Practical Gastroenterology publishes articles for the primary care physician, and your article should therefore have a nuts-and-bolts slant. We urge you to keep the nonspecialist in mind as you write your article. We cannot stress strongly enough the importance of focusing your article on information that will be useful and instructive to the primary care physician. In this regard, it would be helpful for you to emphasize prevention and cost (of tests, drugs, surgery, hospital stay, procedures, techniques. etc.) whenever and wherever possible.
INFLAMMATORY BOWEL DISEASE: A PRACTICAL APPROACH, SERIES #107INFLAMMATORY BOWEL DISEASE: A PRACTICAL APPROACH, SERIES #107
The Use of Vedolizumab in Pregnancy and Breastfeeding in Women with Inflammatory Bowel Disease
Kerri Glassner
Bincy P. AbrahamKerri Glassner DO. Bincy P. Abraham MD MS AGAF FACG Fondren Inflammatory Bowel Disease Program, Lynda K and David M Underwood Center for Digestive Disorders, Division of Gastroenterology and Hepatology, Houston Methodist Hospital and Weill Cornell Medical College.
Many women with inflammatory bowel disease (IBD) are diagnosed during their childbearing years. Accordingly, patients and their providers may experience significant anxiety about the impact of their disease course and medications on pregnancy and breastfeeding. Unfortunately, it is not uncommon to encounter a pregnant IBD patient who presents with active disease after stopping their medications, either by personal choice or on advice from their obstetrician or gastroenterologist. Active disease at the time of conception and during pregnancy has been associated with poor pregnancy outcomes including an increased risk of preterm birth, low birth weight, and fetal loss.1-7 In fact, the greatest risk to the mother and fetus during pregnancy has been found to be active IBD, not the medications used to treat it.7-10 Recently, there has been increased attention on the importance of preemptive counseling of women with IBD and their providers on the role of disease control on pregnancy outcomes.
Vedolizumab is a humanized immunoglobulin G1 monoclonal antibody targeting a4ß7 integrin. Unlike anti-TNF agents or immunomodulators, vedolizumab is gut selective, and works by inhibiting leukocyte migration into inflamed intestinal tissue via the a4ß7 integrin expressed on circulating B and T lymphocytes.11-14 Vedolizumab is FDA approved for the treatment of both ulcerative colitis and Crohn’s disease. Animal studies have shown that mucosal vascular addressin cell adhesion molecule 1 (MAdCAM-1) is expressed by maternal vessels in the placenta and recruits a4ß7 expressing cells that are considered important for maternal fetal tolerance.15 However, it is unknown what effect, if any, vedolizumab use in humans has on maternal placental vessels.
Pregnancy Outcomes
Few studies have been published on the use of vedolizumab during pregnancy. A review of the literature identified 6 publications with a total of 104 pregnancies in women who received vedolizumab during conception and/or pregnancy with available birth outcome data which can be seen in Table 1.16-21 Among these, there were 80 (77%) of pregnancies that resulted in at least one live birth, 6 (6%) congenital anomalies, 11 (11%) spontaneous abortions, 8 (8%) elective terminations, and 1 (1%) stillbirth or pregnancy loss at greater than 20 weeks gestation. There were 15 (14%) preterm births (less than 37 weeks gestation).
Based upon data from the CDC, the live birth rate in 2008 among an estimated 6,578,000 pregnancies in women in the United States was 65% and is shown in Table 1.22 Previous data from pregnancies among women with IBD showed the live birth rate to be 60%, lower than in women without IBD.5 More recent data from the TREAT registry following women with IBD who were treated with infliximab as well as those without biologic exposure, found a much higher live birth rate of 91.1%, Table 1.23 Based upon the available data in women with IBD using vedolizumab during pregnancy, the live birth rate of 77% appears to be higher than that of both women in the general United States population, and of that seen in certain studies of pregnancy outcomes in women with IBD.5,16- 22 The differences seen in live birth rates may be related to other factors including small sample size, outcome reporting, or closer monitoring of mothers on biologic therapy. In the general population, the rate of spontaneous abortion or miscarriage is 10-17% of clinically recognized pregnancies, and typically occurs in the first trimester.24 In the TREAT registry, the spontaneous abortion rate was 16% in infliximab exposed pregnancies, and 8.9% in non-biologic exposed pregnancies.23 The overall rate of spontaneous abortions in women with IBD on vedolizumab appears to be similar to that of both the general population and infliximab exposed pregnancies.16-21, 23-24 The pre-term birth rate in the United States has been found to be 9-10%.25 Previous studies have shown that there is an increased risk of premature delivery in patients with inflammatory bowel disease, especially those with more severe disease activity.1,26-27 The TREAT registry had a low pre term birth rate of 3.8% in infliximab, and 4.7% in non-biologic exposed pregnancies.23 In vedolizumab exposed women the pre-term birth rate was higher at 14%, but comparable to other studies of pregnancy outcomes in women with IBD.5 The higher rate of pre-term birth seen in the vedolizumab group compared to the TREAT registry may be explained by the fact that vedolizumab is often a second or third-line therapy used in patients with severe disease after having failed multiple biologics. It is known that more severe disease activity correlates with an increased risk of pre-term birth.7 In the United States, the stillbirth rate is 1%.28 The current literature on vedolizumab is similar, with a stillbirth rate of 1%.
Major structural birth defects occur in 3-5% of births in the United States.29-31 In the TREAT registry, women treated with infliximab had a congenital anomaly rate of 1.2% compared to women without biologic exposure who had a rate of 3.7%.23 In women exposed to vedolizumab the rate of congenital anomalies appears to be slightly higher at 6%. However, none of the congenital anomalies were felt to be related to vedolizumab’s mechanism of action. Hip dysplasia, pulmonary valve stenosis, and Hirschsprung’s disease occurred in three infants, one infant was born with congenital hypothyroidism, and there was a single case of agenesis of the corpus callosum and left frontal polymicrogyria in a healthy volunteer who received a single dose of vedolizumab 79 days prior to conception. Investigators felt that all of the congenital anomalies were unrelated to vedolizumab use. Of the elective terminations, one patient underwent an induced abortion due to a neural tube defect. This patient was also taking sulfasalazine which inhibits folate synthesis.32 Folate is known to have protective effects against the development of neural tube defects and supplementation is recommended at conception and during pregnancy.33-34 In analyzing the congenital anomalies as a whole, no two babies developed the same anomaly, and there is no evidence of an association with vedolizumab’s mechanism of action, making any causal effect unlikely.
Infant outcomes
IgG antibody transfer is known to increase from week 16 of gestation, with the majority of transfer occurring during the third trimester.35-36 Studies have demonstrated placental transfer of anti-TNF biologic agents with the exception of certolizumab.37-38 In PIANO registry data, vedolizumab was present in the infant serum at birth, however was less than half that of the mother39. This is in comparison to infliximab, where levels in the infant were found to be double that of the mother.40 Previously physicians have recommended the discontinuation of biologic medications in the third trimester due to fear of an increased risk of infection and other adverse outcomes in the exposed infant. However, exposure to anti-TNF therapy in the third trimester of pregnancy in the PIANO registry was not associated with an increased risk of infection in the infant.41 In addition, recent studies have shown that there is no increased risk of pre-term birth or low birth weight among women receiving anti-TNF therapy during the third trimester of pregnancy.42 An increased risk of pre-term birth or low birth rate was however associated with disease activity. Infants exposed in utero to immunomodulator and biologic therapy did not exhibit developmental delay compared to unexposed infants, with development scores in some categories actually higher in the exposed infants.43-44
In anti-TNF medications other than certolizumab, the rotavirus vaccination is avoided.7,45 Vedolizumab however, is not an immunosuppressant and does not pose the same risk for live virus vaccinations. In fact, patients themselves are able to receive live virus vaccines.46-47 Thus, infants exposed to vedolizumab should be able to receive the rotavirus vaccination. In one study of vedolizumab exposed mothers, 9 infants were inadvertently administered the rotavirus vaccination and none experienced any adverse outcomes.16
Of the studies that reported data on the outcomes of infants born to vedolizumab exposed mothers with IBD, there was only 1 hospitalization reported during the first year of life for fever of unknown origin.16 No other known hospitalizations occurred for the other 99 infants despite in utero vedolizumab exposure.
Breastfeeding
Breastfeeding is highly recommended by the American Academy of Pediatrics for at least the first 6 months of a babies’ life.48 In women with inflammatory bowel disease, breastfeeding has not been shown to have any adverse effects on disease activity.49 Several studies have assessed the transfer of anti-TNF agents, the anti-integrin natalizumab, and interleukin 12-23 inhibitor ustekinumab to breast milk. Although all were found to be present at low levels, exposed infants had similar rates of growth, milestone achievement, and infection risk compared with non-breastfed or unexposed infants.50 In two small studies of ten lactating women with inflammatory bowel disease, vedolizumab has been found to be present in breast milk in small amounts.51-52 Drug levels have not exceeded 480 ng/ml or 1/100th of the comparable serum levels. Fully breastfed infants primarily consume secretory IgA with a low content of IgG. Although vedolizumab may be transferred to breast milk, it is not used orally and is unlikely to be bioavailable in the newborn’s stomach. The minute quantity of vedolizumab is suspected to undergo proteolysis in the newborn’s digestive tract. All infants exposed to vedolizumab in breast milk reached normal developmental milestones at 3.5 to 10 months and there was no increase in general or intestinal infections.51-52
CONCLUSIONS
Current evidence supports the safety and benefits of continuing vedolizumab during pregnancy and breastfeeding. The rates of live birth and miscarriage are similar to those seen in the general population. The pre-term birth rate appears to be higher than that of the general population, but similar to previous studies of pregnancy outcomes in women with IBD. The congenital anomaly rate is slightly higher than that of the general population, however reassuringly, none of the anomalies have been felt to be related to vedolizumab’s mechanism of action. Although vedolizumab has been found to be present in infant serum at birth and in breastmilk, there has not been any increase in infection, and infants reached normal developmental milestones. In addition, newborns administered the live rotavirus vaccination did not have any adverse effects. Although the current data is reassuring, larger studies are still needed to confirm these findings.
NUTRITION ISSUES IN GASTROENTEROLOGY, SERIES #190
High Output Ileostomies: The Stakes are Higher than the Output
Meagan Bridges
Roseann Nasser
Carol Rees ParrishMeagan Bridges, RD Clinical Dietitian and Nutrition Support Specialist, University of Virginia Health System Charlottesville, VA. Roseann Nasser, MSc RD CNSC FDC Research Dietitian – Nutrition and Food Services Pasqua Hospital – Regina, Saskatchewan Health Authority. Carol Rees Parrish, MS, RDN Nutrition Support Specialist University of Virginia Health System Digestive Health Center Charlottesville, VA.
Recent years have seen a dramatic increase in readmission rates among patients with ileostomies who present with dehydration and/or kidney injury. High readmission rates are often the result of a failure to anticipate what will happen after discharge. Preventing readmission and preserving kidney function in these patients starts with reliable and accurate data collection – including not just stool output, but urine as well – and continues with detailed follow-ups to optimize medications, fluid, and food intake. Supporting patients through the entire process also requires educating them and equipping them with tools to gather and track their output. As clinicians, it is incumbent upon us to develop and execute a practical plan for adequate hydration and output management to not only prevent kidney injury, but also improve the quality of life for these patients.
CASE STUDY
A 46-year-old male with history of ulcerative colitis (diagnosed at age 26), status post total proctocolectomy with J-pouch (1998), proximal diversion with loop ileostomy with 270cm small bowel remaining (2005), presented in 2018 following 5 days of emesis and high output from his ileostomy, ultimately found to be secondary to a narrowing in his ileum causing outflow diarrhea.
His medications at the time of consult included a PPI BID, 5 mg Lomotil QID, 8 mg Imodium QID, Metamucil TID, cholestyramine BID, oxycodone PRN, and 50 mcg sandostatin q8h.
He is 6’ 2” and has maintained a weight of 250-255 lb for years. His BUN and creatinine were 28 and 1.3 respectively, with a reported 24 hour urine output of 1 liter during the winter months, but stated he sometimes goes a day or so without urinating in the summer. He also reports over 300 kidney stones – the first one occurring within 3 months of his loop ileostomy – with over 20 lithotripsies, all of which were managed at an outside facility hence, the surgeon who performed the loop ileostomy was unaware of any of this. After years of failing to meet his hydration needs and repeated bouts of nephrolithiasis, he finally lost his left kidney. During this admission, it was determined that he needed 3 L of IV fluids nightly to prevent dehydration and to protect his remaining kidney.
INTRODUCTION
Cases like the one above are not uncommon among patients with ileostomies. As Table 1 shows, recent years have seen a growing focus on readmission rates for dehydration and/or acute kidney injury (AKI) among this population (possibly as a result of stipulations in the Affordable Care Act aimed to decrease hospital readmissions in general).1 New ileostomy patients are often sent home well hydrated from IV fluids while admitted and with minimal output owing to decreased post-op appetite and intake, but this often does not reflect what will happen after discharge when patients are left to hydrate themselves and their appetite and oral intake picks up. In one study, it was shown that patients readmitted for AKI presented with a 3-fold increase in ileostomy output between post-op discharge and readmission (an average of 13 days later).2 Another study found that patients had significantly decreased GFR at ileostomy closure compared to pre-op ileostomy creation for any cause.3 Finally, Li et al. showed that 25% of patients with ileostomies develop CKD within 2 years, likely due to recurrent, sub-clinical dehydration.4
As clinicians, we are tasked with intervening not only to prevent kidney injury, but also to ease the other clinical and psychological burdens as well as quality of life challenges that so many patients with high-output ileostomies face (Table 2).
High Output Defined
As Table 3 shows, there can be many causes of high output, which in turn may lead to dehydration and kidney injury. A normal, mature ileostomy should only make about 1200mL of output each day (Table 4). Jejunostomies can initially put out up to 6 L, but this too will decrease with the help of medication. On the other hand, colostomies usually only put out 200-600mL/day. In the literature, “high output” is loosely defined as > 1500mL/day.
Acute Kidney Injury and Dehydration
As of 2011, expanded guidelines have been proposed based on serum creatinine levels and urine volume, widening the scope of what it means to have an AKI (Table 5). Dehydration, however, is a bit more nebulous. While there is no single way to define it, one of the best indicators is whether a patient is able to make enough urine (>1200mL/ day). Other indicators are listed in Table 6. Note that dark urine can sometimes be a side effect of a particular medication, rather than a sign of dehydration. Make sure to ask patients if they have ever been admitted for dehydration (whether at your own or another outside facility) and/or been to the emergency department and received IV fluids or experienced a kidney stone.
Treating and Preventing Dehydration: What to Do When an Ileostomy Patient is Readmitted
Treatment for dehydration will look different in ileostomy patients vs. those without ileostomies. In addition to fluid resuscitation with IV fluids, high output ileostomy patients are often told to drink more by mouth. Drinking more, however, does not mean absorbing more fluid and in fact, in some, will drive ileostomy losses further, resulting in even worse dehydration or volume depletion. In patients suffering from ongoing malnutrition, sweetened liquid nutrition supplements (such as Ensure/Boost, etc.) are often recommended, but these too are known to drive stool losses in those with high output. Some patients may notice that if they drink less fluid, their bothersome ileostomy output decreases, but then so does their urine output, often to a volume well below a liter per day. Unfortunately, while many patients are taught to record their stool or ileostomy volume, most are not educated to measure urine also, and this is the most important guide to hydration in these patients. Stool or ileostomy output may look great, but it may come at the expense of an adequate urine output, which may ultimately result in renal demise and chronic kidney insult.
Data Collection
Importance of Ins and Outs (I&O)
For dehydrated, high output ileostomy patients, the first step is to ascertain the patient’s true GI anatomy (if not known). If the operative report is unclear, consider ordering an abdominal CT to determine a patient’s anatomy and/or the presence of any strictures. If this is not an option, a small bowel follow-through can help determine gross anatomy and transit time through the GI tract.
For an accurate 24-hr I&O while an ileostomy patient is admitted, an order for “Strict or Measured I&O” vs. just “I&O” will ensure greater accuracy–i.e., not just if/when the patient stooled or emptied their ileostomy bag, but the actual volume of each occurrence. In many cases, it is worth having a discussion with the nursing staff to clarify the difference between I&O and Strict I&O. It is also very important that both floor and wound and ostomy nurses document if a patient’s ostomy is leaking, or bursting, so all know that the ostomy volume recorded in the medical record is less than what the losses really are. In general, goal urine output should be around 1200mL (or in the case of kidney stone formers, at least 1500mL) each day. Ideally, a goal stool output should be < 1500mL/ day, not just to reduce the risk for dehydration, AKI or kidney stones, but also to improve the patient’s overall quality of life. Providing patients with the tools to measure both urine and ostomy output is essential (see Figures 1-4).
Sodium
Patients with high ostomy output are at risk for sodium depletion as jejunal and ileal effluent contain 80-140mEq sodium per liter respectively. It will be important to provide enough sodium in the patients IV fluids to reflect this and adjust as the output is brought down under control. One way to determine if your patient is sodium replete is to obtain a 24 hour or random urine Na level; < 10mmol/L suggests Na depletion.5,6
Osmotic vs Secretory Diarrhea
Some patients who present with high output will require differentiating between osmotic and secretory diarrhea. These patients will need to be NPO for 24 hours with IV fluids and possibly parenteral nutrition (PN), if also malnourished. If ileostomy output significantly drops during this time, then it is osmotic in nature and can at least be partially managed by reducing food and/or fluid intake (and replacing with IV fluids as needed). The added benefit of this approach is that your patients will be able to see for themselves how eating and drinking directly drive output. If, on the other hand, ileostomy output remains over 500-800mL/24 hours, then it is considered a secretory diarrhea and will require a different medication and treatment approach.
Determining a Malabsorptive Component If you suspect malabsorption, collect a 48-72 hour fecal fat to determine the degree. A patient with severe malabsorption may require PN, whereas a patient with mild to moderate malabsorption may see enough improvement with diet/beverage changes, along with antidiarrheal and antisecretory medications. For younger patients, a 48-hour sample is usually sufficient, but Medicare beneficiaries will need to complete a 72-hour collection. Whichever test you use, ensure that your patients are ingesting/infusing 100 g fat per day either orally or enterally. A patient cannot malabsorb fat if they do not ingest it.
Food and Fluid Considerations
There is limited data on specialized diets for ileostomy patients other than those with known short bowel syndrome. Our clinical experience, however, suggests that these patients may benefit from a “relative” short bowel diet, at least until their output is well under control. In general, this diet is high in complex carbohydrates and low in sugar alcohols (contained in many liquid medications7), sugar, and sugary beverages (Table 7).8-11 Those with an end jejunostomy or ileostomy will need additional salt. Once a patient’s output is under control, it is important to begin liberalizing the diet as tolerated to avoid unnecessary restrictions and potential nutrient deficiencies.
Overall fluid intake is patient-specific. In general, hypertonic fluids, which pull water into the small bowel and thereby increase stool volume, should be avoided altogether.12 This includes fruit juice/drinks, regular sodas, sweet tea, syrup, ice cream, sherbet, sweetened gelatin, and liquid nutrition supplements such as Ensure, Boost or store brand equivalents. Small amounts of hypotonic fluids, such as water, tea, coffee, alcohol, and diet sodas, are allowed. However, bear in mind that hypotonic fluids will pull sodium into the small bowel; sodium in turn will pull water with it, thereby increasing stool volume as well. Initially, a drastic fluid restriction (e.g. <120mL with meals plus sips of water with meds for 24 hours) can be a powerful demonstrator to the patient regarding just how much oral fluids can drive output. Remember that all patients will still need to maintain a urine output of at least 1200mL/day; hence, some patients will need the addition of IV fluids while undergoing this trial.
Oral rehydration solutions (ORS) will not reduce stool output, but can be helpful in enhancing absorption of fluid in select patients. Consider trialing ORS with a small amount at first (e.g. 500mL sipped throughout the day). Some patients may prefer ORS in the form of ice cubes or popsicles. Other patients may benefit from a nocturnal infusion via gastric feeding tube as an alternative to IV fluids. A nasogastric trial is recommended first before placing more permanent access to ensure success (and not keeping the patient up all night with yet even more output). In addition to several ready-made commercial products available, patients can make their own ORS at home. See “A Patient’s Guide to Managing Short Bowel Syndrome” (available at no cost) for recipes.
Fiber Bulking Agents
Fiber bulking agents may thicken ostomy effluent from a jejunostomy or ileostomy, but they may hinder absorption of nutrients from food in the small bowel. In stable, well-nourished patients who have a colon, fiber bulking agents can be tried if desired by patient to improve the viscosity of stool, which in turn may improve quality of life (although there is a paucity of data to support benefit in this population). However, in the setting of malnutrition or poor appetite and PO intake, avoid filling your patients up on fiber supplements at the expense of other vital nutrients. In addition, fiber bulking agents may exacerbate electrolyte depletion by binding up minerals preventing absorption. Finally, while fiber bulking agents may thicken stool, they do not hydrate the patient as the water is now bound up in the fiber that is excreted in stool.
Medication Considerations
A number of medications can be used to slow down GI transit and reduce ileostomy output. Tables 8 and 9 list specific antidiarrheal and antisecretory agents that are commonly used to slow output. When maximum doses of loperamide (Imodium) (2-3, QID) and diphenoxylate/ atropine (Lomotil) (2, QID) are taken and ostomy output remains >1500mL/day, it is time to consider stronger gut slowing medications like opioids. In addition to the analgesic effects of opioids they:
- Delay gastric emptying
- Disturb the migrating motor complex
- Slow intestinal transit
- Increase anal sphincter pressure
- Inhibit water and electrolyte secretion
- Increase fluid absorption
thereby, allowing more time for fluid absorption to take place with a reduction in stool output.13 Likewise, Histamine-2 receptor blockers will not be as effective in reducing gastric secretions as proton pump inhibitors (PPI). In those patients deemed to be net-secretors, if oral PPI is not effective (possibly due to inadequate surface area for absorption), IV PPI, maximum dose, BID should be tried. Finally, octreotide/sandostatin can be very effective in those who have failed all other interventions. A dose of up to 500mcg q 8 hours may be needed in some.
Bile Acid Binders
Bile acid binders (cholestyramine, etc.) are often ordered in an effort to reduce high output. However, they are not appropriate for patients who have an end jejunostomy or ileostomy. The whole purpose of a bile acid binder is to protect the colon from the caustic effects of bile acids that pass through the ileum (normally, 95% of bile acids are reabsorbed in the last 100cm of ileum through the very efficient process of enterohepatic circulation). Unabsorbed bile salts that escape to the colon reduce transit time, decrease fluid resorption, and increase fluid secretion into the colon. As a result of fat malabsorption and calcium binding, they can also potentially lead to increased absorption of unbound oxalate.14 If one does not have a colon, the only thing bile acid binders will do is aggravate fat malabsorption and bind important minerals, nothing more. Bile acid binders are best reserved for patients with terminal ileal resections of <100cm or those with a diseased ileum, and, a colon segment in continuity. In addition to worsening fat malabsorption, fat-soluble vitamin status will need to be monitored more closely; screening for signs and symptoms of essential fatty acid deficiency annually may be wise as well.
The Total Pill Count
It is essential to review a patient’s medication list thoroughly and reduce the oral pill burden wherever possible. This includes prescription and over-the-counter medications, as well as vitamin/mineral supplements. Remember that the more pills a patient has to swallow, the more fluid he/she will be drinking, which can further increase output. For example, Lomotil and Imodium often require large, frequent doses (16-20 tablets/day) and can still leave a patient with a daily stool output of over 1500mL. Codeine, on the other hand, is much more efficient at slowing the gut with fewer pills (often around 4 tablets/day). Tincture of opium is also effective, but is much more expensive, often not covered by insurance, and is not readily available at pharmacies. It also requires good eyesight to measure the dropper dose, and has a particularly unpleasant taste.
The Curse of “PRN” Orders
In hospitalized patients, “PRN” medications are often not given. Yet, in a patient with high output, to be effective, it is not only imperative to schedule these medications, but to ensure they are taken before meals/snacks to avoid the “wash out effect.” Some will achieve better efficacy if crushed. Medications such as sustained, controlled, and delayed-released, as well as elixirs/suspensions with sugar alcohols should be avoided. Additional pharmacological considerations are listed in Table 10.
Finally, it is worth mentioning glucagon-like peptide 2 (GLP-2), an intestinotrophic, endogenous peptide released from the distal ileum and proximal colon that enhances gut adaptation in response to enteral nutrients. It inhibits gastric acid secretion and may slow emptying; stimulates intestinal blood flow; increases intestinal barrier function; and enhances nutrient and fluid absorption. In recent years, the GLP-2 analog, Teduglutide (Gattex/Revestive), has demonstrated effectiveness in reducing output and IV fluid / PN requirements in those patients with a high output from short bowel syndrome, provided they meet criteria (Table 11).
When an Ileostomy Patient Goes Home: Tools for Success
Before a patient is discharged home or to a facility, if possible, stop all IV fluids and monitor urine output for 2 days to ensure the patient can make adequate urine volume (1200mL). Also make sure patients have been educated on what is “normal” ileostomy output, what is “high output,” what their goal urine output is before they go home, and who to call if questions or problems arise. In addition, provide patients with the proper tools to successfully measure and track their output and manage all their medications. Stool hats, other ileostomy measuring devices, and urinals (see Figures 1-4), are essential for accurate stool and urine output measurements. A chart for tracking stool/ileostomy output, urine output, and date/time is also essential for monitoring trends and optimizing a patient’s regimen. For example, if a patient always records a large stool mid-afternoon, he/she may need to increase gut-slowing medications beforehand. Provide specific volumes for daily fluid/ORS recommendations; if the clinician is not sure a patient understands the volume intended, provide bedside pitchers with graduated markings or other containers to demonstrate the amount needed. Recommend a pill crusher for those medications that can/would benefit from crushing, and send your patients home with the small 30mL medication cups (see Figure 5) to keep the medication at the ready on their bedside table when gut-slowing medications need to be taken in the middle of the night (thus minimizing competition with food, fluid, and other medications).
Once a patient has left the hospital, closer and earlier follow-up (1-2 weeks post-discharge), along with routine labs and ongoing nutrition counseling will also help prevent readmissions. See summary suggestions in Table 12. For articles and handouts that can help clinicians and patients alike, visit the UVAHS GI Nutrition website and click on “Nutrition Articles” and “Patient Education.”
CONCLUSION
Hydration is essential to preventing kidney injury in patients with ileostomies who already have enough challenges to face as is. Most readmissions for dehydration and acute kidney injury can be avoided with proper planning and anticipatory guidance, along with early and thorough follow-up. Paying closer attention to electrolytes, both stool and urine measurements, along with medication and diet management, can dramatically improve our ileostomy patients’ quality of life and reduce readmissions and complications. Also see Table 13 for the new University of Virginia Health System diet/hydration handout developed to help prevent readmission for AKI after new ileostomy creation.
FROM THE PEDIATRIC LITERATURE
Supplements to Prevent Enteric Dysfunction in Children in a Developing Country
Environmental enteric dysfunction (EED) is a common problem in developing countries and is caused by inflammation leading to reduced intestinal absorption and diminished barrier function causing diarrhea and poor growth. Preventative treatments for large populations can be problematic, and this study evaluated the effectiveness of zinc as well as multi-vitamin supplementation to prevent complications of EED.
This double-blind, placebo-controlled study involved 6-week old infants in Tanzania born to mothers without HIV infection. Subjects were divided into 4 groups: infants receiving zinc supplementation, infants receiving multivitamin supplementation (consisting of vitamin C, vitamin E, vitamin B6, vitamin B12, thiamine, riboflavin, niacin, and folate), infants receiving both zinc and multivitamin supplementation, and infants receiving placebo. Compliance was monitored by medication count, and blood work was obtained at 6 weeks, 6 months, and 12 months of age to monitor serum anti-flagellin antibodies (IgA / IgG) and anti-lipopolysaccharide (anti-LPS) antibodies (IgA / IgG) to assess for EED; C-reactive protein (CRP) and alpha-1-acid glycoprotein (AGP) to assess for inflammation; and insulin-like growth factor-1 (IGF-1) and insulin-like growth factor binding protein-3 (IGFBP-3) levels to assess for growth.
In total, 2400 infants were enrolled in the study for which 590 infants with a history of EED had completed the study at 6 months and 162 infants with a history EED completed the study at 12 months. No statistical difference was seen in maternal, child, and socioeconomic factors between the 4 study groups except for the category of maternal mid-upper arm circumference. There was a significant increase in EED biomarker levels as children became older. However, no changes in biomarker concentration for EED, inflammation, and growth were noted between groups except for IGGBP-3 levels which were significantly lower at 6 months of age in children who received zinc supplementation and anti-LPS IgG levels which were significantly higher at 6 months of age in children who received multivitamin supplementation.
This study suggests that zinc and multivitamin supplementation does not improve EED in children, and such supplementation may not be helpful in reducing downstream effects of diarrhea and poor growth for children living in developing countries. Further studies are needed to validate these findings so that appropriate use of nutritional resources can serve such populations.
Lauer J, McDonald C, Kisenge R, Aboud S, Fawzi W, Liu E, Tran H, Gerwirtz A, Manji K, Duggan C. Markers of systemic inflammation and environmental enteric dysfunction are not reduced by zinc or multivitamins in Tanzanian infants: a randomized, placebo-controlled trial. Journal of Pediatrics 2019; 210: 34-40.
A SPECIAL ARTICLE
Chronic Hepatitis C Assessment and Treatment: Algorithms for Primary Care Providers
Susan FergusonSusan Ferguson, PA-C Physician Assistant Santa Rosa Gastroenterology, Santa Rosa Beach, FL.
In the past, chronic hepatitis C (CHC) was a challenge to manage as treatment protocols were complicated and medications had severe side effects. Primary care providers are often the first medical professionals who screen for CHC, but patients are frequently referred to specialists after hepatitis C is diagnosed. Moreover, health care providers in safety-net clinics should be screening their patients as low-income individuals are disproportionately affected by CHC. The new generation of CHC medications makes treatment options more feasible and cost-effective. Assessment and treatment protocols for CHC have been developed to serve the needs of primary care providers in private practice as well as safety-net clinics.
INTRODUCTION
Chronic hepatitis C (CHC) affects 3.4 to 6 million Americans.1 The prevalence of CHC is four times higher in populations that are below the federal poverty level in comparison to those who are two times the federal poverty level.2 Therefore, clinics that serve low-income patients should have a high priority for screening and treatment of CHC.2 Moreover, the cure rates for CHC increased when medical providers within safety net clinics had a comprehensive care plan when treating patients for CHC.1
The new CHC therapies have lower side effect profiles and can be prescribed to a larger selection of patients.3 The advent of pangenotypic medications for CHC enable primary care providers, whether in safety-net clinics or private practice, to effectively treat a larger cohort of patients.
Moreover, assessment and treatment algorithms can aid primary care providers in determining when it is appropriate to refer patients to specialists for CHC treatment.
Epidemiology
In 2016, there were 2,967 reported cases of acute hepatitis C in the United States.4 The overall incidence rate for new cases was 1.0 per 100,000, which is an increase from 0.8 cases per 100,000 in 2015.4 Furthermore, it is estimated that the actual number of cases are most likely 13.9 times greater than reported.4
There are more than 71 million people worldwide who have CHC.5 Approximately 70% of individuals with CHC were born between the years of 1945 to 1965.6 In the United States, males have approximately double the prevalence rate of CHC compared to women.6
Etiology
Hepatitis C virus (HCV) is the most common blood-borne infection in the United States.6 Common exposures to the virus include injection drug use, needle sticks, transfused infected blood product before 1992, and maternal transmission during birth.4 The hepatitis C virus can infrequently be transmitted through the use of personal hygiene products belonging to infected individuals, unregulated tattoos, and sexual intercourse with infected individuals.4 There is a common misconception that CHC is prevalent in the “baby boomer” generation as a result of high-risk behaviors, such as tattooing, sexual practices, and injection drug use.7 To the contrary, members of the high-risk group born between 1945 and 1965 were approximately 5 years old when the peak of Genotype 1A HCV occurred in 1950.7 The likely peak in 1950 occurred as a result of medical procedures performed during and after World War II, and the transition to disposable plastic syringes in the 1950s through the 1960s correlated with a decline in HCV incidence rates.7 However, the role of recreational injection drug use from 1920 to the late 1960s was contributory.7
Pathophysiology
Hepatitis C is transmitted parentally through blood and body fluids.8 Acute presentations of the illness may result in jaundice, which occurs in approximately 20% of individuals.8 The average incubation period for HCV is between six and seven weeks. The incubation time can be difficult to assess as infected individuals are often asymptomatic, and there is not a precise serological marker of early infection.9 Spontaneous resolution of the virus occurs in approximately 15% to 45% of infected individuals, and clearance usually happens within six months of contracting the virus.8 Moreover, clearance of HCV is bifurcated by gender with females having a 40% spontaneous clearance rate compared with 19% of males.6 Spontaneous clearance is influenced by genetic inheritance of the IL28b and DQB1 0301 allele on the class II2 major histocompatibility complex.10
Interestingly, patients who present with clinical symptoms of acute hepatitis C demonstrate higher rates of spontaneous clearance of the virus.10 For individuals who do not clear the virus, the annual rates of fibrosis progression are between 0.1 and 0.2 stages per year.10 Approximately 20% to 25% of CHC cases progress to cirrhosis.8 Out of every 100 infected individuals with CHC, 10 to 20 will develop cirrhosis, and there is a 1% to 5% annual risk of developing hepatocellular carcinoma.4 The progression from cirrhosis to hepatocellular carcinoma is due to several factors. Specific viral proteins act upon cell signaling pathways and inhibit tumor suppressor genes, signaling pathways that up-regulate cell growth and division are activated, and the loss of the tumor suppressor genes p53 and retinoblastoma cause increased carcinogenesis.11
Signs and Symptoms
Most individuals with CHC are asymptomatic, or they may have non-specific symptoms, such as lethargy or depression.4 However, CHC is linked to extrahepatic disorders, such as arthritic-like pain, kidney, cardiovascular, neuropsychiatric, and gastrointestinal conditions.12 Many individuals with CHC will eventually present with liver disease. Chronic hepatitis C was the known etiology for 17% of the reported cases of hepatocellular carcinoma in the United States from the years 2000 to 2010.13 Patients with advanced liver disease may manifest symptoms, such as ascites, jaundice, variceal bleeding, altered mental status, and pruritus.14
Diagnosis
Serology
The presence of HCV antibody (anti-HCV) can indicate active HCV infection, either acute or chronic; past infection that resolved with spontaneous clearance or treatment; or a false positive.15 Therefore, the World Health Organization (WHO) and the Centers for Disease Control and Prevention (CDC) recommend that serological testing for CHC should include both anti-HCV and nucleic acid testing for the presence of HCV RNA.16,17 The American Gastroenterology Association (AGA) recommends testing for anti-HCV, HCV RNA, HCV genotype, complete metabolic panel (CMP), INR, hepatitis B studies, and HIV antibody.14
Assessment of liver disease
The 2016 WHO guidelines state that assessment for fibrosis and cirrhosis in CHC infected individuals is necessary to properly stratify individuals into appropriate treatment protocols.17 The American Association for the Study of Liver Diseases (AASLD) and the Infectious Diseases Society of America (IDSA) collectively composed guidelines for evaluation of advanced liver disease, including fibrosis and cirrhosis. Liver disease can be assessed with liver biopsy, imaging, and/or non-invasive markers.18 The AGA recommends right upper quadrant ultrasound for initial imaging before treatment.14 Fibrosis staging can also be assessed with newer non-invasive technologies, such as transient elastography and magnetic resonance elastography.14 However, aspartate aminotransferase to platelet ratio index (APRI) or fibrosis-4 (FIB-4) testing can be utilized when medical resources are scarce.14,17 Both the APRI and FIB-4 tests have low and high cutoff values that can be utilized to assess the severity of fibrosis, and the APRI scoring system can also be utilized to estimate a patient’s probability of having cirrhosis.17 The cutoff values aid providers in the assessment of liver disease secondary to HCV and can be utilized to stratify patients for treatment.17 An APRI score below the low cut-off value of 0.5 indicates that a patient has only an 18% probability of having advanced fibrosis, and a FIB-4 score lower than 1.45 indicates an 11% probability of having advanced fibrosis.17 Moreover, the WHO guidelines stated that treatment could be deferred in patients who score below the low cut-off values on APRI and FIB-4 tests.17 However, patients with an APRI score above the high cutoff value of 2 have a 94% probability of having cirrhosis, and these patients would benefit greatly from treatment.17 Patients who have APRI or FIB-4 scores that fall between the low and high cut-off values can either be treated or monitored, depending on the availability of medications.17
Treatment Options for Chronic Hepatitis C
Overview
There are six separate genotypes for HCV (1- 6), and approximately 70% of all cases of CHC are Genotype 1.18 The prior treatment for CHC utilized interferon alfa-2a plus ribavirin, but these medications have severe side effects associated with them and achieve cure in only about 40% to 50% of patients.19 In 2009, the first direct-acting antivirals (DAAs), which directly target HCV replication, were approved for the treatment of CHC.20 By 2017, there were six different DAA combination therapies available for specific genotypes of CHC.19 The advent of pangenotypic medications for CHC almost obviates the need to genotype the virus; however, in certain CHC patients, viral genotyping is still necessary. Additionally, cirrhotic patients should be assessed for the severity of cirrhosis utilizing the Child-Pugh score.21 While glecaprevir/pibrentasvir is pangenotypic for non-cirrhotic, treatment-naive patients, genotyping is still necessary for treatment-experienced and/or cirrhotic patients as the duration of therapy has to be adjusted based on these parameters.22 Conversely, sofosbuvir/velpatasvir is a pangenotypic medication that can be utilized for non-cirrhotic or Child-Pugh A cirrhotic patients who are treatment-naive or experienced.23 Although there are other medicinal treatments for CHC, this review will be limited to the pangenotypic class of CHC medications as they enable providers to more easily treat patients at the primary care level.
NS5B inhibitors
The AASLD and IDSA collectively recommend sofosbuvir (SOF) for treatment of all genotypes of HCV.18 Sofosbuvir is administered orally, once a day.19 Sofosbuvir inhibits the HCV NS5B RNA-dependent RNA polymerase by incorporating into the HCV RNA, whereby it terminates the RNA chain.24 Sofosbuvir demonstrated an advantage over the first-generation DAA class of HCV medications because of its high genetic barrier to viral resistance.18 Furthermore, sustained viral response (SVR) improved when SOF was used in combination therapy.25 There is a low risk for medication interactions with SOF as it does not inhibit or induce the P450 cytochrome system; however, amiodarone is contraindicated with SOF because of the risk for fatal bradycardia.19 Additionally, given the fact that SOF is excreted through the renal system, it should be used with caution in renal patients and should not be given to patients with glomerular filtration rates (GFR) less than 30.19
NS5A inhibitors
There are numerous functions of the HCV viral NS5A protein, including roles in viral replication, assembly, and interactions with cellular functions.26 Velpatasvir (VEL), an NS5A inhibitor used in conjunction with SOF, is efficacious on all genotypes of HCV.19,20 Because less than 1% of VEL is renally excreted, there are no dosing adjustments in the context of renal dysfunction; however, caution should still be taken as VEL is given in conjunction with SOF.19 Concomitant administration of rifampin, efavirenz, or St. John’s Wort may decrease levels of SOF/VEL and should not be taken concurrently.19 Conversely, SOF/VEL may decrease therapeutic levels of digoxin so treatment levels should be monitored closely.19 Additionally, proton pump inhibitors (PPIs) should be discontinued during therapy with SOF/VEL as VEL concentration levels decrease with increases in gastric pH.23
NS3/4 protease inhibitor combination therapy
The AASLD and IDSA guidelines recommend the combination of glecaprevir (GLE) and pibrentasvir (PIB) for all genotypes of HCV.18 Glecaprevir is a second-generation NS3/4 protease inhibitor and PIB is a second-generation NS5A inhibitor.27 The GLE/PIB combination regimen has a high genetic barrier to resistance compared with first-generation medications27 and is excreted through the biliary system with minimal renal excretion.28 Moreover, GLE/PIB is designed for once-daily dosing, and treatment-naive patients without cirrhosis need only take the medication for 8 weeks.22 The GLE/PIB combination medication is not recommended for individuals with Child-Pugh B scores and is contraindicated for patients with Child-Pugh C scores.22 Additionally, GLE/PIB is contraindicated with concurrent use of rifampin or atazanavir.22
Challenges of Treating Chronic Hepatitis C Co-infection
Co-infection with hepatitis B (HBV) and/or human immunodeficiency virus (HIV) is associated with worse prognosis of CHC.18 Additionally, there is the possibility of a reactivation of HBV after initiating HCV therapy in patients who are co-infected with hepatitis B and C.18 Therefore, the WHO recommends that individuals who test positive for the hepatitis B surface antigen (HBsAg) be treated for HBV before they are treated for HCV.17
Concomitant alcohol or drug use
The WHO guidelines recommend performing alcohol and illicit drug intake assessments before initiating treatment.17 Moreover, U.S. Department of Veterans Affairs (VA) recommends that patients who have histories of substance or alcohol disorders be considered on a case-by-case basis for treatment.29 The AASLD and IDSA guidelines contend that there is a lack of evidence supporting the practice of withholding treatment for HCV infected individuals who intake alcohol or illicit drugs.18 Moreover, the VA guidelines discourage providers from disqualifying patients for treatment based solely on the length of abstinence from alcohol or illicit drugs. Patients with active substance or alcohol disorders may be treated for CHC but coordination with substance treatment professionals is imperative for treatment success.29
Treatment experienced patients
Treatment-experienced patients need to be stratified by the type of prior failed therapy, the specific HCV genotype, and stage of liver disease.18 Once daily dosing of GLE/PIB can be utilized for treatment-experienced patients; however, the duration of therapy is between 8 and 16 weeks based on prior treatment failures, viral genotype, and liver disease state.22 In contrast to GLE/PIB therapy, SOF/VEL has a once daily dosing for 12 weeks, independent of genotype, prior therapy, or liver disease state.23
Patients with cirrhosis
The VA asserts that all CHC infected individuals are potential candidates for HCV therapy, including individuals with cirrhosis. Furthermore, patients with advanced liver disease are likely to benefit the most from therapy.29 However, HCV infected individuals with decompensated cirrhosis need to be evaluated by specialists in the management of complex liver disease before treatment is initiated.18
Treatment Monitoring and Curative Metrics
The WHO guidelines state that a complete blood count (CBC), renal function test, liver function test, and HCV RNA quantity be performed as a baseline before starting treatment and again at week 4 of treatment with the exception of HCV RNA quantity.17 However, the AGA guidelines assert that patients receiving treatment should be tested at week 4 with a CBC, CMP, and HCV RNA quantity.14 If HCV RNA is detectable at week 4, therapy should continue for 2 more weeks with a repeated HCV RNA quantity test at week 6.18 Therapy should be discontinued if there is a greater than 10-fold increase of HCV RNA from baseline at week 6; however, there are no set guidelines to discontinue therapy if there is less than a 10-fold increase of HCV RNA.18 Moreover, a 10-fold increase in ALT levels from baseline at any time during therapy should prompt discontinuation of treatment.18 For patients on either 12-week or 16-week regimens, HCV RNA quantity should be tested at week 12 and week 16 respectively.14 All patients should be tested for HCV RNA quantity at 12 weeks post-therapy.14,17,18
Sustained viral response at 12 weeks post-therapy (SVR12) is defined as an undetectable HCV RNA level in the blood and is considered the primary endpoint of treatment success.29 Pradat et al.5 stated that the use of direct acting antiviral agents (DAAs) achieved a 90% cure rate. Moreover, Evon et al.12 asserted that that DAA medications for CHC demonstrated a 95% cure rate. Kosloski et al.28 contended that the combination of GLE and PIB achieved a 99% cure rate in compensated cirrhotic patients treated for 12 weeks for all genotypes except genotype 3, for which a greater than 96% rate was achieved. Furthermore, Osawa et al.27 demonstrated a 93% cure rate of GLE and PIB combination therapy in treatment-experienced, non-cirrhotic patients who experienced prior virologic failure on DAA medication regimens. However, the cure rate decreased to 83% in treatment-experienced, cirrhotic patients.27 Treatment-naive patients with or without cirrhosis who were given SOF/VEL had cure rates of 95% to 99% depending on HCV genotype.19 Moreover, Feld et al.30 demonstrated a 99% cure rate with SOF/VEL in treatment-experienced patients.
Figure 1, “The Assessment Algorithm for Chronic Hepatitis C”, is designed in a decision-tree format for ease of use. The decision-tree contains common scenarios that primary care providers may encounter in assessing patients for possible chronic hepatitis C therapy. There may be circumstances in rural or resource scarce areas where specialty providers, including infectious disease specialists, are limited. In such cases, providers can refer patients to their local health department for further evaluation if need be. However, most areas do have gastroenterologists who can assist with both liver disease and complex treatment protocols.
Figure 2, “The Treatment Algorithm for Chronic Hepatitis C”, utilizes the currently available combination therapy of sofosbuvir/velpatisvir. Sofosbuvir/velpatisvir has a relative ease of use because it can be prescribed for any genotype of the hepatitis C virus. Additionally, the dosing durations are the same regardless of whether a patient is non-cirrhotic or has Child-Pugh A cirrhosis. Moreover, the medication has the same dosing duration whether or not the patient is treatment-naive or had unsuccessful treatments for CHC in the past. The side effects are minimal in comparison to prior CHC therapies and dosing is once a day.
SUMMARY
Patients with cirrhosis are in more urgent need of HCV therapy as their risk for conversion to decompensated cirrhosis or hepatocellular carcinoma is elevated, in comparison to non-cirrhotic individuals with CHC.29 However, the AASLD and IDSA guidelines recommend treatment for all individuals with CHC except those with short life expectancies who cannot be reasonably and positively affected by HCV therapy, liver transplant, or other directed therapy.18
Primary care providers are on the front lines for hepatitis C screening and need to have the proper tools to manage the disease process. Given the new medications available to treat CHC, treatment protocols have never been easier to administer. One of the main challenges of treating CHC patients at the primary care level is knowing when to refer patients for more specialized care. The assessment algorithm included in this article addresses the situations that may require specialty referrals. Additionally, the treatment algorithm utilizes SOF/VEL, which has a low side-effect profile; less onerous monitoring schedules; and does not require viral genotyping. Primary care providers can effectively utilize these algorithms to treat a much larger cohort of CHC patients both in private practices and safety-net clinics.
FROM THE PEDIATRIC LITERATURE
Detergent Pod Ingestions in Children
Concentrated detergent pods are a cause of caustic ingestions in young children due to their similar appearance to candy. The authors of this study attempted to describe both endoscopy as well as bronchoscopy findings in children who had injuries from detergent pods by reviewing concentrated detergent pod ingestions occurring at a single, tertiary children’s hospital over 7 years (2010-2016). This retrospective study included children between 0 to 18 years of age who were exposed to caustic agents, including concentrated detergent pods. Children were excluded if they had a pre-existing disease of the esophagus (such as gastroesophageal reflux disease) or if they had a concurrent foreign body ingestion occurring with a caustic ingestion. Patient demographics, including esophagogastroduodenoscopy (EGD) and direct laryngoscopy-bronchoscopy findings were reviewed.
In total, 83 caustic ingestions occurred during this time period, and 23 of these cases (28%) were due to detergent pod ingestion. Detergent pod ingestions mainly occurred in males (61%), and most patients had gastrointestinal symptoms after ingestion (91%). The most common laboratory abnormality was metabolic acidosis which occurred in 39% of patients. Although no gastrointestinal complications such as esophageal stricturing occurred, 13% of patients had respiratory failure requiring intubation / mechanical ventilation. EGD was performed in 91% of patients with detergent pod ingestion, and 30% of these patients had esophageal edema, erythema, or ulcerations. There was a significant association between positive oral-pharyngeal findings and esophageal damage. Direct laryngoscopy-bronchoscopy was performed in 26% of the patient cohort, and damage to the upper airway was noted in 67% of such patients, including epiglottitis and glottis edema although there was no association between respiratory symptoms and airway findings.
Although this is a small study, the results suggest that EGD may not be necessary in children who undergo accidental concentrated detergent pod ingestion. On the other hand, it appears that respiratory failure is a risk and should be considered when such children present in the emergency room, hospital, or clinic setting.
Singh A, Anderson M, Altaf M. Clinical and endoscopy findings in children with accidental exposure to concentrated detergent pods. Journal of Pediatric Gastroenterology and Nutrition 2019; 68: 824-828.
Fellow’s Corner Submission Guidelines
The Fellow’s Corner is open to Trainees and Residents only.
Guidelines:
- Send in a brief case report.
- No more than one double-spaced page.
- One or two illustrations, up to four questions and answers and a three-quarter to one-page discussion of the case.
- Case to include no more than two authors
Email your Case Report to Section Editor C.S. Pitchumoni, MD
Nutrition Issues In Gastroenterology, Series #156
Sterile Water and Enteral Feeding: Fear Over Logic
Todd W. RiceMany practitioners believe they must utilize sterile water for administration into enteral feeding tubes, yet the data supporting this practice are very limited. This manuscript will expose the flaws in the rationale behind the practice and outline why other forms of potable water are not only acceptable, but preferred as the type of water to administer to patients with enteral feeding tubes.
Many practitioners believe they must utilize sterile water for administration into enteral feeding tubes, due to a fear of exposing the patient to potentially pathogenic infectious organisms, especially in critically ill patients, immunocompromised patients, or those with post-pyloric feeding tubes. However, the data supporting this practice are very limited. Enteral feeding tubes are not sterile devices; they are not placed or maintained under sterile conditions. Furthermore, the gastrointestinal tract is designed to handle foreign material and infectious organisms. This does not change in patients receiving enteral feedings. The recommendation to utilize sterile water for administration into enteral feeding tubes is both unjustifiable and costly. This manuscript will expose the flaws in the rationale behind the practice and outline why other forms of potable water are not only acceptable, but preferred as the type of water to administer to patients with enteral feeding tubes.
Todd W. Rice, MD, MSc, Associate Professor of Medicine Division of Allergy, Pulmonary and Critical Care Medicine Vanderbilt University School of Medicine, Nashville, TN
INTRODUCTION
A number of practitioners and healthcare systems have insisted on the use of sterile water in enteral feeding tubes. This use of sterile water is sometimes limited to administration of free water, and other times also includes any mixing of powder formula or medications. Many reserve this practice for enterally fed critically ill patients, immunocompromised patients, or those receiving post-pyloric enteral feedings that bypass the stomach. This is an interesting practice pattern, which has its origins in a few anecdotal reports of infections from contaminated water in patients who happen to be receiving enteral feeding. In fact, the last American Society of Parenteral and Enteral Nutrition (ASPEN) Safe Practices for Enteral Nutrition Therapy guidelines published in 2009 recommend the use of sterile water in certain populations of patients receiving enteral feeding.1 However, the use of sterile water in these situations is neither logical nor practical, but instead based on an irrational fear of harming patients.
Enteral Feeding
First, let’s examine the universal use of sterile water in enteral feeding tubes. The gastrointestinal system is not a sterile environment. From the oropharynx through the rectum, and every location in between, is saturated with commensal bacteria forming the normal flora and individual microbiome of each patient.2 When these bacteria are decreased, opportunistic organisms, such as Clostridium difficile are more easily able to multiply and cause infections. In addition, placement of enteral feeding tubes is not a sterile procedure – while gloves are donned for the placement, the gloves are not sterile and are intended to minimize soiling of the person placing the tube over preventing contamination of the tube. The patient is not taken to an operating room, the nares (or oropharynx) are not sterilized prior to placement, and a whole sterile field is not used during the placement procedure. Instead, these tubes are often placed at the bedside under non-sterile conditions by the bedside nurse. Furthermore, once placed, the enteral feeding tube is not maintained with sterility. It is not covered with a sterile dressing, nor is the hub sterilized with chlorhexidine or alcohol prior to access.
Medication Administration
Furthermore, the enteral feeding tube is often used for medication administration – medications which are delivered from the pharmacy (or kept in a medication dispensing unit on the floor), are not sterile. They are touched by numerous human hands, often without gloves, prior to administration to the patient. In fact, in order to be administered through an enteric feeding tube, medication in pill form often has to be crushed – which occurs using a non-sterile pestle and mortar or pill crusher kept on each floor or unit. The mortar and pestle, or pill crusher, are washed after each use, but not sterilized. Liquid medications are often dispensed in smaller quantity aliquots from large quantity storage containers in a non-sterile fashion. While the bottles used to dispense the liquid medications are clean, they are not handled under strictly sterile conditions. While this delineation of all of the non-sterile interactions with the enteral feeding tube may startle some practitioners, it should not cause concern. The gastrointestinal tract is meant to handle non-sterile conditions.
In addition to the huge number of bacteria present as its normal flora, the GI tract is also designed to handle exposure to extraneous organisms. The GI tract secretes a number of molecules, which help to protect against infectious insults as part of its normal function. Digestive enzymes may help kill some bacteria, bile salts may bind some bacteria, and IgA antibodies provide a level of immunity against bacteria that are not part of the normal flora.3 This allows us to eat without having to sterilize our food. While we often wash fruits, vegetables and other non-packaged food products, we do not wash them with sterile water, nor worry that they must be sterile prior to consumption. Similarly, we do not limit our consumption of water to only sterile water. Imagine having to find (or carry with you) sterile water for every time that you wanted or needed a drink of water.
Critical Illness
Yes, but that is in normal, non-sick humans. Is the critically ill patient different? Of course, the critically ill patient is different than a healthy individual. Many critically ill patients are not eating on their own and are dependent on enteral feedings. Therefore, they are not ingesting fruits or vegetables or non-packaged products. While this is true, many of the same facts above are also true. Medications administered to critically ill patients through enteral feeding tubes are not sterile – they are not handled with sterility in the pharmacy or ICU, they are not crushed under sterile conditions, and they are not administered using sterile technique. As soon as the enteral feeding tube is removed from its package, it loses any sterility that it had. It is placed through the nose or oropharynx, which have their own microbiome and are not sterile. The placement is done with non-sterile gloves, without chlorhexidine or povidone-iodine prep. Also every time the enteral feeding tube is accessed, it is not done under sterile conditions. The feeding tubes are not thoroughly washed with chlorhexidine or alcohol prior to touching them, administrations are not done using sterile gloves, the connectors or insertion end of the feeding tube is not sterilized with chlorhexidine or alcohol wipes prior to administration of anything through the tube, and the feeding tube is not maintained in a sterile sleeve or dressing (like the sterile protective sleeve that covers pulmonary artery catheters or dressing covering intravenous catheters to maintain their sterility). Some practitioners administer probiotics through the enteral feeding tube4 in certain critically ill patient populations, purposefully introducing bacteria into the enteral tube in an effort to replenish normal flora in the gastrointestinal tract, in order to prevent the overgrowth of pathogenic bacteria such as Clostridium difficile.
Immunocompromised Patients
What about immunocompromised patients receiving enteral feedings? They are a bit more complicated as they, by definition, do not have normal immune function. The use of probiotics in immunocompromised patients is currently discouraged as there are reports of bacteremia from the specific bacteria in the probiotic.4 However, like all patients, these patients do not have a sterile gastrointestinal tract. When they eat, they do not eat sterile food – despite a lack of evidence to support the practice, their diets may be modified to avoid fresh fruits or vegetables. However, their diet is not limited to sterile food. Their enteral feeding tubes are not placed, nor maintained, in a sterile fashion. In addition, the medications that they receive are also not sterile. While caution should be taken to not introduce known contaminated materials, including contaminated water, into their enteral system, their gastrointestinal tract still has adequate defense mechanisms to handle bacteria.
Post-pyloric Feeding
Lastly, some have advocated for the use of sterile water for post-pyloric tubes, or when the distal end of the enteral feeding tube terminates somewhere beyond the stomach. While the acid from the stomach represents one of the first lines of defense against bacteria, it is not the only line of defense, and the bile salts and IgA protective mechanisms of the gut are present in the small intestine and not the stomach. In fact, these are more effective at countering potential infectious organisms than the acid of the stomach.5 Furthermore, most patients receiving enteral feedings are also receiving some sort of acid suppression (i.e. histamine blocker, proton pump inhibitor, etc), so even patients who have gastric tubes likely lack much of the natural protection afforded by the acidity of the stomach. If there is concern that post-pyloric feeding bypasses this protective mechanism, there should be equal concern for our gastrically fed patients receiving histamine receptor blockade or proton pump inhibitors.
The caution about using non-sterile water, and recommendation for sterile water use, appears to come from two misunderstandings. First, there is an irrational fear of harming the patient by either introducing an infection with contaminated water or precipitating bowel necrosis. However, infections documented from contaminated water are not from enteral administration. The vast majority are pulmonary infections such as legionella, pseudomonas, or mycobacteria6-10 and according to Smith et al.., these respiratory infections almost assuredly were obtained by inhalation of contaminated droplets from the air and not from hematogenous spread from an initial GI source. Washing hands (with subsequent aerosolization of the water source) is more likely the culprit than enteral administration of the water, where the gastrointestinal tract has numerous defense mechanisms in place to prevent contraction of infectious organisms. Secondly, bowel necrosis is a rare event in patients receiving enteral feedings. One case report associates distilled water administration into the jejunum with bowel necrosis and perforation in a burn injury patient.11 Due to hypernatremia, the patient was receiving 400 mL of distilled water flushes every 2 hours. Data from a study in one rat suggest that electrolyte-free water may permit digestion of the bowel wall and predispose to perforation, compared to infusion of salt water.11,12 Even if these limited data are true, administration of sterile water does not ameliorate this risk. Sterile water is still electrolyte-free, and in fact, is likely more electrolyte free due to its sterile processing than other forms of drinkable water. Furthermore, although distilled water was utilized in the case report, there is no evidence that the use of sterile water in that patient would have prevented the necrosis. The two are as likely unrelated as they are coincidentally related. In addition, this case report does not provide evidence that tap water flushes into the jejunum in reasonable volumes pose any danger to humans.
In addition, there is a misunderstanding of different types of water (Table 1). There are not merely two options, sterile vs. contaminated. These represent two ends of the spectrum, with numerous options in between. Sterile water is verified to be free of all infectious organisms, is produced for use in sterile medical procedures, and must meet USP regulations.13 Due to this, it carries a higher cost compared to other forms of water. Potable (drinking) water is not sterile. It can come from many sources, including tap water, spring water, and filtered water.13 Bottled water, which is defined as water sealed in containers without added ingredients, is regulated by the Food and Drug Administration and is also not sterile. It often comes from springs and is treated to remove most of the infectious organisms and heavy metals and other impurities, but it is not sterile. Other types of water in the spectrum between sterile and contaminated include filtered, purified, distilled, disinfected, and tap water. These water types are also not sterile and should not be used in place of sterile water in medical situations where sterility is needed, like an operating room or lavaging a sterile body cavity (like the abdomen, thorax, urinary bladder, etc). Filtered water has been filtered through a physical, chemical, or biological process to remove many of the impurities14 and is an acceptable grade for drinking. Purified water is similar to filtered. It has been processed to remove impurities and is also acceptable for drinking.14 Distilled water is the steam from boiled water, which is allowed to condense in a separate container, leaving many of the solid contaminants behind. However, it is not sterile; it can still have some low level of bacteria present. Distilled water is also potable, meaning it is an acceptable grade for drinking. Disinfected water has been processed with a disinfectant, often chlorine, fluorine, iodine, or ultraviolet light to kill bacteria. While this process greatly diminishes the number of live infectious organisms in the water, it does not ensure sterility. Tap water simply describes water that is obtained from the tap, or spigot. The quality of tap water varies greatly, depending on the source of the water and any processing that occurs prior to delivery at the tap. Potable water should be used for administration into enteral feeding tubes. Often, tap water is potable and can be used. Most tap water in the United States is acceptable for drinking, but contaminated taps do exist. If there is any concern, tap water should not be used for drinking, even by healthy people. Caution should be taken to avoid using known contaminated tap water or water that personnel on the unit would not feel comfortable drinking. In these situations, tap water should also not be used for enteral feeding tube administration, regardless of whether or not the patient is critically ill, immunocompromised or has a feeding tube whose distal end terminates in the small bowel. However, even these situations do not require the use of sterile water. Other forms of potable water, namely filtered, purified, distilled, or even bottled water bought at the store can be used instead, and often at considerably less cost. Any water that healthcare personnel or other people drink (filtered, purified) is more than adequate for administration through enteral feeding tubes.
Despite this, one may ask, even if the risk of non-sterile water is very low, why not be safe and use sterile water for the highest risk patients receiving enteral feeds? Sterile water is not without downside. It is expensive, often costing up to $4 per liter (not to mention personnel to deliver to unit, storage space required on the unit, as well as nursing time spent retrieving, etc.). While this may sound like a minor expense, in patients receiving one to two liters of free water via feeding tubes each day (not an uncommon amount for a patient to be receiving), this would amount to $4-8 or more per day, or almost $1500-$3000 per year per patient. An added healthcare cost without benefit that will have to come out of someone’s budget – either the hospital’s or the patient’s. Many patients receive more than two liters of water each day, especially with medication administration, formula reconstitution, and feeding tube flushes. In addition, the use of sterile water is not maintaining sterility of the feeding tube – other things administered via the feeding tube, including medications, supplemental protein, and unclogging agents, are not sterile.15 Furthermore, obtaining sterile water is more difficult than other forms of potable water. Given its medical grade, and requirement for meeting USP standards, it is only available from places that sell medical goods or medications. It is not routinely available at the grocery, nutrition, or convenience store. This inconvenience may be onerous for the patient or caretaker, and many times, the patient (or caretaker) simply stops using sterile water, despite the guilt, which may occur. Furthermore, the use of sterile water in this fashion also creates environmental concerns as unneeded trash from empty containers (i.e. plastic bottles) must be disposed of someplace and is likely to end up in our landfills.
CONCLUSION
Given the soaring health care costs in this country, clinicians should always weigh the cost against demonstrated benefits of practices prior to implementation. The routine use of sterile water in enteral feeding tubes increases cost, is more labor intensive and is harmful to the environment, without any added benefit to our patients. Regardless of the condition of the patient or location of the distal ports of the feeding tube, the mandated use of sterile water is illogical, unfounded, and expensive. Instead, we should recommend against using known contaminated water (including contaminated tap water) or water which is known or thought to be non-potable. Any water administered into an enteral feeding tube should be potable, just like any water drunk by patients able to ingest on their own. When safe potable tap water is not available, numerous cheaper, more practical, and more easily accessible forms of potable drinking water exist than medical grade sterile water. Therefore, we should stop recommending the use of sterile water in our patients with enteral feeding tubes. Finally, see Table 2 for practical interventions before considering a switch to bottled water for enterally fed patients.
Nutrition Issues In Gastroenterology, Series #155
Refeeding the Malnourished Patient: Lessons Learned
Stacey McCraySymptoms of Refeeding Syndrome (RS) can vary from a mild fall in serum electrolytes to critical electrolyte disarray and even death in the most severe cases. The goals of this article are to help clinicians better understand the mechanism of RS, recognize patients at risk, and identify the clinical circumstances that may require special attention.
Refeeding Syndrome (RS) was first recognized in the 1940s in starved prisoners of war who suffered complications after being refed. Today, the problem has become more widely appreciated due to current advances in medical care and nutritional support. However, despite the increased recognition, no standard definition or treatment approach has been established by randomized clinical trials. Symptoms of RS can vary from a mild fall in serum electrolytes to critical electrolyte disarray and even death in the most severe cases. The goals of this article are to help clinicians better understand the mechanism of RS, recognize patients at risk, and identify the clinical circumstances that may require special attention.
Stacey McCray RD Program Coordinator, Medicine Nutrition Support Team. Carol Rees Parrish MS, RD Nutrition Support Specialist. University of Virginia Health System Digestive Health Center of Excellence Charlottesville, VA
CASES
Which of the following cases are refeeding?
(Answers at the end)
For all cases, normal (UVA) reference ranges for
electrolytes are as follows:
Phosphorus (Phos): 2.3 – 4.5 mg/dL (0.74-1.45 mmol/L)
Magnesium (Mg): 1.6 -2.6 mg/dL (0.66-1.07 mmol/L)
Potassium (K+): 3.4 – 4.8 mEq/L (3.4 – 4.8mmol/L)
Case #1
65 year old male admitted to the ICU with COPD exacerbation. Patient was well nourished prior to admission (just returned from a Caribbean cruise with his family). Now intubated and sedated. Enteral feeding initiated at a low rate within 24-48 hours of admission. Phosphorus level on hospital days 2 and 3, respectively: 1.7 mg/dL (0.55mmol/L) and 1.9 mg/dL (0.61mmol/L). Magnesium and potassium levels were within normal limits.
Case #2
65 year old female admitted with fever, UTI, and
dehydration. History of hypertension and stroke four
months ago when she was discharged to a skilled
facility on a pureed diet with thickened liquids. Her
weight on discharge to the facility was 65 kg; at the
time of this admission she was 56 kg. She failed a
swallow evaluation and enteral feeding was initiated
via nasogastric tube. Next morning labs revealed:
Phos: 1.8 mg/dL (0.58 mmol/L)
Mg: 1.4 mg/dL (0.58 mmol/L)
K+: 3.1 mEq/L (3.1 mmol/L)
Case #3
45 year old female admitted from the ER with diabetic ketoacidosis. Eating well until 2 days ago when she became ill from a virus and stopped taking food and medications. Current weight 62 kg; usual weight: 65 kg. Receiving IV fluids at 125mL/hr, an insulin drip with potassium replacement. On admission her potassium level was 5.8 mEq/dL (5.8 mmol/L), phosphorus level was 4.6 mg/dL (1.49 mmol/L), and magnesium level was 2.5 mg/dL (1.03 mmol/L). Phosphorus level now: 1.4 mg/dL (0.45 mmol/L).
INTRODUCTION
Refeeding syndrome (RS) is the metabolic response to nutrient provision in a malnourished patient. The driving force behind RS is the physiologic shift from a starved, catabolic state to a fed, anabolic state. Under normal conditions, the body’s preferred fuel is carbohydrate. Carbohydrate is stored as glycogen in the liver for readily available energy. During starvation, glycogen stores are depleted, and the body responds by utilizing protein and lipid as the primary fuel source. This shift in fuel source results in decreased insulin levels and increased glucagon levels. Prolonged starvation will lead to decreased lean body mass as muscle is burned for energy. This results in decreased skeletal, cardiac, and respiratory muscle mass, as well as overall strength.
Prolonged periods without nutrition also result in total body loss of electrolytes (including phosphorus, magnesium, potassium), as well as vitamins and minerals. Serum electrolyte levels may not reflect total body stores as only about 1% of phosphorus and magnesium stores are reflected in the serum level.1,2 Serum electrolyte levels may remain normal despite overall depletion; this can be attributed to adaptation, intracellular contraction, decreased renal excretion, and/or dehydration.3-5
Insulin, in response to carbohydrate provision, is the primary stimulus for the cascade of events associated with RS. Insulin not only drives glucose into the cells, but also vitamins and electrolytes required for utilization of the substrate. This intracellular shift of electrolytes (and resulting decreased serum levels) account for many of the clinical complications associated with RS.
Signs and Symptoms
Symptoms of RS will vary from mild drops in serum electrolytes to severe electrolyte disorders with complications, or even death. Most symptoms will first occur between 1-3 days after refeeding is initiated,6 although in some cases up to 5 days.7 The duration of symptoms will vary based on the degree of malnutrition, feeding advancement and other factors. There is no standard definition for what defines RS or how many symptoms must be present to constitute RS. The majority of symptoms associated with RS are due to electrolyte dysregulation with cardiac, respiratory, neurologic and other systems affected (see Table 1). Cardiac arrhythmia is the most common cause of death from RS.3
Hypophosphatemia
Hypophosphatemia is the classic sign associated with RS. In fact, some authors have suggested that the term “Refeeding Hypophosphatemia” may be more appropriate for cases where hypophosphatemia is observed, and no other electrolyte disorders or symptoms of RS are present.7 In a review of 27 cases of RS, hypophosphatemia was documented in 96% of the cases.7
Phosphorus is required for a number of systems including the respiratory, neuromuscular, cardiac, endocrine, and hematologic systems.2,5,8 Phosphate is a component of adenosine triphosphate (ATP) and therefore is critical to providing energy to the cells. Phosphate is important in respiratory and cardiac muscle function, white blood cell function, nerve conduction, and oxygen delivery. Phosphorus is required for the pathway which allows for the release of oxygen from hemoglobin.9 Respiratory alkalosis or metabolic alkalosis can cause phosphorus redistribution, resulting in decreased serum phosphorus concentration.8 Hypophosphatemia has been shown to result in longer length of stay, longer ICU and ventilator days, and a higher mortality rate.10,11
Hypomagnesemia and Hypokalemia
Other serum electrolyte abnormalities are associated with RS, primarily magnesium and potassium. Magnesium is required for more than 300 enzyme pathways.1 Among its many functions, it is important in the synthesis of proteins and is required for normal muscle, cardiac and nerve function. Hypomagnesemia is defined as serum Mg < 1.8 mg/dL (0.74 mmol/L), although symptoms most often occur with Mg < 1.2 mg/ dL (0.5 mmol/L).1 Hypomagnesemia can lead to muscle weakness, ventricular arrhythmia, neuromuscular problems, metabolic acidosis and anorexia.
Hypokalemia (serum potassium < 3.5 mEq/L [3.5mmol/L]) can lead to weakness, paralysis, and confusion. Severe hypokalemia can lead to life threatening arrhythmias, cardiac arrest or sudden death. Because of the severity of potential complications, hypokalemia is rarely left unattended by the medical team and is usually replaced promptly.
A full list of complications associated with hypophosphatemia, hypomagnesemia, and hypokalemia is available.12
Other Complications
Complications other than electrolyte disarray may also occur. Increased carbohydrate provision may decrease water and sodium excretion, resulting in fluid overload. This is most common in severely malnourished patients, such as those with anorexia nervosa. Hyperglycemia can be seen as carbohydrate is provided to a body adapted to fat metabolism. Micronutrient deficiencies are likely if the patient has been without adequate nutrition for a prolonged period.
Thiamine should also be of primary concern in the patient at risk for RS. Depleted thiamine stores can lead to neurological compromise and other complications (Wernicke’s encephalopathy). Thiamine supplementation should be provided to patients with a history of alcohol abuse, as well as patients who are markedly malnourished for any reason. In addition, until better data is available, thiamine is provided in our institution both before and for the first few days of feeding in these patients. Theoretically, if thiamine is given without concurrent nutrient delivery, there may not be “recruitment” (i.e. demand for thiamine), and it is unclear whether the thiamine would be utilized. Several recent reviews of thiamine and Wernicke’s encephalopathy are available.13-15
Incidence
The true incidence of RS is difficult to determine, as there is not a standard definition for RS. The incidence reported in the literature varies greatly and is often based solely on the appearance of hypophosphatemia. Reported rates in specific populations include:
- 34% of all ICU patients10
- 10% in anorexic patients admitted to the ICU16
- 15% of hospitalized patients17
- 9.5% of patients hospitalized for malnutrition from gastrointestinal fistulae18
- 48% of severely malnourished patients being refed19
This broad range in reported incidence is likely due the wide variety of patient populations reported upon; varying degrees of malnutrition among the populations, different criteria used to diagnose malnutrition, different definitions of RS, and varied refeeding protocols among institutions.
Confounding the identification of RS is the fact that electrolyte disorders have many causes in the hospitalized setting. Therefore, it is important to remember that not all low electrolyte levels are a result of RS. Metabolic or respiratory acidosis, sepsis, volume repletion, changing renal function, initiation or stopping of insulin drips, or other factors may affect phosphorus levels. Many medications may lower serum phosphorus as well. Patients, such as those with COPD, may experience hypophosphatemia when mechanical ventilation is initiated. This is due to the intracellular shift of phosphorus that occurs when pH normalizes as respiratory acidosis corrects.8 Table 2 lists some of the causes of hypophosphatemia in the hospitalized patient. The myriad of factors altering potassium and magnesium are outlined in an earlier article.12
Patients at Risk 9,12,20
Any patient who has been without adequate nutrition for a prolonged period of time may be at risk for RS. Critically ill patients may experience hypophosphatemia upon refeeding after a relatively short period of time (48 hours) without nutrition.9 Table 3 identifies some conditions that put patients at risk for RS. Interestingly, and likely to become more prevalent, is RS seen after severe weight loss from gastric bypass surgery.21,22 A recent study by Manning cites a low incidence of RS in alcoholics; however, it is important to note that these patients were not identified as malnourished, presented voluntarily, and were provided with an oral diet as desired.23 Patients with chronic alcohol abuse should be presumed to have a component of malnutrition and be provided with thiamine (opinion of authors).
The National Institute for Health and Clinical Excellence (NICE) in England and Wales published guidelines in 2006 for identifying patients at high risk for RS.24 While such screening tools may be helpful, it is often difficult to determine which patients will show signs and symptoms of RS. Zeki, et al. retrospectively reviewed the records of 321 hospitalized patients.17 The authors evaluated the risk for RS based on the NICE guidelines, and looked at serum phosphate levels before and after feeding initiation. Ninety-two patients (29%) were identified at risk of RS; of these, 23 patients (25%) developed refeeding hypophosphatemia (RH) compared with 26 patients (11%) who were not identified at risk, but still developed refeeding hypophosphatemia (p=0.003). This study demonstrates that not all patients identified at risk will show symptoms, and some patients not identified at risk will experience signs of refeeding. Other authors have also found that patients identified at risk do not always go on to develop RS.25
RS can occur when consistent nutrients are provided regardless of source-oral, enteral, parenteral nutrition or IV dextrose. While in the past, overzealous PN was associated with RS, other reports have shown that RS can occur when any source of nutrition is provided.7,17 In the Zeki article discussed above, the authors found that at-risk patients in the enteral group were more likely to develop hypophosphatemia than at-risk patients in the PN group.17 The authors postulate that lower levels of phosphorus seen in enteral feeding compared to PN may play a role, as well as increased stimulation of insulin secretion with enteral compared to PN due to first pass metabolism may be responsible.
Treatment
There is no one regimen that has been proven to prevent RS. Recently, one group undertook one of the first randomized, controlled trials to assess outcomes associated with a treatment regimen for RS in critically ill patients.26 Patients who experienced hypophosphatemia upon feeding (<2 mg/dL [0.65mmol/L] within 72 hours of feeding) were randomized to either the control group vs. calorie restricted group. Patients in the control group (n=170) continued on nutrition support per standard protocol. Patients in the calorie restricted group (n=169) received 20 kcal/hour for 2 days, and then advanced to goal in a stepwise manner over several days. The results of this study were mixed. In the short term, the calorie restricted group had higher phosphorus levels on days 1 and 2, and less hyperglycemia on days 1-4. There was no difference in the primary study outcome of days alive after discharge from the ICU. This study provides some additional information; however, nutrition support was initiated at standard levels and the study group had calories decreased if hypophosphatemia occurred. This is different than the more standard practice of beginning nutrition support conservatively in patients at risk for RS and replace electrolytes as the need arises (see Table 4).
The NICE guidelines recommend initiating calories at 10 kcal/kg in patients at high risk for RS.24 In the most severe cases, such as patients with anorexia nervosa, even lower levels may be recommended.16,24 Hofer, et al. evaluated 86 cases of severely malnourished anorexia nervosa patients (in 74.4% of cases, patients were less than 70% IBW).27 The authors evaluated the use of a refeeding regimen which included initiation of feeding at 10 kcal/kg/day, fluid and sodium restriction, and electrolyte supplementation and monitoring (for full protocol see article). They found a low incidence of complications with this protocol, and no deaths were reported.
For patients deemed to be at mild to moderate risk of RS (or those patients where risk is unclear and clinicians just want to err on the safe side), such low calorie levels may not be necessary. Prior to the NICE guidelines, a calorie level of 15-20 kcals/kg was generally recommended for most patients at risk for RS.12 and is currently used at our institution unless a patient is deemed to be at severe risk.
While repleting the malnourished patient is essential, repletion can only occur so quickly and a “rush” to refeed may lead to complications. On the other hand, once the refeeding calorie level has been established, there is no need to also initiate nutrition support below the designated refeeding calorie goal. For example: if the refeeding level is determined to be 1000 kcals/day, a continuous tube feeding rate for a 1 cal/mL product would be ~45mL/hr. We base our flow rates on 20-22 hours per day, as this is what is typically received.
When initiating nutrition support, all calorie sources should be taken into account such as: D5%, D10%, or calories coming from glucose or lipids in medication administration. These calories alone can cause RS in a malnourished patient. If it is not possible to stop any of these additional calorie sources, the nutrition support regimen should be adjusted to take these calories into account. Protein calories should always be included as part of the total calories.
Calories should be increased slowly as the refeeding risk subsides. An advancement of 200-300 calories every 1-3 days is generally recommended.28,29 It is important to make sure this advancement takes place so that patients are not left on hypocaloric feeding levels for a prolonged period of time. This is especially important if a patient is discharged home during the advancement period. It is also important to monitor whether patients are actually receiving and utilizing (see special circumstances discussed later), the nutrition prescribed during this period before further calorie advancement. Feeding interruptions, NPO status, hyperglycemia, malabsorption, or other issues may thwart efforts at nutrition support and leave patients still at risk for RS and further malnutrition.
Electrolyte Monitoring and Replacement
Electrolytes should be checked prior to initiation of nutrition support and low levels replaced. However, there is no need to withhold nutrition support until electrolyte levels are normal.20,24 Some guidelines recommend proactive supplementation of electrolytes, vitamins, and minerals.24,27 Thiamine should be provided to severely malnourished patients and those with a history of alcohol abuse or chronic vomiting prior to the initiation of nutrition support.
Serum electrolytes should be checked after 8 – 12 hours of nutrition support initially, then daily during the refeeding period (first 48-72 hours). The frequency and duration of electrolyte monitoring will vary depending on the degree of malnutrition and whether electrolyte disorders occur, as well as their severity.
Mild to moderate drops in electrolytes can be replaced orally/enterally in patients with a functioning GI tract. Severely low levels should be replaced intravenously. IV replacement may also be necessary for patients without a functional GI tract, those who do not seem to be responding to enteral replacement, or in other situations where oral replacement is not possible or is contraindicated. Specific guidelines for phosphorus and magnesium replacement are available.12
Oral magnesium replacement may be poorly absorbed and can have a cathartic effect, causing or exacerbating diarrhea. As an example, some forms of magnesium (such as Magnesium Citrate and Magnesium Sulfate) are available over the counter as laxatives. Consider slower, more gradual dosing of oral magnesium (smaller doses over the day, or given at night on an empty stomach before bed), or forms of magnesium which provide more free magnesium per dose to the GI tract (such as magnesium oxide).
Intravenous magnesium is often given as a bolus over 60 minutes. This exceeds the renal threshold for magnesium and the kidneys will dump 50% or more of this dose.30 Experience at our institution indicates a slower IV infusion of magnesium (over 10-12 hours) will be better utilized and retained. In this era of shortages and increasing healthcare costs, it is important to ensure the therapy being provided is being utilized.
Hypomagnesemia can exacerbate hypokalemia and make it more difficult to replace potassium. Potassium levels may not normalize until the corresponding hypomagnesemia is corrected. According to one report, 42% of patients with a low potassium level will also have a low magnesium level.31 Concurrent hypomagnesemia may also worsen the symptoms associated with hypokalemia. In addition, a recent review reports that patients with a history of alcohol abuse are often deficient in magnesium and discusses the role of magnesium in the treatment of Wernicke’s encephalopathy.14
SPECIAL CONSIDERATIONS
Renal Failure
Patients with renal failure are at high risk for malnutrition.32,33 However, due to the underlying disease state, these patients may have elevated serum electrolyte levels when feeding is first initiated. Serum levels may drop more gradually or over a longer period of time due to the “protective” effect of the renal failure. Therefore, in such patients, there may be a delayed response to refeeding. Electrolyte levels may need to be monitored over a longer period of time, and replacement may be needed after several days of nutrition support rather than in the first few days. Electrolyte replacement must be done carefully in patients with renal failure.
Hepatic Failure
Patients with severe end-stage liver disease will have depleted glycogen stores and may be unable to maintain serum glucose levels within a safe range.34 This also may occur in patients with anorexia nervosa or severe malnutrition from any cause. In patients unable to maintain their serum glucose level in a safe range, dextrose infusion (D10%) may be required. In some, this infusion may provide calories that exceed the refeeding calorie goal. However, maintaining serum glucose levels above 80mg/dL (4.44mmol/L) trumps any concern for RS. Electrolytes should be monitored closely and replaced as needed. Adequate thiamine, as well as other vitamins and minerals, should also be given during the first few days.
Hyperglycemia or Diabetic Ketoacidosis
If a patient presents with hyperglycemia, or becomes hyperglycemic after nutrition support is initiated, the refeeding process may be delayed as there is inadequate insulin to drive glucose and electrolytes into the cells. Hyperglycemia is essentially a continuation of the starved state (“starvation in the midst of plenty”). When insulin therapy is initiated, the refeeding response is accelerated. Clinicians should anticipate this response, and monitor for signs of RS. Note that if insulin therapy is delayed and the patient is hyperglycemic for the first days of feeding, the signs of RS may be seen later, once insulin therapy is initiated and glycemic control is achieved.
Treatment for diabetic ketoacidosis is a form of refeeding.12,35,36 Exogenous insulin provided to treat diabetic ketoacidosis will power glucose and electrolytes into the cells causing electrolyte levels to drop and supplementation will be needed. Of note, exaggeration of hypocalcemia can be seen with aggressive phosphorus repletion (8.5mmol/hr, or 6 g inorganic phosphate); therefore caution is required.37
GI Tract issues
Patients with gastrointestinal disease or malabsorptive disorders also may face unique challenges related to RS. Several scenarios can be seen.
first, it is not unusual for patients undergoing work- up on a GI service to frequently be made NPO for any number of reasons (procedures, GI bleed, access issues and symptoms). If the patient is to receive oral or enteral nutrition support, delivery may be inconsistent at best, and the total amount of nutrition provided may vary greatly from what is ordered. The amount of nutrition actually received needs to be determined in order to evaluate whether the patient has received enough nutrition for RS to occur. For example, a patient identified at refeeding risk may have an order to receive nutrition support on day #1, but be NPO off and on the next several days (while refeeding electrolytes are being monitored). If consistent nutrition actually starts several days later, RS may occur at that time. Ongoing monitoring should be coordinated with this timing. Also, clinicians may need to consider continuing thiamine supplementation, as the demand for thiamine occurs when patients are fed, not while they are NPO. It is unknown if patients will just excrete supplemental thiamine in the non-fed state, so until better data is available, it may be prudent to continue dosing until feeding is consistent for 3-5 days (opinion of the authors).
In patients with possible malabsorption, RS will not occur if the nutrition delivered is not absorbed (unless an IV source of nutrition is provided). For example, if a malnourished patient with suspected malabsorption shows signs of RS when enteral nutrition is provided, at least some absorption is occurring. This is certainly a �gross’ test at best, but it does provide some clues to the level of absorption. Patients with malabsorption may receive enteral nutrition for a period of time, but at some point require the initiation of parenteral nutrition due to failure to thrive, change in status, etc. Patients should be monitored for RS at this time also (even though they have been “fed” for some time).
BACK TO CASES
Case #1
Unlikely. Patient well-nourished prior to admission. However, mechanical ventilation and correction of respiratory acidosis in a patient with COPD can lead to significant hypophosphatemia (8).
Case #2
Most likely RS. Multiple electrolyte disorders in a patient with significant recent weight loss.
Case #3
Possibly. Resolution of DKA will mimic RS due to the potent effects of insulin driving electrolytes intracellular. Although this patient may be refeeding after experiencing excess loss of potassium, magnesium and phosphorus in the urine due to the catabolic effects of DKA, the exogenous insulin provided will accelerate serum drops as glucose and electrolytes move intracellularly.
SUMMARY
RS is a concern for any patient who has been without consistent or adequate nutrition for a prolonged period of time. Serious complications can be avoided with appropriate identification of patients at risk, slow initiation of feeding, and careful monitoring. An understanding of the causes and mechanisms of RS can aid the clinician in better caring for patients, as well as recognizing when special circumstances arise or additional care and monitoring may be needed.
Nutrition Issues In Gastroenterology, Series #154
Immunonutrition in 2016: Benefit, Harm or Neither?
Kelly RoehlClinicians remain intrigued by the potential to alter the immune response through nutrition, yet there is much debate on what is considered efficacious use of immunonutrition (IN). Here we discuss evaluation of out
Over the past two decades, there have been numerous clinical trials, meta-analyses, and systematic reviews on the use of immunonutrition (IN) in a variety of populations. Although clinicians remain intrigued by the potential to alter the immune response through nutrition, there remains much debate on what is considered appropriate and efficacious use of IN, including lack of consensus from critical care guidelines and the international nutrition support community. Clinicians practicing in nutrition support must first evaluate outcome benefit, as well as consider the patient population and cost when determining whether IN is appropriate. While administration of IN prior to or following elective GI surgery, may be beneficial in preventing post-op infectious complications and reduce hospital length of stay (LOS), there is inadequate evidence to support the routine use of IN among the critically ill population as a whole.
Kelly Roehl, MS, RDN, LDN, CNSC, Advanced Level Dietitian, Rush University Medical Center, Chicago, IL
INTRODUCTION
Infection is the most common cause of morbidity and mortality following surgery1 and during critical illness,2 potentially resulting in prolonged length of stay and increased hospital costs.3,4 Enteral nutrition (EN) support is currently provided as the standard of care in an effort to prevent degradation of lean body mass (LBM) for gluconeogenesis and prevent malnutrition, a risk factor for infectious complications. Over the past two decades, interest has moved to not only prevention of malnutrition, but also modulating the immune response through nutrition, often referred to as immunonutrition (IN). The potential for altering the immune system and associated clinical outcomes is exciting, but current research and practical implications are not robust enough to drive practice. The aim of this article is to review evidence to date on the safety, efficacy and recommendations for use of IN.
Overview of immunonutrition (IN)
Specific nutrients and dietary components, including arginine, glutamine, selenium, omega-3 (n-3) fatty acids, (eicosapentaenoic acid [EPA] and docosahexaenoic acid [DHA]), the omega-6 gamma-linolenic acid [GLA], nucleotides and/or antioxidants have been implicated for their potential to modulate the metabolic response to surgery or stress by enhancing immune function. Specialty enteral products have been developed to include nutrients that are believed to enhance or modulate the immune response (Table 1). Many of the IN enteral formulations currently available were designed for use among those undergoing gastrointestinal (GI) surgery, and are therefore elemental or semi-elemental as a presumed necessary criteria.
The composition of the IN enteral and oral products available varies greatly, not only in nutrients, but also the concentration of each specific component. Unfortunately, clinical trials of the individual potentially immune-modulating nutrients have either not been conducted, or have failed to demonstrate benefit.5,6 It has yet to be established which, how much (if any), when, and for whom IN may provide benefit.
Meet the “Immune-Modulating” Nutrients
Glutamine, most notably known as the primary fuel for enterocytes, lymphocytes and macrophages,5 is also a conditionally essential amino acid during metabolic stress. It serves as a substrate for gluconeogenesis, and may be oxidized for fuel for rapidly proliferating cells.8 Additionally, it is a precursor for renal ammoniagenesis, the process by which ammonia is excreted from the body.8
Arginine is a conditionally essential amino acid during metabolic stress as it is a precursor for many compounds within the human body. It is required for normal T- and B-lymphocyte and macrophage functions, and can be metabolized and utilized in collagen production by way of proline synthesis.9
Arginine stimulates secretion of growth hormone, insulin, and glucagon,10 and can be metabolized to nitric oxide, thereby altering blood flow, angiogenesis, epithelialization, and tissue granulation.11
Omega-3 fatty acids, specifically EPA and DHA, are believed to be immunosupressive by reducing the production of the pro-inflammatory omega-6 fatty acid, arachanonic acid, whose production results in higher levels of the pro-inflammatory eicosanoids, prostaglandins, leukotrienes, and thromboxanes.12 Furthermore, EPA and DHA are postulated to reduce macrophage adhesion, alter T-cell proliferation, and stabilize the cytokine response.13 Some have suggested that arginine and n-3 fatty acids may synergistically improve immune function with:
- 1. arginine delivery improving cytokine and nitric oxide production,
- 2. n-3 fatty acids reducing pro- inflammatory eicosanoid production, and
- 3. increasing arginine availability by decreasing expression of arginase I, an enzyme responsible for degradation of arginine.14,15
Given the role of nucleotides in structural integrity of DNA and RNA, and involvement in the transfer of energy and coordination of hormonal signals, they are often added to IN formulas intended for use during times of stress and/or rapid tissue proliferation.7 Interestingly, the processing techniques utilized in the production of commercial EN formula results in the removal of nucleotides;therefore, some have suggested that standard EN products do not provide adequate nucleotide content for those experiencing metabolic stress.13
Antioxidants, including vitamins C and E, beta- carotene, and selenium are often added in an effort to reduce oxidative stress among patients with acute metabolic stress.
A number of formulas with varying IN compositions are available in the United States (Table 1). Some of these products have been used in research in attempts to demonstrate efficacy for their use, but many of the products have never been tested for efficacy or safety in the populations for which they are marketed or in a clinical trial of any kind.
Reviewing the Evidence
Although immune-enhancing nutrition has been explored in a variety of settings, including pulmonary, trauma, neurology, oncology, and critical care, much of the research has been conducted among patients with GI disorders, specifically elective surgeries for cancers of the GI tract. Those undergoing elective surgery are an attractive and easy group to study because enteral and/or oral nutrition support is often utilized to prevent unintended complications related to malnutrition as many patients struggle to meet nutrition requirements orally during the pre- and post-operative periods.
Over the past two decades, there have been at least 16 meta-analyses and systematic reviews to evaluate the efficacy of IN among patients undergoing elective surgery (Table 2) and the critically ill (Table 3), yet use of IN remains controversial, particularly among the critically ill. In fact, the most recent Guidelines for the Provision and Assessment of Nutrition Support Therapy in the Adult Critically Ill Patient, jointly published by the American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) and Society of Critical Care Medicine (SCCM), recommends that immune-modulating EN formulations not be used routinely among medical ICU patients, reserving it for those with traumatic brain injuries and perioperatively in the surgical ICU populations.31 Additionally, they do not recommend routine use of fish-oil and antioxidant-containing EN among patients with ARDS or ALI, citing insufficient evidence and conflicting data. Much of the backing behind these recommendations stems from research with wide heterogeneity and inconsistency in outcomes, as well as meta-analyses. Since methodologic and funding concerns blanket much of the IN research, it is an important point to consider that the strength of any meta-analysis or systemic review is only as strong as the studies that they are comprised of.
Review of Efficacy for Use of IN Among Elective Surgical Populations
Among those undergoing elective surgery, most commonly for GI malignancy, improvements in post- operative infectious complications and LOS may result in reduction in cost of care. Additionally, pre-operative nutrition status, a topic that itself has a murky array of definitions, may explain the differences found in pre- op versus post-op IN outcomes.32
Despite at least 10 meta-analyses and systematic reviews (Table 2), it remains unclear which nutrients, how much, timing, length of treatment, and specific surgical populations may benefit from IN. Researchers generally conclude that provision of IN among patients undergoing elective surgery may reduce incidence of infection and decrease hospital LOS, but find no reduction in mortality. A more critical evaluation of the meta-analyses reveals wide heterogeneity with regards to population and volumes of feeding delivered, therefore potential differences in amount of IN components delivered. According to one group, perioperative administration of 500-1000 mL/day of an IN formula for 5-7 days prior to surgery, with continuation into the post-op period reduces infection, other complications and hospital LOS, regardless of preexisting nutrition status.33 Although they conclude that single-substrate administration does not impact clinical outcome, and describe a potential synergistic effect between arginine and fish oils, recommending that these nutrients be used together, this has yet to be proven. Given the variation in formula composition and actual amounts delivered in various studies, it is impossible to determine which specific nutrient is potentially improving outcomes, if any.
To offer fair comparisons between groups where nutrition is provided to both, allowing IN to be the intervention or treatment, nearly all of the randomized controlled trials (RCTs) that provide the basis for the meta-analyses and systematic reviews described in Table 2 compare administration of IN to standard EN. Similar reductions in LOS have been reported when IN was utilized in the pre- versus post-operative periods.34 Hegazi, et al. reported that pre-op oral IN only provided benefit when compared to those that received non- supplemented oral diets,25 suggesting that adequate delivery of basic nutrients results in prevention of post- op complications. However, given that the standard of care (control) is no nutrition intervention, perhaps the benefits of preoperative nutrition can be attributed to carbohydrate loading to maximize glycogen stores as recommended for Enhanced Recovery After Surgery (ERAS), which has been shown to significantly reduce complications and hospital LOS.35,36 Though some researchers have reported that pre-op carbohydrate loading may prevent loss of LBM,37-39 reduce insulin resistance, tissue glycosylation in the operative period, and optimize glycemic control post-op,40-42 direct comparisons have not yet been made. Is it simply the provision of extra (or adequate) calories above the ‘standard’ intake the patient would be able to consume usually in the pre-op period that is resulting in benefits? More research is needed.
Review of Efficacy for Use of IN Among Critically Ill Populations
Given the role infectious complications play in the critically ill population, any intervention that might decrease that risk is worthy of investigation. Generally, the outcomes of meta-analyses examining efficacy of IN among the critically ill (Table 3) are similar to those for the elective surgical population with regards to reduced incidence of infection and decreased hospital LOS, with no difference in mortality; however, some researchers16 suggest that provision of IN among the critically ill may result in adverse outcomes, and therefore, be a safety concern. Like that in the elective surgical population, the research for use of IN in the critical care arena is full of methodologic and heterogeneity concerns.
Much of the debate regarding efficacy of IN among critically ill patients surrounds the safety of its use – specifically relating to arginine. In a 2001 meta-analysis, Heyland et al.,16 concluded that arginine-supplemented IN provided no benefit among the critically ill, and may potentially result in adverse outcomes, a conclusion made due to a trend toward increased mortality among those receiving IN; however, these results were not statistically significant. Since this time, concerns regarding the safety of IN, specifically arginine supplementation, among septic patients has been hotly debated; however, research remains limited, and the debate has mainly surrounded three theories (though none confirmed):
- 1. Sepsis results in arginine deficiency and supplementation may improve septic state.43
- 2. Sepsis is caused by excess nitric oxide (NO) production. Since NO is the end-product of arginine metabolism that causes vasodilation, arginine supplementation may exacerbate the septic syndrome.43
- 3. Arginine infusion among septic medical and surgical patients does not cause hemodynamic instability.44
As many of the IN products available contain a number of potentially immune-modulating components, and it remains unclear which (if any) nutrient may be providing the most benefit, researchers have attempted to scrutinize immune-modulating nutrients independent from nutrition delivery.
IN, Individual Delivery, Biomarkers and Outcomes Among the Critically Ill
As the goal of IN is to enhance the immune response, researchers have examined inflammatory biomarkers concurrently with clinical outcomes in attempts to demonstrate potential changes in outcomes. However, it is imperative to remember that changes in surrogate markers do not necessarily translate to differences in clinical outcomes, a point that is often missed in interpretation. One group concluded that delivery of IN EN containing n-3 fatty acids, glutamine and arginine among those with esophageal cancer undergoing concurrent chemotherapy and radiation resulted in a reduced rise in the inflammatory cytokines C-reactive protein (CRP) (p=0.001) and tumor-necrosis factor-alpha (TNF-a) (p=0.014) compared to those receiving standard EN support.45 It is important to note that although statistically significant change in markers of inflammation were found, these authors failed to connect their results to clinical outcomes, which is necessary to drive change in practice.
To further illustrate this point, researchers of the highly publicized ARDS Network Omega Trial (n= 272), administered n-3, GLA, and antioxidants separate from the enteral formulas twice daily.5 Although delivery of n-3 fatty acid increased plasma EPA concentration 8-fold, there were no differences in ventilator-free or ICU-free days among those receiving the supplemental immune-enhancing nutrients.
In the Reducing Deaths Due to Oxidative Stress (REDOX) trial comparing the effects of glutamine and/ or selenium administered separate from the EN formula, unexpectedly, researchers reported longer time to ICU and hospital discharge.6 Interestingly, post hoc analysis revealed that high dose glutamine and/or antioxidants may be associated with increased mortality, especially in those with multiorgan failure. Furthermore, Van Zanten et al.46 found that after adjusting for Acute Physiology and Chronic Health Evaluation II (APACHE II) scores, patients requiring mechanical ventilation that received an IN formula containing glutamine, n-3 fatty acids and antioxidants were found to have significantly higher 6-month mortality than those receiving an isocaloric, high protein formula (54% vs 35% in the control EN group, p=0.04). Conversely, a systematic review concluded that use of fish oil/antioxidant containing enteral formulas or supplements were associated with a reduction in ICU LOS and ventilator days; However, after excluding the Omega trials5 where fish oil was administered as a twice daily bolus outside of the EN, use of continuously administered EN containing fish oil was associated with a significant reduction in mortality (P=0.004).30
The influence of IN nutrients glutamine and selenium among patients requiring both enteral and parenteral support remains inconclusive. Although most have concluded that glutamine and selenium supplementation may result in reduction of nosocomial infections among the critically ill, researchers of one meta-analysis concluded that glutamine supplementation via enteral, parenteral, or a combination of these routes posed no benefit in overall mortality or hospital LOS, but did result in lower incidence of noscocomial infections among the critically ill.47 Furthermore, these researchers, as well as a separate group48 concluded that high-dose supplementation (>0.5 g/kg/day) significantly increased mortality among the critically ill, resulting in higher rates of infection, and longer ICU and hospital LOS. Appropriately, the A.S.P.E.N./SCCM 2016 guidelines suggest that supplemental enteral glutamine (above what is standard in EN formulas) NOT be supplemented in critically ill adults.31
Cost
The potential ability to reduce the cost of medical care was one of the driving forces behind initial efforts to study the impact effect(s) of IN on post-op morbidity, and continues to influence the decision to use IN. Researchers have suggested that IN enteral formulas may be cost-effective when used in specific populations and healthcare settings;49,50 However, this is only if they work, which remains unclear. Products with IN properties are significantly more expensive than standard preparations (Table 1) with some IN EN formulations costing up to 6 times that of a standard formula. Although nutrition support is widely accepted as a life-sustaining therapy, insurance coverage differs among payers and administration settings, making cost-benefit analyses complicated. Differences in coverage may depend on route of administration (oral, enteral or parenteral).51 Therefore, clinicians must be cognizant of coverage to prevent a cost burden not only to the patient, but also the healthcare system as a whole.
CONCLUSION
Despite the large volume of research conducted on efficacy of IN products over the past three decades, there is still no consensus on whether or not they provide benefit. More concerning, some suggest potential risk to the critically ill. Researchers have attempted to find a pattern of potential benefit by conducting meta-analyses and systematic reviews. However, overall, these have revealed no difference in the ultimate outcome of mortality in a variety of populations with enteral IN was compared to standard EN support, in either the surgical and critically ill populations. The literature is riddled with limitations, including research design, heterogeneity, and possible bias from conflicts of interest, thereby preventing the ability to draw solid conclusions and make specific recommendations for clinical practice.
Guidelines and recommendations for use are derived from research conducted by a relatively small group of individuals, many of which receive financial gain/funding from the makers of IN formulas. Given the lack of consensus and exorbitant cost associated with IN, clinicians must demand a well constructed, multi-center, non-biased robust study that addresses the limitations of previous research, and is designed to test the true efficacy of these formulas among critically ill patients.