy determine whether or which opioid to use? What is the treatment endpoint? Would the use of opioid be in violation of federal regulations? Could drug addiction become a problem for the patient? How much opioid can I prescribe? In the following, we attempt to provide some guidance and clarification on these points.

Threshold to Start Systemic Opioid

There is no established threshold and high quality research is sorely needed to guide practice decisions. Based on the pharmacology and physiology, it is our opinion that starting a systemic opioid drug is reasonable when the following criteria are met:

The patient has received the maximal
recommended doses of the first-line
antimotility agents and:
- Stool output remains over 1500 mL per day, OR
- Recurrent admissions for dehydration with stool/ostomy output < 1500mL, OR
- The reduction of stool output from baseline is less than 50%, OR
- The patient remains dehydrated with persistent orthostatic hypotension, OR
- The patient develops intolerable side effects from the first-line agent(s) that are unrelated to the systemic side effects of narcotics (e.g., bloating), OR
- The total pill count reveals that it is unreasonable to ask the patient to take 16-24 pills to slow their gut, when potentially a single 30 mg tablet of codeine QID will be as effective, or even more so.
The patient cannot tolerate or has developed adverse events from the first-line antimotility agents;
There are no clinical contraindications to using systemic narcotics (e.g., allergic reactions, delirium, underlying respiratory problem such as obstructive sleep apnea); and
The patient has no known history of drug abuse (although this should not be an absolute contraindication; 1-2 septic episodes on PN/IV fluids and the potential for recurrent abuse may be the lesser of evils).

Is there benefit to use both loperamide and systemic narcotics at the same time? What if a patient does not have an ileum µ can loperamide still be utilized?

Opioids interact and regulate the enteric nervous system primarily through 3 different opioid receptors µ, δ, and k receptors. Binding of opioids to these receptors results in a decrease in intestinal motility and secretion. Loperamide is only selective to m opioid receptors. Therefore, the addition of the non-selective agonist, such as morphine, may provide a synergistic effect in reducing GI motility and secretion, especially in patients with insufficient response to loperamide alone. The combined use of loperamide and codeine was demonstrated in a randomized, controlled study to be safe and exert a synergistic effect on ileostomy output.¹⁰ Although animal data have shown that loperamide has a more profound effect in the small intestine than in the colon, it can still be used in patients with a significant portion of the ileum removed because opioid receptors are also present in the colon.^3,11 Loperamide will provide some antisecretory effect and can reduce the frequency and severity of diarrhea.

What is the treatment endpoint and duration of therapy?

The treatment endpoint is patient-specific but should be directed to improving the frequency and severity of diarrhea and preventing dehydration. The patient should be assessed regularly for treatment response and side effects. The goal when using a narcotic agent as an anti-diarrheal is to use the lowest effective dose to achieve the overall therapeutic endpoint. It is reasonable to titrate the narcotic down every few weeks as long as the patient is clinically stable.

What are some of the reasons clinicians give for not using narcotics in the treatment of patients with short bowel syndrome?

Risk of narcotic addiction
- This begs the question, is this risk higher than the risks involved with keeping a patient on PN or IV fluids requiring a central venous catheter? The goal of using any narcotic drug is to use the minimal effective dose and continually reassess the need for this therapy. The risk of causing narcotic addiction in a patient with no history of substance abuse is minimal.

Drug abuse potential
- There is no doubt that a clinician should carefully select the “best” patients to receive narcotic therapy. Patients with active drug seeking or narcotic abuse behavior should generally not be considered appropriate candidates to receive prescription narcotic drugs as antimotility, antidiarrheal agents. At the same time, clinicians must carefully assess whether it is appropriate to discharge a patient with a documented history of narcotic abuse with a central catheter in place even though the intention of the catheter is for IV fluids or PN.
Narcotics require a monthly prescription to be handed directly to the patient. Prescriptions cannot be faxed/mailed directly to the pharmacy.
There are concerns that using high doses of narcotics may result in an audit from the Drug Enforcement Agency (DEA). While it may be easy to support this practice from a busy physician’s office that sees a fair number of SBS patients, it may require the office to be closed down while the audit is completed.

Overall, there is much misconception and lack of understanding among healthcare providers about the use of narcotic agents in the management of chronic diarrhea. Although this indication is an off-label use of narcotic drugs, the rationale is well-supported by science and the efficacy is supported by clinical literature. In the process of evaluating the appropriateness of narcotic therapy, clinicians should follow federal and state regulations, exercise good clinical judgment in patient selection, and implement a patient-specific monitoring plan. Our experience suggests that a carefully executed management plan with low-dose narcotic therapy can reduce PN use and unnecessary hospitalizations.

SUMMARY

Anti-diarrheal agents are part of the cornerstone of
therapy in the treatment of short bowel syndrome. If
not well controlled, not only are hydration and nutrient
utilization at risk, but also overall quality of life is
seriously threatened. It is of particular importance in
the SBS population to consider the site of absorption,
formulation, frequency, and timing of each drug, as well
as cost and availability before it is ordered. Setting and
monitoring endpoints for each intervention (i.e., how
long you will give each intervention time to achieve
efficacy or not), is also fundamental. Keep in mind that
there are only so many hours in the day, and patients
need to fit “life” in between all their medications,
meals/snacks and fluids. When the “Total Pill Count”
of the lesser gut slowing medications exceeds a certain
number, it is time to use stronger agents that will not
only be more effective, but also require fewer pills per
day.

FRONTIERS IN ENDOSCOPY, SERIES #17

Evaluation and Therapy of Pancreatic Cysts

Serge A. Sorser,

Ali Siddiqui,

Douglas G. Adler

Multiple societal guidelines have been established to assist clinicians in directing the care of patients with pancreatic cysts, but the natural history and malignant potential of all cystic lesions is not entirely understood. In this article, we aim to describe the common types of pancreatic cysts, their natural history, means of follow up, and possible modes of endoscopic and non-endoscopic intervention.

Serge Sorser, MD,¹ Ali Siddiqui, MD² and Douglas G. Adler, MD, MD, FACG, AGAF, FASGE, Associate Professor of Medicine, Director of Therapeutic Endoscopy, Director, GI Fellowship Program, Gastroenterology and Hepatology3 ¹Providence Hospital, Southfield Michigan ²Jefferson University School of Medicine 3University of Utah School of Medicine, Huntsman Cancer Center, Salt Lake City, UT

Pancreatic cysts are a common incidental finding as a result of increased use of diagnostic cross sectional imaging. They run the gamut from benign disease processes to pre-cancerous entities to frank malignant lesions. Much has been written in regards to the natural history of pancreatic cysts, their presentation, diagnostic evaluation, endoscopic and surgical intervention and surveillance. In this article, we aim to summarize the natural history/epidemiology of cysts, their evaluation, and future endeavors in the management of pancreatic cysts.

INTRODUCTION

The increased use of cross-sectional imaging in recent years has led to more incidental findings being noted.¹ The prevalence of pancreatic cysts has been documented to range from 2.6% to 13.5%.^2,3 The majority of these cysts are asymptomatic, but further work-up is often warranted once they are noted.⁴

Subsequent evaluation may be performed in the form of further imaging, but fluid acquisition/analysis helps delineate the nature of the cyst, its potential for malignant transformation, possible need for surgical intervention and further surveillance. Endoscopic Ultrasound guided fine Needle Aspiration (EUS- FNA) plays a crucial role in cyst evaluation.^5,6 Multiple societal guidelines have been established to assist clinicians in directing the care of these patients, but the natural history and malignant potential of all cystic lesions is not entirely understood.^7,8,9 In this article, we aim to describe the common types of pancreatic cysts, their natural history, means of follow up, and possible modes of endoscopic and non-endoscopic intervention.

Serous Cystadenoma (SCA)

SCA, previously known as a microcystic adenomas, are benign entities and the second most common cystic tumors of the pancreas, accounting for up to 30% of pancreatic cysts.^10,11 These tumors occur most commonly in the body and tail of the pancreas and are often seen in middle-aged women, but can occur in both sexes and at any age.^12,13 Patients are commonly asymptomatic, but may present with abdominal pain and a palpable mass, depending on the size of the tumor.¹⁴

Cross sectional imaging may help make the diagnosis, as the tumor appears as a multi-septated cyst with so-called “honeycombing.”¹⁵ A central, spiculated (“sunburst”) calcification may also be seen.¹⁶ EUS FNA is commonly performed if the diagnosis of SCA is less than certain, but some advocate that if a lesion is classic for SCA, an EUS guided-FNA may not be needed.^17,18 (Figure 1) When fluid for analysis is obtained, it is often noted to be clear with cuboidal cells lining the cyst cavity, although acellular fluid is also commonly obtained.¹⁹ A low cyst fluid CEA and bland cytology are frequently noted on fluid analysis.

Due to the benign nature of SCA, no surveillance is generally felt to be warranted. SCA have a low rate of malignant transformation, quoted as less than 3%.²⁰ Surgical intervention is not indicated for serous cystadenomas, unless they are symptomatic. However, recent literature recommends consideration of surgical intervention if the patient has symptoms that can be attributed to the lesion and if the cyst in aggregate is greater than 4 cm.^21,22 It should be stated that many lesions that are, in fact, SCAs do not have all of the classic findings of these lesions and it can sometimes be hard to distinguish SCA from other, more ominous lesions.

Mucinous Cystic Neoplasm (MCN)

MCNs are the most common type of pancreatic cysts. They constitute up to one half of all known cystic lesions of the pancreas. They range in size from 5 to 35 cm and are predominantly found in females.^23,24 The age of onset is usually in the fifth or sixth decade of life and the tumor tends to localize in the body or tail of the pancreas.^{25, 26, 27} These cysts are defined strictly by the presence of ovarian type stroma within the tumor.^28,29 On cross sectional imaging, no communication with the main pancreatic duct is typically noted. On fluid analysis, thick and mucoid material is typically found, with a low amylase and an elevated cyst fluid CEA level.^30,31 Histologically, these cysts are mostly benign. An adenoma was noted in 72% of the cases, borderline neoplasm in 10.5%, carcinoma in situ in 5.5% and invasive cancer in 12% of patients in a series of 163 patients.³² Given that malignant transformation may occur via K-ras and p53 mutations, surgical resection should be considered for MCNs in patients who are suitable operative candidates.^33,34,35,36 The cysts are typically unifocal and when the lesion is resected and is noted to be non-invasive, no surveillance is typically required although in practice many patients undergo post-treatment imaging periodically.⁸

Pancreatic Pseudocysts

Pancreatic pseudocysts are most commonly a complication of acute or chronic pancreatitis, although they can also occur following trauma to the pancreas. Pseudocysts are rich in amylase and are not lined by an epithelium.³⁷ The underlying etiology is multifactorial, but ultimately leads to ductal disruption and an increase in pancreatic ductal pressure.³⁸ Patients tend to present with ongoing abdominal pain and anorexia weeks after their initial presentation, with rare complications such as jaundice or sepsis also noted.³⁹ Jaundice can result from extrinsic compression of the bile ducts, and sepsis from secondary infection of the cyst itself can be seen. Large cysts frequently compress the stomach and/or duodenum and can cause gastric outlet obstruction. Pseudocysts are usually distinguished from other pancreatic fluid collections by the lack of significant solid debris (as is more commonly seen in walled off pancreatic necrosis (WOPN). The diagnosis of a pseudocyst is made by cross-sectional imaging. CT scans will, in general, underestimate the amount of solid debris within a lesion.⁴⁰ Other modes of evaluation, including MRI or EUS may be considered.⁴¹ MRI and EUS will give a more accurate assessment of the amount of solid debris with in a pancreatic fluid collection.

Numerous studies have been published regarding cyst fluid analysis to help distinguish pseudocysts from mucinous cystic neoplasms.^42,43,44 In general, pancreatic pseudocysts will have a high cyst fluid amylase with a low cyst fluid CEA. Pancreatic cyst fluid is often laden with debris and macrophages and often has a “dirty” chocolate brown color. Endoscopic, surgical and interventional radiology approaches may be offered for treatment/drainage of the cysts.⁴⁵

Symptoms from pseudocysts that require treatment include pain, infection, hemorrhage into the cyst, and compression of the stomach, bowel, and/or bile duct. Many symptomatic pancreatic pseudocysts can be treated endoscopically with transampullary drainage via pancreatic duct stent placement. In some patients the cyst decompresses through the pancreatic duct via the stent, while in other patients the pancreatic stent simply relieves pressure on the pancreatic duct; this stops the backfilling of the cyst, allowing it to resolve over time.⁴⁶ A recent paper by Lin et al. showed that transpapillary drainage may be an adequate approach in up to 79.5% of patients.⁴⁷

If transmural drainage is desired, technique usually involves puncturing the stomach or the duodenum to gain access to the cyst via EUS under fluoroscopic guidance, delineating the cyst cavity, dilating the tract and placing multiple plastic double pigtail stents or single metal stents. ^48,49,50,51 (Figure 2) Some patients warrant both transmural and transampullary drainage simultaneously.⁵² Surgical management involves creation of a cyst-enterostomy in the most dependent part of the cyst cavity.^53,54 The percutaneous approach involves finding the most appropriate window (transperitoneal, retroperitoneal, transgastric, transduodenal or transhepatic) and placement of a Percutaneous External Drain.^55,56 Percutaneous approaches are the least invasive and can be favored in patients who are poor candidates for other interventions, although they result in external drainage and there is a risk of chronic cutaneous fistula development.

Walled-off Pancreatic Necrosis (WOPN)

WOPN, a complication of necrotizing pancreatitis, is defined as a collection of fluid and solid components that tends to develop 3 to 6 weeks after an episode of pancreatitis, although some patients can develop an immature form of this lesion in a shorter timeframe.⁵⁷ WOPN is seen in 1 to 9% of cases of acute pancreatitis and occurs most commonly after biliary pancreatitis.⁵⁸

Patients typically present with ongoing abdominal pain as well as fever and leukocytosis.⁵⁹ Fever and leukocytosis can be present even in the absence of infection. The diagnosis of WOPN is typically made on cross-sectional imaging by visualizing a non-enhancing pancreatic fluid collection, which may contain solid and liquid debris, correlated with the age of the fluid collection and the presence of a surrounding capsule.⁶⁰ Indications for intervention include clinical suspicion of, or documented, infected necrotizing pancreatitis with clinical deterioration, gastric outlet, intestinal, or biliary obstruction due to mass effect of walled-off necrosis or persistent symptoms in patients with walled- off necrosis without signs of infection.⁶¹

While many therapeutic modalities for WOPN exist, not all infected pancreatic necrosis requires intervention. Multiple case series have reported good clinical outcome in patients treated conservatively with a prolonged course of antibiotics and supportive care.^62,63 This was later supported by a meta-analysis of eight studies, including 324 patients, which noted that conservative management without necrosectomy is a successful approach in 64% of patients.⁶⁴

In patients who are deemed candidates for intervention, multiple treatment modalities are available. In stable patients, therapy should be delayed a suitable amount of time (usually 4 weeks or more) to allow liquefaction of the contents and the development of a fibrous wall around the necrosis.⁶⁵ This timeframe also allows the fibrous wall to adhere to the stomach or duodenum if endoscopic approaches are to be undertaken.

Multiple endoscopic drainage methods are available for patients with WOPN. One option is endoscopic transmural drainage, in which one or more transmural tracts are created with EUS guidance between the necrotic cavity and the GI lumen. These tracts can be flushed with saline or a mixture of saline and hydrogen peroxide via endoscopic means or via a nasocystic catheter. The tracts can be held open via plastic or metal stents per physician preference.⁶⁶ In the combined percutaneous/endoscopic techniques, a large caliber percutaneous catheter can be used for irrigation of a cavity that has been accessed endoscopically to provide multiple routes for irrigation and drainage.⁶⁷

Recently, the use of dedicated, covered, transmural self-expanding metal stents has been described, in which a short, barbell shaped metal stent is deployed and apposes the pancreatic cyst to the gastric cavity.⁶⁸ (Figure 3) The use of fully covered esophageal and biliary stents has also been noted for these purposes.⁶⁹ A tailored endoscopic approach has also been proposed, which is based on size and extent of the walled-off necrosis and stepwise response to intervention.⁷⁰

Patients undergoing necrosectomy by any route constitute a high risk population; procedure related complications are as high as 25%. These complications include bleeding, sepsis and perforation and the procedure has an overall success rate of 82-93%.^71,72 It should be noted that some patients will fail endoscopic approaches and still require a traditional surgical necrosectomy and/or percutaneous drains.

The radiologic approach has been noted to be safe and feasible. This technique is minimally invasive and has an overall success rate of 33-56% in resolving the WOPN.^73,74 Complications of the percutaneous approach include internal and external pancreatic fistulas, with an overall mortality rate of 17.4%.⁷⁵

Surgical approaches to patients with WOPN are well described and are now often performed through minimally invasive/laparoscopic techniques, although some patients still require an open necrosectomy. With the laparoscopic approach, a transgastric endolumenal cystogastrostomy is created.⁷⁶ Common adverse events include pancreatic fistulae (28.6%), debris recollection (10.7%) and wound infection (10.7%).⁷⁷

While each individual approach has its advantages and disadvantages, a combined/multidisciplinary approach may be needed. A 2012 study by Gluck et al. showed that dual modality (endoscopic and percutaneous) drainage reduced length of stay, number of radiological procedures and number of ERCPs with a durable long-term outcome (100/103 patients did not require surgery at two years).^78,79 A multidisciplinary approach should be undertaken and the treatment modalities selected should rely on individual center expertise, but also depend on the anatomical position, the ratio of solid to fluid components within the collection, and the degree of systemic organ dysfunction.^80,81

Intraductal Papillary Mucinous Neoplasm (IPMN)

IPMNs are mucin producing lesions of the exocrine pancreas. They account for up to one third of pancreatic cysts, but are felt to be responsible for only one percent of pancreatic cancers.⁸² They may be subcategorized in terms of their ductal involvement: main duct (16-30%), side branch (40-65%) or mixed type (15-23%).^83,84 Most IPMN are solitary and are located in the pancreatic head, but 20-40% may be multifocal.⁸⁵ Histologically, the tumors are graded as having low-grade dysplasia, intermediate grade dysplasia and high-grade dysplasia. IMPN are also sub-classified into four different types: gastric, intestinal, pancreaticobiliary and oncocytic. This classification is descriptive and indicative of different pathways of differentiation and progression to carcinoma.^58,86

While IPMN are usually incidentally found on imaging, diagnostic evaluation with EUS is commonly undertaken for a more detailed evaluation and for cyst fluid aspiration and analysis.⁸⁷ (Figure 4) The cyst content may be analyzed in a number of ways, including mucin stain and viscosity, and cyst fluid CEA level, although all of these tests can be limited when attempting to identify malignancy.^88,89 DNA studies of cyst fluid are also available but are not in widespread use. IPMN may be malignant at presentation, but they carry a better prognosis than pancreatic adenocarcinoma in this setting.⁹⁰ Worrisome features of IPMN lesions include size greater than 3 cm, presence of mural nodules, dilation and/or involvement of the main pancreatic duct and cyst location (main duct versus side branch).⁹¹ Surgical intervention includes pancreaticodeuodenectomy, distal pancreatectomy, total pancreatectomy, segmental resection, enucleations and duodenum preserving resections.^92,93,94 For patients who are not surgical candidates, endoscopic ablation of the cyst cavity with ethanol has been described but can only be considered experimental at this time.^95,96,97 Surveillance strategies after definitive therapy are guided by the Sendai Criteria, and are based on clean surgical margins, extent of dysplasia and whether known cystic lesions remain in the pancreas.⁸

Rare Pancreatic Cystic Lesions

There are a number of relatively rare cystic lesions of the pancreas or solid lesions with cystic degeneration or solid lesions with cystic components that are also worthy of brief discussion.

Cystic lymphangioma of the pancreas arise from lymphatic vessels, and this is thought to be developmental aberrancies. They account for 0.2% of all pancreatic cysts and are most often noted incidentally.⁹⁸ These tumors are benign, but may be locally invasive, and are more commonly found in women.⁹⁹ Symptomatic lymphangioma patients usually present with epigastric pain and a palpable mass although they can be asymptomatic as well.¹⁰⁰ Review of histology yields interconnecting cysts separated by septa, lined by epithelial cells, and contain serous, serosanguineous, or chylous fluid (elevated triglyceride level).¹⁰¹ Given their benign nature, no further work-up is needed and lesions can be resected based on symptoms as needed.¹⁰²

Lymphoepithelial cysts are also benign cystic entities, most commonly seen in men and also typically discovered incidentally.¹⁰³ The diagnosis is made by EUS/FNA showing abundant anucleated squamous cells, multinucleated giant cells, mature lymphocytes in a background of keratinaceous debris and a lack of neoplastic cells.¹⁰⁴ Surgery is not recommended in most patients unless they are symptomatic.¹⁰⁵

Other rare cystic tumors include cystic degeneration of ductal adenocarcinoma¹⁰⁶ and solid pseudo-papillary tumors of the pancreas^107,108 and other mesenchymal origin cysts.¹⁰⁹

CONCLUSION

Cystic lesions of the pancreas are more commonly
encountered with increased use of cross sectional
imaging for evaluation of gastrointestinal and other
symptoms. These lesions have a wide range of presenting
symptoms, while most are asymptomatic. EUS/FNA
plays a key role in the diagnostic work-up, and offers
prognostic value, with surveillance recommendations
made based on the cyst size and fluid characteristics.
Depending on the type of cyst, endoscopic, radiologic
and/or surgical modalities may be employed in treating
the underlying pathology. No definitive guidelines
exist for surveillance of all the known cyst types, and
a tailored approach is recommended in many cases. In
the future, genetic profiles and tumor markers may play
a role in improving treatment strategies.