Page 74 – Practical Gastro

Blenderized tube feeding has been gaining momentum among patients despite the availability of commercially prepared and nutritionally complete enteral formulas. This article will review perspectives and provide clinicians with basic guidelines for planning and implementing blenderized enteral feedings when our patients request this feeding option.

INTRODUCTION

Before the availability of commercial enteral formulas, “blenderized” foods were prepared in hospital kitchens to create liquid mixtures given by bolus syringe method through large-bore nasogastric and gastric tubes. As technology continued to advance in the 1970s, commercial formulas of defined composition were introduced for tube feedings.¹ Blenderized tube feeding (BTF) became a less desirable option when commercial formulas became more affordable, easy to use, and widely available. The risk of microbial contamination and labor intensity were the primary reasons BTF were abandoned by most healthcare facilities. Commercial enteral formulas are packaged aseptically and are approved to hang for 48 hours as a closed system reducing administration error and time involved with tube feeding. Unfortunately, the only tubing available in the U. S. for use with these products cannot hang for more than 24 hours (so this touted 48 hour benefit is not really a benefit). Studies have demonstrated that the microbial quality of hospital- prepared BTF was not within the published guidelines for safety.^2-3 However, in many parts of the world, especially in developing countries, the use of hospital- prepared formula is a routine practice due to economic or cultural reasons.^2-4

Compared with commercial formulas, preparation of BTF can be labor intensive and the cost (is rarely, if ever) covered in medical plans. In addition, there are food safety concerns and uncertainty about the nutritional value of non-standardized recipes. For these reasons, clinicians are hesitant to recommend this option to an already stressed and exhausted patient and /or caregiver. Several studies have shown that hospital- prepared formulas provided unpredictable levels of macro and micronutrients, delivered less than the desired amounts of nutrients, and increased the risk of tube occlusion due to viscosity, making it unsuitable for reliable infusion through feeding tubes.^2,4

However, there remains a subset of the population who are interested in this feeding option. BTF has been gaining popularity among home enteral nutrition (HEN) patients, particularly the pediatric population. Many patients and families choose BTF because of perceived health benefits, intolerance to commercial feeding formulas, food allergies, improved bowel function, psychosocial reasons, or personal preference (desire for “real” food, organic, vegetarian, etc.).^5-6

Use of Blenderized Tube Feeding

BTF is defined as the use of blended foods and liquids given directly via the feeding tube. Historically, these types of enteral formulas have been called “blended diet”, “blenderized feeding,” “blended formula,” or “homemade blended formula.” The practice of incorporating BTF into a feeding regimen could be a combination of a commercial formula and commercial pureed baby food, or three meals a day of homemade blenderized foods supplemented by nocturnal feeding of a commercial formula, or complete feedings using homemade BTF made from recipes, and many variations in between. The introduction of ready to use BTF in the marketplace has also provided HEN patients with an option over commercial enteral formulas. Regardless of how a patient uses BTF, it is essential to identify a commercial enteral formula for emergency situations, or when traveling when refrigeration is not available.⁷

There is little published research available to support the efficacy of BTF that translate into any type of beneficial outcomes of this feeding technique. There are numerous anecdotal reports from patients, caregivers, and medical professionals of positive experiences that have been shared through informal patient questionnaires, feeding support groups on the internet, social media, professional discussions and clinical experiences.^5-9

One feeding clinic reported both medical and emotional benefits from the use of BTF.⁷ It allowed for some normalization of the feeding process for gastrostomy tube-fed patients, greater volume tolerance, and improvement in reflux and constipation, and it facilitated the transition from tube to oral feeding. The use of blenderized foods allowed for inclusion of a tube-fed patient in family mealtimes and a sense of “normalization” of gastrostomy tube feedings. It also promoted the view of the G tube as another mouth, thereby priming the gastrointestinal system for the complexities of food.

The interdisciplinary feeding team at the Cincinnati Children’s Hospital Medical Center (CCHMC) conducted a feeding trial using a pureed diet given directly into the feeding tube, referred to as the pureed by gastrostomy tube (PBGT) diet. It was designed for children with gagging and retching after fundoplication surgery.⁸ This was the first clinical trial using BTF to manage the complications associated with enteral feeding. Fifty-two percent of parents reported a decrease in gagging and retching after their child started the PBGT diet and 57% of children were reported to have an increased oral intake.⁸

When to Consider Blenderized Tube Feeding

Table 1 outlines the prerequisites before seriously considering the use of BTF. Ideally, a clinician must first determine if a patient is a good candidate for BTF. However, a tube-fed patient may already have transitioned to either partial or full homemade BTF regimen before a referral is made. Working with a dietitian is essential to ensure that the homemade diet is adequate and whether the current homemade BTF recipe plan needs modification. The best candidate is a patient and/or caregiver who made the decision to “try” this feeding option, and is willing to commit their time and effort for instruction and preparation of BTF. A patient who is having tolerance issues or allergy to a commercial enteral formula may also be a candidate after discussion with the medical team and dietitian.

When Not to Consider Blenderized Tube Feeding

BTF is not an option for all tube-fed patients. Patients with complicated medical and gastrointestinal issues and those who require frequent hospitalization may not tolerate and sustain a BTF regimen. Often, these patients require specialized enteral formulas. A patient who requires continuous feeding is not a good candidate since a homemade blended formula is not recommended for feedings that will last for more than 2 hours due to concerns over food safety and bacterial contamination.⁵

Homemade BTF is generally thicker and can potentially clog the feeding pump making it difficult to flow through the feeding set. Some patients have successfully infused BTF through a feeding pump as long as the mixture is thinned with additional fluid, blended, and strained sufficiently. Bolus syringe method works best and provides the pressure needed to move a homemade blended formula down a feeding tube. It is recommended for gastrostomy tubes 14 French size or larger to prevent clogging and for ease of administering the diet.

Tools for Success

Heavy duty blende
- Blendtec® HP3 blender: www.blendtec.com (800) 253-6383
- Vitamix®: Inquire about Vitamix® Medical Needs Discount Program which is available to all eligible candidates at (800) 848-2469 or email: household@vitamix.com reference code 07-0036-0011
Strainer or fine sieve if using a regular kitchen blender or stick blender (see Figure 1)
Airtight storage containers, ice cube trays for freezing individual portions
Adequate refrigeration/ freezer space
60 mL syringe with plunger
Bolus extension set for low-profile gastrostomy tube
Feeding pump (if using)
- Make sure BTF is thin enough to flow easily through the pump
- Discard BTF after 2 hours maximum if kept at room temperature for that long
Insulated bag or ice chest with ice packs when traveling
Patient education on food safety guidelines (www.fsis.usda.gov, www.foodsafety.gov, www.homefoodsafety.org)

Getting Started with a Homemade Blenderized Tube Feeding

Evaluate the patient’s medical history, success (or not) with current feeding regimen (feeding tube, oral intake), food tolerances, lifestyle/ethnic/religious preferences, ability to obtain individual ingredients and tools needed for preparing a homemade BTF.
Discuss with medical team and determine if patient is ready to start the transition. Most patients transition slowly and use a commercial formula for part of nutrient requirements.
Determine goals for calories, protein, fluids, and vitamin, mineral, and electrolyte supplementation.
Develop a meal plan and starter recipe. The following tools can be used as starting point to create a meal plan based on caloric goals:
- USDA Choose My Plate http://www.choosemyplate.gov/ supertracker-tools/supertracker.html The supertracker feature provides individualized worksheets where a profile can be created to calculate and track a menu plan. A sample 1000 calorie meal plan will include:
  - Grains: 3 servings
  - Fruits: 1 serving
  - Vegetables: 1 serving
  - Protein (meat, beans, or nuts): 2 servings
  - Milk or milk substitute: 2 servings
  - Fats: 3 servings
- Homemade blended formula worksheets by Dunn Klein M, Morris SE. Homemade blended formula handbook. Mealtime Notions, LCC, Tucson, AZ, 2007; 117-128; www.mealtimenotions.com The worksheets were adapted from the USDA My Pyramid.
- Sample Blenderized Tube Feeding Recipes at www.ginutrition.virginia.edu under Resources for Nutrition Support Clinicians
- Nutrition and recipe analysis applications. Computerized nutrition programs such as Food Processor®, Nutritionist Pro^TM allows the clinician to add and modify foods while monitoring the total caloric and protein levels along with vitamin and mineral profile. The clinician can save and retrieve the data for future reference.
- USDA National Nutrient Database for Standard Reference: http://www.nal.usda. gov/fnic/foodcomp/search
- Exchange Method
- Food company websites/food labels
Determine macronutrient food sources. The following examples work well in blenderized tube feedings:
Grains: cooked cereals, boiled white or brown rice, cooked quinoa, oats, regular or whole grain bread
Fruits: avocado, applesauce, peach, pear, banana, papaya, blueberries, 100% fruit juice (pulp free). Commercial pureed baby food (stage 2) can be used for variety, consistent nutritional value, and it also avoids the potential complication of clogging the feeding tube. It is convenient especially when traveling, and avoids the need to purchase an expensive heavy-duty blender. The patient can eventually advance and transition to blenderized table foods.
Vegetables: white potato, sweet potato, carrots, squash, well-cooked broccoli. Can use commercial pureed baby food (stage 2).
Protein: chicken, beef, legumes, soft tofu, smooth 100% peanut butter, cooked eggs, canned tuna or other fish without bones. A commercial enteral formula can be used as the base for a blended diet instead of meat, milk, or yogurt as the sole protein source (see Table 2). Milk and yogurt tend to blend more easily than cheese.
Milk or milk substitute: cow’s milk, soy milk, almond milk, rice milk, yogurt, non- fat milk powder.
Fats: canola, olive, flaxseed, hemp and corn oils.
Example of a 1000 calorie blend:
- Grains: 1 cup cooked oatmeal and 1/2 cup cooked brown rice
- Fruits: 1/2 cup peaches (canned or fresh) and 1/2 cup unsweetened apple juice
- Vegetables: 1/2 cup cooked carrots and 1/2 cup cooked sweet potato, butternut squash or pumpkin
- Protein: 2 ounces cooked chicken
- Dairy: 1 cup whole milk and 1 cup yogurt (plain)
- Fats: 3 tsp olive oil
Review and modify the recipe using a recipe analysis program to adjust the ratios of macronutrients to the desired composition.
- A modular product (glucose, MCT oil, protein, fiber such as Benefiber®) can be added to enhance the nutrient profile of blenderized enteral feedings.
Determine fluid goals. The percentage of free water is calculated from the blended recipe by multiplying the total volume of fluid- containing ingredients (commercial formula, meats, fruits, vegetables, milk, and yogurt) in ounces by 0.75. This is based on the assumption that most infant foods contain roughly 75% free water (9). Additional free water is determined from the difference between estimated daily fluid requirement and amount of free water of the blended recipe.
Determine need for vitamin, mineral, and electrolyte supplementation. Perform a recipe analysis. Request labs as appropriate for any nutrients of concern. A multivitamin may be crushed and added to one of the bolus feeds. A liquid multivitamin is an option, but is not always complete – make sure patient gets the right one. Oral rehydration solution (ORS) can be given instead of free water flushes between bolus feeds.
Patient and/or caregiver education
- meal plan, ingredients, starter recipe
- equipment: blender, strainer, 60 mL syringe, storage containers
- preparation details
- proper sanitation method
- proper storage and refrigeration
- administration method and feeding schedule (bolus syringe, pump)
- water or oral rehydration solution flushes
- vitamin/mineral/electrolyte supplementation if needed
- travel/emergency plan
Patient monitoring and follow-up. Schedule call-back, follow-up visit to monitor weights/ weight changes, modify/adjust BTF recipe.

Commercial Ready to Use Blenderized Tube Feeding Products – See Table 3

Liquid Hope^TM. Ready to use organic whole foods. The manufacturer recommends switching to Liquid Hope^TM slowly, one meal per day until the transition is complete. The unopened formula is shelf stable for up to 2 years. It has a 3-hour room temperature hold time and unused formula can be refrigerated for 48 hours. The formula can be diluted with water to achieve the desired consistency. Patients can order online ($7.99 per 12-oz serving; available in units of 6, 12, or 24). Medicare approved code for this product is B4149 (blenderized natural foods with intact nutrients). http://www.functionalformularies.com
Real Food Blends^TM. Pre-made blenderized meals, available in 3-meal varieties (Salmon Oats and Squash, Orange Chicken Barley and Carrots, Quinoa Kale and Hemp) and is shelf stable for 18 months. Unused formula should be refrigerated and used within 24 hours. The quinoa meal is slightly thicker and may require additional fluid to flow easily. Patients can order online ($49.95 for 12-pack meals); approved by CMS HCPCS for code B4149 (blenderized natural foods with intact nutrients).
Discover 100% Real Food for Tube Feeding with Real Food Blends
Compleat©. Formulated with real food ingredients: chicken, peas, carrots, tomatoes, and cranberry juice. Available in 250 mL carton and 1 liter closed system. HCPCS code B4149 (blenderized natural foods with intact nutrients). http://www. nestlehealthscience.us/products/compleat

Use of Blenderized Tube Feeding and the New ENFit Connectors

The new enteral connector (ENFit) system is being introduced in phases and is expected to be completed in 2015. GEDSA (Global Enteral Device Supplier Association), Kimberly-Clark, and A.S.P.E.N. (American Society for Parenteral and Enteral Nutrition) recently conducted experiments to determine the pressure required to dispense a BTF through a 60 mL syringe (catheter tip and ENFit) and to check the gravity flow (mL/min) of BTF through the connectors.¹⁰ The BTF was tested just coming out of the refrigerator to simulate clinical conditions. It took about the same pressure to push the BTF through the catheter tip and ENFit syringe, but was actually easier with the ENFit connector. The testing also showed that the gravity flow (mL/min) of BTF for ENFit and catheter tip syringes were essentially equivalent through two gastrostomy tubes (Kimberly-Clark 18fr/20fr PEG with solid bolster and balloon retained G tube feeding systems).

SUMMARY

The process of transitioning to BTF can be an overwhelming task for both the patient and clinician. Homemade blended diets are often described as nutritionally unbalanced, have increased risk for food contamination, and may compromise the enteral access devices. BTF can be used for partial, supplemental, or complete nutrition support. It can be safely used and implemented with the involvement of the medical team and support from the RD to assess, educate, and monitor the patient’s progress toward this feeding option. See Table 4 for additional resources.

DISEASES OF THE PANCREAS, SERIES #9

The Physiology of the Pancreas

Satyanisth Agrawal,

Elie Aoun

Here we discuss the pancreas, a complex organ that plays a critical role in the digestion process. Knowledge of normal pancreatic secretory functions can help clinicians order appropriate tests, which eventually assist in diagnosing specific pathologies. New imaging techniques such as computed tomography (CT) and magnetic resonance imaging (MRI) have enhanced the understanding of pancreas anatomy and made possible the surgical, endoscopic and percutaneous manipulation of pancreas.

INTRODUCTION

When originally studied, the pancreas was one of the last organs in the abdomen to catch critical attention of anatomists and physiologists.¹ New imaging techniques such as computed tomography (CT) and magnetic resonance imaging (MRI) have enhanced the understanding of pancreas anatomy and made possible the surgical, endoscopic and percutaneous manipulation of pancreas.²

Anatomy

The pancreas is a flat, long, and soft gland that is roughly 6 inches in length and weighs between 70 and 110 g. It lies obliquely in the retroperitoneal space of the upper abdomen and is covered by the stomach, transverse colon, and transverse mesocolon. The head of the pancreas lies next to the duodenal sweep at the level of the body of L2.^3,4 The neck, body, and tail span leftward, with the tail extending close to the spleen.⁵

Functional Anatomy

The pancreas consists of ≥80% of acini that are arranged in clusters that form lobules separated by loose connective tissue. A circular shaped acinus and its tubular draining ductule form the functional unit of exocrine pancreas.⁶ Many acini arranged like a bouquet secrete digestive enzymes into the ductule which drains into interlobular ducts and finally into the main pancreatic ductal system.

Pancreatic acinar cells are specialized exocrine secretory cells that synthesize, store, and secrete the digestive enzyme component of the pancreatic juice. An acinar cell is shaped like a triangle, with the basal membrane pointed outward for neurohormonal receptors and the apical membrane located inward forming the lumen of an acinus. The nucleus and rough endoplasmic reticulum (RER) are located near the basal membrane for protein synthesis.7 Zymogen granules that store digestive enzymes are located near the apical membrane and hence close to the lumen. Tight junctions between acinar cells form a barrier between the lumen and apical membrane to prevent inappropriate passage of enzymes but allow water and ions to go through.^8,9 Secretagogues stimulate acinar cells causing the granules to fuse with each other and the apical membrane. Microvilli covering the apical surface of acinar cells facilitate exocytosis of enzymes into the lumen. Gap junctions between adjacent acinar cells allow coordinated chemical and electrical communication between cells for passage of small molecules such as calcium and other ions important for digestive enzyme secretion.

The ducts collect pancreatic enzymes, and the activity of the ductular cells dilutes and alkalinizes pancreatic juice before it is washed out into the small intestine. The duct epithelium is made of cuboidal columnar cells held together by intercellular tight junctions. These ductular cells are packed with mitochondria to supply energy for ion transport. Once stimulated, these cells transport bicarbonate ions into the pancreatic juice as it passes along the duct, with water following in response to the resulting transepithelial osmotic gradient.¹⁰

Exocrine Secretions

Pancreatic exocrine secretions, nearly 2.5L/day in volume, can be classified in two groups: organic and inorganic. Organic secretions are proteins such as digestive enzymes and inorganic secretions consist mostly of water and electrolytes. Acinar cells secrete digestive enzymes and ductal cells secrete bicarbonate rich electrolyte solution.¹¹ Depending on the organic secretion, the enzyme component of pancreatic juice is mixed in various proportions with the aqueous component. Greater than 75% of proteins in organic secretions are enzymes and proenzymes; the rest are plasma proteins, trypsin inhibitors, and mucoproteins (Table 1).

Organic Secretions

One of the major purposes of the pancreas is to synthesize digestive enzymes and deliver them to the intestine where they play a critical role in digestion. The four major enzyme groups are proteolytic (eg, chymotrypsin), amylolytic (eg, amylase), lipolytic (eg, lipase), and nuclease digestive enzymes. Some of the enzymes are present in more than one form (e.g., cationic trypsinogen, anionic trypsinogen, and mesotrypsinogen).¹² To prevent auto digestion of the pancreas, and hence pancreatitis, enzymes are stored and secreted as inactive precursor forms. Enterokinase, secreted by duodenal mucosa, converts trypsinogen to its active form trypsin, which then catalyzes the activation of the other inactive proenzymes. The acinar cells also secrete a trypsin inhibitor, which inactivates trypsin by disabling this catalytic action.¹³

Inorganic Secretions

Pancreatic electrolytes (sodium, potassium, chloride, and bicarbonate) mixed with water form an alkaline fluid that is isosmotic with extracellular fluid and helps neutralize gastric acid entering the duodenum.¹⁴ Postprandial stimulation mediated mainly by secretin increases secretory flow rate from an average of 0.3 mL/minute in the resting (interdigestive) state to 4.0 mL/minute in the digestive state. The concentrations of bicarbonate and chloride in pancreatic juice change reciprocally as secretory flow rate increases making the osmolality of pancreatic juice independent of flow rate.¹⁵

Digestive Enzyme Functions

Amylase The salivary glands and pancreas make amylases. Pancreatic amylase hydrolyzes the 1,4-glycoside linkages of complex carbohydrates and starches. This produces short dextrins, which can then be digested by brush border enzymes like maltose and maltotriose into glucose.

Lipases The majority of dietary lipids in western diets are triglycerides, which cannot be digested by brush border enzymes. Pancreatic triglyceride lipase binds to the oil-water interface of the triglyceride oil droplet where it cleaves the majority of fatty acids from dietary triglycerides.

Bile acids and colipase are important for the full activity of lipases. Bile acids emulsify triglyceride molecules to expand surface area for lipase to act on.¹⁶ Colipase forms a complex with lipase and bile salts and anchors lipase to allow it to act in a more hydrophilic environment on the hydrophobic surface of the oil droplet. Carboxyl ester lipase can act on a variety of substrates and is important in digestion of cholesterol esters, lipid-soluble vitamins such as Vitamin A, and triglycerides.

Proteases Pancreatic proteases and gastric pepsin digest all of the complex dietary proteins into short peptides and amino acids for further digestion and absorption in the intestine. The most abundant enzyme is trypsin, which is present in three forms. Cationic trypsinogen, coded by PRSS1 gene, is present in a large proportion, and anionic trypsinogen and mesotrypsinogen, which are coded by PRSS2 and PRSS3 genes, respectively, are present in smaller proportions. All trypsinogens act similarly by attacking the exposed arginine and lysine residues within a peptide chain. Chymotrypsin and elastase are endopeptidases, just like trypsin, that cleave specific peptide bonds adjacent to specific amino acids. These amino acids eventually have greater effects on stimulating pancreatic secretion, inhibiting gastric emptying, regulating small bowel motility, and causing satiety.

Synthesis and Transport of Digestive Enzymes

Protein synthesis occurs in the ribosomes close to the rough endoplasmic reticulum (RER) of acinar cells.¹⁷ The cell’s messenger RNA (mRNA) then translates these newly synthesized proteins into exportable proteins. A terminal peptide extension on pancreatic enzymes, known as signal protein, allows attachment and entry of the enzyme into the RER.¹⁸ The enzyme and signal protein interact with a membrane protein called a docking protein. This process permits the completion of the translocation, dissociation of the signal protein and mRNA from enzyme, and allows the enzyme to enter RER. Newly synthesized proteins can undergo modifications and conformational changes in the endoplasmic reticulum before being transported to the Golgi complex where further post-translational modification (glycosylation), sorting, and concentration occur.¹⁹

Digestive enzymes are then transported to the zymogen granules. A given zymogen granule has various pancreatic proteins mixed in relative proportions depending on their rates of synthesis. The rate of synthesis of a particular enzyme is related to the type of diet. For example, dietary increase in carbohydrates will result in increased expression of amylase compared to other pancreatic enzymes. Zymogen granules then move towards the apical membrane of acinar cells via microtubules and await appropriate neurohormonal stimulus to trigger exocytosis.²⁰

Facilitated by microvilli covering apical surface of acinar cells, exocytosis is a process where the zymogen granule fuses with the apical surface and allows its contents to be released in the ducts.²¹ This entire process takes less than 1 hour allowing the pancreas to be ready for the next meal by repeating synthesis and packaging of enzymes.

Cellular Regulation of Enzyme Secretion

At the cellular level, secretion of pancreatic juice can be divided into organic and inorganic secretions. Organic secretions containing pancreatic enzymes occurs by regulating acinar cells, and inorganic secretions containing bicarbonate and other electrolytes occurs by regulating ductal cells. Hormonal regulation of acinar and ductal cells is explained in this section and the integrated neurohormonal control of pancreatic secretion is discussed later.

Acinar Cells

Acinar cells express receptors on their basolateral membranes for the following hormones: cholecystokinin (CCK), acetylcholine (ACh), gastrin-releasing peptide (GRP), vasoactive intestinal peptide (VIP), and secretin.^22,23 These receptors are divided into two groups based on their mode of stimulating acinar cells (Figure 1). VIP and secretin activate adenylate cyclase, which increases cellular cAMP and facilitates enzyme secretion through cAMP dependent protein kinase A. The other group consisting of acetylcholine, GRP, and CCK lead to an increase in intracellular free calcium concentrations via stimulating cellular metabolism of membrane phosphoinositides.²⁴ This phospholipase C-dependent pathway is the primary stimulus for significant acinar secretion, with cAMP-dependent signaling playing a secondary role.

Ductular Cells

Ductular cells contribute the fluid and bicarbonate components of pancreatic juice. Bicarboanate is predominantly derived from plasma rather than intracellular metabolism. Both the apical and basolateral membranes have polarized epithelial cells and membrane transport proteins that help with ion transportation (Figure 2). Ductal cells are very sensitive to secretin and VIP, both of which increase intracellular cAMP, which in turn leads to opening of CFTR chloride channels initiating secretion.^14,15 Bicarbonate enters through the sodium-bicarbonate cotransporter on the basolateral membrane and exits through the CFTR channel on the apical membrane.²⁵ Concomitantly, the sodium-potassium pump keeps the intracellular sodium low thus creating a continual electrochemical force and driving bicarbonate into the ductal cell through the sodium-bicarbonate cotransporter. Water and sodium ions follow paracellularly in response to the electrochemical gradient across the epithelium.

Organ Physiology

Exocrine pancreatic secretion happens during two states: fasting (interdigestive) and after ingestion of a meal (digestive). The interdigestive pattern of secretion begins when the stomach is empty. Secretory activity related to eating (digestive state) occurs in phases: cephalic (20-25%), gastric (10%), and intestinal (approximately 60%-70%). Pancreatic secretion is activated by a combination of neural and hormonal effectors.

Interdigestive Secretions

The interdigestive pancreatic secretions are governed by the cholinergic nervous system, motilin, and pancreatic polypeptide. Secretions follow the cyclical pattern of the migrating myoelectric complex (MMC) [26, 27]. Enzyme secretion occurs every 1 to 2 hours and is associated with the periods of increased motor activity in the stomach and duodenum. In addition to pancreatic enzyme secretion, there is increased secretion of bicarbonate and bile (secondary to partial gallbladder contraction) into the duodenum. The pancreatic secretion during the interdigestive phase is integral to the “housekeeping” function of the MMC to clear the stomach and small intestine of debris including bacteria between meals.²⁶

Digestive Secretions

Secretion with ingestion of a meal is divided into three phases: cephalic, gastric, and intestinal. During the cephalic and gastric phases, secretions are low in volume with high concentrations of digestive enzymes, reflecting stimulation primarily of acinar cells. This stimulation arises from cholinergic vagal input during the cephalic phase, and vago-vagal reflexes activated by gastric distension during the gastric phase. Gastric distention predominantly causes secretion of enzymes with little secretion of water and bicarbonate.²⁸

The intestinal phase begins when chyme leaves stomach and enters the small intestine. During the intestinal phase, ductular secretion is strongly activated, resulting in the production of high volumes of pancreatic juice with decreased concentrations of protein, although the total quantity of enzymes secreted during this phase is actually also markedly increased. Ductular secretion during this phase is driven primarily by the endocrine action of secretin on receptors localized to the basolateral pole of duct epithelial cells. The inputs to the acinar cells during the intestinal phase include CCK as well as neurotransmitters including acetylcholine (ACh), GRP, and VIP.^29-31 The large magnitude of the intestinal phase is also attributable to amplification by so-called enteropancreatic reflexes transmitted via the enteric nervous system.

CCK is released from the upper small intestinal mucosa by digestion products of fat, protein, and starch. CCK is a potent stimulus of acinar secretion, acting both directly on CCK-B receptors localized to the basolateral membranes of acinar cells (Figure 1), and via stimulation of vagal afferents close to its site of release in the duodenum, thereby evoking vago-vagal reflexes that stimulate acinar cell secretion via cholinergic and noncholinergic neurotransmitters (the latter including both GRP and VIP).³² In addition to its effects on the pancreas, CCK coordinates the activity of other GI segments and draining organs, including contraction of the gallbladder, relaxation of the sphincter of Oddi, and the slowing of gastric motility to retard gastric emptying and thereby control the rate of delivery of partially digested nutrients to more distal segments of the gut.³³ Finally, CCK modulates the activity of secretin in a synergistic fashion by markedly potentiating the effect of secretin as an agonist of pancreatic ductular secretion of bicarbonate.³¹

The other major regulator of pancreatic secretion is secretin, which is released from S cells in the duodenal mucosa by gastric acid, with a pH threshold of 4.5.³⁴ When the meal enters the small intestine from the stomach, the volume of pancreatic secretions increases rapidly, shifting from a low-volume, protein-rich fluid to a high volume secretion in which enzymes are present at lower concentrations (although in greater absolute amounts, reflecting the effect of CCK and neural effectors on acinar cell secretion). As the secretory rate rises, the pH and bicarbonate concentration in the pancreatic juice rises, with a reciprocal fall in the concentration of chloride ions.³⁴ These latter effects on the composition of the pancreatic juice are mediated predominantly by the endocrine mediator, secretin.³⁵ After the meal, trypsin is free and inhibits intestinal CCK release and pancreatic enzyme secretion. This process is known as feedback inhibition and this effect of trypsin is mediated by intraluminal CCK-releasing factors (CCK-RF) present on the intestinal epithelium.³⁶ In the presence of a meal the digestive proteases are occupied and CCK-RF promotes CCK release and more digestive enzyme secretion. However, after digestion of a meal when there is an excess of digestive proteases in the intestinal lumen, CCK-RF is in turn digested and inactivated so that its ability to augment CCK release and stimulate further pancreatic enzyme secretion ceases (Table 2).

SUMMARY

The pancreas is a complex organ that plays a critical role in the digestion process. Understanding the cellular physiology of acinar and ductal cells lends to grasping the concepts of pancreatic exocrine secretions in various phases of digestion. Knowledge of normal pancreatic secretory functions can help clinicians order appropriate tests, which eventually assist in diagnosing specific pathologies.

GASTROINTESTINAL MOTILITY AND FUNCTIONAL BOWEL DISORDERS SERIES #4

Cannabis in Gastrointestinal Disorders

Mohammad Bashashati,

Richard W. McCallum

For thousands of years, cannabis and its derivatives have been used for the treatment of human diseases including those that present with gastrointestinal (GI) symptoms. Despite the strong evidence supporting the therapeutic role of cannabis in nausea and vomiting related to chemotherapy and cachexia of AIDS, studies on the use of cannabis for other GI disorders are limited and sparse. In this article, we review available clinical evidence in supporting medical cannabis for GI diseases.

Mohammad Bashashati, MD1 Richard W. McCallum, MD2 1Research Associate, Snyder Institute for Chronic Diseases and Hotchkiss Brain Institute, University of Calgary, Calgary, AB, Canada 2Professor and Founding Chair, Department of Internal Medicine, Texas Tech University Health Sciences Center, El Paso, TX

INTRODUCTION

With over 5000 years of use, cannabis or marijuana, i.e. dried leaves, stems, flowers, and seeds of the plant Cannabis sativa and its compressed resin form, i.e. hashish are the most common consumed illicit drugs worldwide and in the United States.^1,2

The chemical derivatives of Cannabis sativa are called “phytocannabinoids”, while there are also synthetic cannabinoids which have cannabinomimetic effects.³

Besides being used recreationally, cannabis and its derivatives have been valuable for the treatment of human diseases for centuries. The term “medical cannabis” or “medical marijuana” describes medicinal use of cannabis or cannabinoids.⁴ However, based on their potential for being abused, it is very important to define a border between medical and recreational cannabis.⁵

Under federal law of the United States, cannabis is a Schedule I substance and its use is illegal; however, the District of Columbia and 23 states have legalized medical cannabis. Moreover, in some states recreational use of marijuana is legal and possession of limited amounts of marijuana has been decriminalized. Medical cannabis is approved for chronic, debilitating and long- lasting disorders such as cancer, AIDS and multiple sclerosis and occasionally, this is legally allowed for gastrointestinal (GI) patients with Hepatitis C, Crohn’s disease, nausea and cachexia.⁶

Due to the legalization of medical cannabis in several states, physicians who see patients with GI diseases may need to prescribe cannabis. Alternatively, they might be asked about the benefits and adverse effects of medical marijuana. In addition, due to decriminalization of cannabis in some states, a practicing physician may see more patients who use cannabis recreationally. Therefore, it is important for a physician to know the therapeutic potentials of cannabis and its natural or synthetic derivatives and the possible adverse effects of cannabis use.

This article highlights the benefits of medical cannabis in GI diseases and discusses possible side effects of medical cannabis and cannabis intoxication.

Cannabis and Cannabinoids

D9-tetrahydrocannabinol (THC) and cannabidiol (CBD) are two major components of cannabis. While THC has psychoactive effects, CBD is devoid of central side effects.⁷ For medicinal purposes, cannabis can be smoked or vaporized or its extracts/juice can be used orally; however, if taken orally, its absorption is slow and its therapeutic components are less bio-available. More reliable forms of medicinal cannabis in terms of dosage are synthetic compounds which are structurally related to THC, e.g. Nabilone (Cesamet) and Dronabinol (Marinol), as well as accurately measured mixture of THC and CBD extracted from the cannabis plant, e.g. Nabiximols (Sativex). Nabilone and Dronabinol are used orally; however, Nabiximols is an oromucosal spray.⁸ Besides these, there are synthetic cannabinoid agonists which have not yet been used in the clinic, but have shown significant therapeutic potential in pre-clinical studies as well as the potential for being abused.⁹

The discovery of endocannabinoid system in 1990s shed light into our understanding of the mechanisms of cannabis function in the body. Moreover, it introduced new potential therapeutic strategies through modulating endocannabinoid system.¹⁰ Such an approach is still under clinical evaluation for different clinical conditions and hopefully their results will be released in the early future.

Endocannabinoid System

Basically, our body has endogenous cannabinoids, which are present both in periphery and the central nervous system (CNS). Two main endogenous cannabinoids are anandamide (AEA) and 2-arachydonilglycerol (2-AG). AEA and 2-AG are synthesized on demand from the membrane phospholipids and activate presynaptic cannabinoid receptors 1 and 2 (CB₁ and CB₂). Endogenous cannabinoids, the enzymes that are involved in the biosynthesis and degradation of them besides CB₁ and CB₂ receptors are the components of the “endocannabinoid system”.^11,12

N-arachidonoylphosphatidylethanolamine phospholipase D (NAPE-PLD) and diacylglycerol lipase (DAGL) are the major enzymes in the biosynthesis of AEA and 2-AG, respectively. On the other hand, AEA is degraded by fatty acid amide hydrolase (FAAH), while monoacylglycerol lipase (MAGL) hydrolyses 2-AG. In addition to these, cyclooxygenase 2 (COX-2) is an important enzyme which metabolizes both AEA and 2-AG and produces prostaglandins.^11-13

As mentioned, AEA and 2-AG activate CB₁ and CB₂ receptors. Another potential receptor for AEA is the transient receptor potential vanilloid type-1 (TRPV1). CB₁ and, to some extent, CB₂ receptors are present in the central and enteric nervous system. On the other hand, CB₂ receptors are more abundant on immune cells, making them a good drug target during inflammation. Despite their variable distribution in different organs, the function of CB₁ and CB₂ receptors are not distinct and both can either regulate neuronal signaling or might be involved in immune mediated mechanisms.^12,14

Endocannabinoid system is present in the GI tract. Although its components have been observed in almost all layers of intestinal sections, they are majorly localized to the enteric nervous system including the myenteric and submucosal plexi. Moreover, there is some evidence to support the presence of CB₁ and CB₂ receptors on the epithelial cells.10,12 The localization of CB₁ and CB₂ receptors potentially defines the pharmacophysiology of cannabis and cannabinoids in the GI tract making them a potential target for the treatment of GI disorders such as abdominal pain, diarrhea, nausea and vomiting as well as GI inflammation.

Figure 1 shows the distribution of cannabinoid receptors in the GI tract and CNS regions that control GI function.

The Physiology of Endocannabinoid System in the GI Tract

Our understanding of the physiology of cannabinoids and endocannabinoids relies heavily on findings from the animal studies. In brief, cannabinoids are usually inhibitory in the GI tract through inhibiting vagal pathway.^10,12,15 Here, we have summarized the major functions of endocannabinoid system in the GI tract:

Cannabinoids are anti-nociceptive through both CB₁ and CB₂ receptors. Moreover, enhanced endocannabinoid tone decreases visceral pain in animal models.^16,17
Cannabinoids inhibit gastrointestinal motility through CB₁ receptors in physiologic conditions and CB₁/ CB₂ receptors during inflammation.¹⁸ Cannabinoids inhibit transient lower esophageal sphincter relaxation,¹⁹ delay gastric emptying²⁰ and inhibit intestinal transit or contractility.¹⁸
Phytocannabinoids and exogenous cannabinoid agonists inhibit the disordered intestinal permeability during inflammation or after exposure to cholera toxin.²¹ Moreover, activating CB₁ receptors reduces acid secretion in the stomach.²²
Cannabinoids induce hyperphagia and increase appetite resulting in weight gain. In addition, cannabinoids including both THC and CBD reduce nausea and vomiting through interacting with CB₁, CB₂ and TRPV1 receptors as well as through possible interaction with 5-hydroxytryptaminergic (5-HT; serotoninergic) system.^12,23
Cannabinoids are anti-inflammatory, and this makes them a potential candidate for the treatment of colitis.²⁴
Cannabinoids induce apoptosis, are antiproliferative and anti-cancerous.²⁵
In the liver, the expression of cannabinoid receptors is low. However, the activation of CB₁ receptors is profibrogenic, proinflammatory and pro-steatotic. On the other hand, CB₂ receptors inhibit hepatic fibrogenesis.²⁶

Understanding the physiology of endocannabinoid system in the GI tract helps us to define the therapeutic potentials of cannabis and its derivatives in different GI diseases.

Cannabinoids and Gastrointestinal Disorders

While many pre-clinical studies have proved the benefits of cannabinoids and endocannabinoids in GI diseases, clinical studies targeting endocannabinoid system are restricted. This is basically because of the psychoactive effects of these compounds as well as the legal restrictions.

Cannabinoids and Nausea/Vomiting

Conventional antiemetics such as 5-HT₃ and NK1 antagonists are helpful in reducing episodes of vomiting; however, they are not usually helpful in reducing the unpleasant sensation of nausea. In contrast, cannabinoids can inhibit both nausea and vomiting. This has been proved in different clinical trials on chemotherapy induced nausea and vomiting (CINV) by using synthetic cannabinoid agonists such as dronabinol, nabilone and levonantradol as well as with THC/CBD mixture (i.e. Sativex). In an earlier study back in 1985, nabilone was compared to prochlorprazine for the treatment of chemotherapy induced emesis. Based on this study, nabilone was superior to prochlorpazine in reducing vomiting episodes.²⁷ Moreover, a recent study that compared dronabinol and ondansetron for delayed CINV showed these medications are similarly effective and their combination is not superior.²⁸ Despite these benefits, due to central side effects, cannabinoid based therapy is not considered the first line treatment in patients with CINV.^8,23

Studies on the effects of smoked cannabis and nausea/vomiting are limited. In a study which compared a group of patients who smoked marijuana before/after chemotherapy and another group who had used THC capsule orally, the former group experienced 70-100% symptom relief, while symptom relief in those who had used THC was 76-88%, suggesting a favorable therapeutic role for smoked cannabis in CINV.29 In another study, SÖderpalm et al. tested the effects of smoked cannabis compared to ondansetron in controlling Ipecac induced nausea and vomiting. Smoked cannabis reduced both nausea and vomiting; however, its effects were modest compared to ondansetron.³⁰

A side effect of long-term (generally>5 years) daily marijuana smoking is cannabinoid-induced hyperemesis syndrome, which is particularly observed in male patients and presents with repeated cyclical vomiting and frequent hot bathing. The symptoms usually alleviate after cessation of cannabis smoking. Not all cannabis users develop hyperemesis syndrome. Therefore, the total (cumulative) dose of marijuana, genetic factors, and psychological parameters may contribute to this condition. The pathophysiology of cannabinoid-induced hyperemesis syndrome remains unknown. The few hypotheses which may explain this syndrome are: (a) accumulation of the cannabis derivatives in the brain based on their lipid solubility and long-term half-life, (b) degradation of the cannabis ingredients to some potential emetic metabolites or toxins, (c) delayed gastric emptying induced by cannabis and (d) down-regulation or desensitization of the cannabinoid receptors due to chronic cannabis use.^31,32

Cyclic vomiting syndrome (CVS) is a functional GI disorder which presents with nausea and vomiting and epigastric abdominal pain. CVS presents with stereotypical and recurrent episodes of vomiting and abdominal pain with relatively nausea and vomiting free intervals. The pathophysiology of CVS is unknown.³² Studies from our group have shown a high rate of cannabis use in a subset of patients with cyclic vomiting syndrome. These patients needed a higher dose of amitriptyline for the control of their CVS attacks compared to non-cannabis users.^33,34 Therefore, it is important to identify cannabinoid induced hyperemesis syndrome as part of the CVS, since a long term goal is decreasing and stopping cannabis use in these patients.

The acute treatment regimen for vomiting attacks in both CVS and cannabinoid-induced hyperemesis syndrome are based on intravenous hydration, intravenous lorazepam due to its sedative effects, anti- emetics and non-narcotic pain medications. Tricyclic antidepressants (TCA) particularly amitriptyline in high doses are recommended for the long-term prophylaxis. Gradual reduction and stopping cannabis use, psychological interventions, anti-anxiety medications and relaxation techniques are necessary in these patients.³²

Overall, despite their favorable effects in controlling nausea, cannabinoids are not the first line treatment of nausea and vomiting including CINV as their chronic use may also induce hyperemesis syndrome. Their anti-emetic effects are modest compared to other conventional medications. Moreover, their side effects make them unfavorable treatment for nausea and vomiting. On the other hand, medical cannabis including dronabinol which is available in the United States and has antiemetic effects at 5-10 mg per dose three times a day may be effective in a subset of patients with gastroparesis who present with nausea and vomiting. It is important to recognize that prescribing cannabis based medications is very different from chronic daily cannabis smoking, which can lead to the entity termed cannabinoid-induced hyperemesis syndrome. Whether manipulating endocannabinoid levels is effective in controlling nausea and vomiting is the goal of future studies in this field.

Cannabinoids and Appetite

Cannabis stimulates appetite and increases food intake. Therefore, cannabis-derivatives (e.g. dronabinol) or smoking cannabis are potential treatments for patients with AIDS-associated loss of appetite and cachexia.^6,8 On the other hand, to our knowledge, no trial has tested the effects of smoked cannabis in inducing appetite or increasing weight among patients with cancer. Moreover, the data on oral THC as the stimulator of appetite in patients with cancer are not conclusive. Interestingly, recent studies have questioned the general belief of higher rate of obesity in cannabis smokers. Based on two large cohorts including 43,093 and 9282 respondents among US adults aged 18 years or older (2001-2003), the adjusted prevalence of obesity was about 8% lower in participants reporting the use of cannabis at least 3 days per week.³⁵ Therefore, although cannabis may stimulate appetite, its effect on weight gain is not well confirmed. Delayed gastric emptying²⁰ or potential tolerance to chronic cannabis use³⁶ may explain lower weight in chronic cannabis users.

Cannabinoids and IBS

IBS is a functional GI disorder presenting with abdominal pain/discomfort and disturbed bowel habits. The pathophysiology of IBS is not well understood and its treatment is underdeveloped. In IBS endocannabinoid signaling is altered compared to normal population.³⁷ Moreover, cannabinoids can affect both GI motility and visceral sensation.¹² Therefore, they are potential candidates for the treatment of IBS especially when it presents with diarrhea.

Studies on the effect of smoked cannabis in IBS are lacking. Based on a clinical trial, dronabinol (5mg) increased colonic compliance and decreased colonic contraction in IBS patients with diarrhea or alternating bowel habits. These responses were varied based on polymorphisms of FAAH and CB₁ encoding genes. In this study, dronabinol did not change sensation scores for pain and gas.³⁸ In another study from the same group, treatment with dronabinol did not alter gut transit in IBS-D, but tended to decrease colonic transit in subjects with a specific polymorphism at CB₁ encoding gene.³⁹Another study, which tested the effects of dronabinol (up to 10mg) on visceral perception to rectal distension in IBS vs. health controls, showed dronabinol does not affect baseline and stimulated rectal perception.⁴⁰

Overall the effects of THC as an effective component of cannabis on GI motility in IBS are variable and its effect on abdominal pain and visceral sensation is not promising. Whether manipulating endocannabinoid system is effective in IBS needs further investigations.

Cannabinoids and IBD

Although there is not a well designed case-control study, smoking cannabis looks to be common among patients with inflammatory bowel disease (IBD). Based on the available studies, 39-49% of patients were past/lifetime and 12-14% of them were current marijuana users,^41,42 and 18-32% of IBD patients have used marijuana for their IBD symptoms.^41,43 Smoking cannabis was more common among patients with Crohn’s disease and a better response to cannabis was observed in these patients.⁴² On the other hand, long-term cannabis use was correlated with the risk of surgery in patients with Crohn’s disease,⁴³ questioning the therapeutic benefits of cannabis in IBD. Recently, a clinical trial on a small number of patients with Crohn’s disease showed that 8 weeks of smoking cannabis (standardized to 115mg of THC; twice daily) decreased Crohn’s disease activity index (CDAI) more significantly compared to placebo; however, after 2 weeks wash-out period, CDAI was not different in both groups suggesting a temporary but not sustained benefit for smoking cannabis in these patients. C-reactive protein levels did not change after smoking cannabis and endoscopic findings were not collected.⁴⁴ Although these findings are somehow promising, the temporary effect of cannabis as well as the increased risk of surgery in patients with Crohn’s disease who were long-term users of cannabis questions whether medical cannabis is effective in IBD as advertised in the media and among patients and physicians.

At the moment, we should be cautious in prescribing cannabis for patients with IBD. Larger studies with cannabis or synthetic cannabinoids as well as the modulators of endocannabinoid levels (e.g. FAAH inhibitors) are recommended.

Cannabinoids as Anti-Nocieceptive Agents

As mentioned above, the effect of THC on abdominal pain and visceral sensation in IBS patients is not promising. However, interestingly a recent double- blind controlled trial showed that dronabinol (5mg b.i.d. for 4 weeks) was superior to placebo in increasing pain threshold and decreasing the frequency and intensity of non-reflux related non-cardiac chest pain. No significant adverse effects were noted in this study.⁴⁵

Therefore, the origin of pain may define the therapeutic efficacy of cannabinoid based medicine in GI diseases with pain. More studies with different regimen are necessary to evaluate the effects of cannabis and its derivatives on visceral sensation and pain.

Cannabinoids and Hepatitis C

Using cannabis as a street drug is common among people with or without hepatitis C. There is a general belief among patients with hepatitis C that cannabis helps with the adverse effects of their medications. This has been shown in studies indicating less adverse effects of interferon and the adherence to HCV treatment in cannabis users or those who were receiving oral cannabinoid-containing medications.^46,47 On the other hand, a recent study has shown no association between cannabis use and the likelihood of completing a full course of HCV therapy, interruption of therapy or sustained virological response.⁴⁸ As the newer and much more effective HCV therapeutic regimens which lack interferon are introduced, the side effects of interferon therapy (e.g. nausea, cachexia, depression) are disappearing and therefore, there will not be a significant indication for the medical cannabis in HCV just based on its potentials in alleviating therapeutic side effects in early future.

Studies on the effects of cannabis and its effect on the progression of liver damage are not conclusive. While some studies have shown that daily cannabis use is associated with liver fibrosis and steatosis in patients with hepatitis C,^49-50 others have shown no association between the progression of liver damage and smoking cannabis.^48-52

Adverse Effects of Cannabis Use

The disorders related to cannabis are majorly classified as: (1) cannabis intoxication, (2) cannabis use disorder, and (3) cannabis withdrawal.^53-55 When recreational cannabis is allowed and medical cannabis is legalized in different states, practicing physicians may see more patients presenting with adverse effects of cannabis use. Many of the adverse effects are self-limited and mild, although psychological interventions, replacement therapies and symptomatic treatment might be necessary.

Cannabis intoxication presents with physiologic and psychiatric presentations of smoking cannabis. Briefly, patients present with anxiety, panic and psychotic symptoms, red eye, tachycardia, dry mouth and increased appetite.^53,54

Cannabis use disorder includes daily or near daily heavy or regular cannabis use for a long duration which may present with cannabis dependence syndrome (which is majorly psychological) in around 1 in 10 users, tolerance as what possibly occurs in cannabis hyperemesis syndrome and disturbed personal and social function. Chronic cannabis smoking may cause chronic bronchitis, psychotic symptoms especially in those with other risk factors of these disorders, cognitive impairment in long-term users and after abstinence and impaired educational attainment among adolescents.^53,54

Cannabis withdrawal presents with psychiatric symptoms such as insomnia, anxiety, depression, appetite disturbance and physical symptoms of impaired cannabis signaling such as abdominal pain, headache, tremor, and restlessness. Most symptoms usually occur on the first day after abstinence, peak on day 2-6 and are abolished within 14 days.⁵⁵

Another big concern regarding cannabis abuse is related to synthetic cannabinoids. These drugs, which are predominantly full and potent agonists of CB₁ receptors, may induce dangerous side effects such as seizure, hallucination, tachycardia and arrhythmia. They may also affect non-cannabinoid pathways and produce severe intoxication.⁹ Therefore, we should also be aware of the clinical manifestations of synthetic cannabinoid abuse.

Cannabinoid Receptors Antagonists

Reducing the endocannabiod tone enhances GI motility and decreases the appetite. Therefore, while the purpose of medical cannabis is the induction of endocannabinoid response, reducing endocannabinoid tone with the cannabinoid receptors antagonists is a potential treatment for constipation and obesity. The most famous compound in this area is the CB₁ inverse agonist Rimonabant, which was approved in Europe as an anti-obesity medication. This drug was withdrawn from the market based on its serious side effects such as depression and increased suicidal tendency.⁵⁶

With the invention of peripherally restricted cannabinoid receptors antagonists, we may see other substitutes for Rimonabant with clinical potentials in the future.

RECOMMENDATIONS AND CONCLUSIONS

The pharmacological function of cannabinoid agonists and antagonists may define the potential therapeutic indication of these compounds. These have been summarized in Table 1.

Medical cannabis is effective in patients with chemotherapy induced nausea and vomiting and in cachexia associated with AIDS; however, the evidence to support the role of medical cannabis in other GI diseases is poor. In some cases, cannabis use may associate with the progression of the disease (e.g. hepatitis C) or poor outcome (e.g. risk of surgery in Crohn’s disease). Despite the misguided advertisements for cannabis as a therapy for different GI diseases, we should be cautious in prescribing medical cannabis and should communicate with our patients regarding the adverse effects as well as the limitations of current studies in the field. This may change by larger and well- controlled studies in future.

For patients with CVS who are active cannabis users and patients with cannabinoid-induced hyperemesis syndrome, we recommend gradual decrease and stopping cannabis. During vomiting episodes, lorazepam plus anti-emetics are helpful and for the prophylaxis of vomiting attacks, high-doses of TCAs are recommended.

We believe the compounds that selectively target endocannabinoid metabolism and degradation are the potential medications of future for human diseases including GI disorders.

FRONTIERS IN ENDOSCOPY SERIES # 14

Office Management of Hemorrhoid Disease

Andy Tau,

Waqar Qureshi

Hemorrhoids account for approximately 3.5 million physician visits and 500 million dollars in costs per annum in the US. The gastroenterologist maintains a unique position between the conservative treatment offered by the primary care physician and the more invasive hemorroidectomy offered by the surgeon. Here we discuss the etiology, classification and diagnosis of hemorrhoids and the procedures for treatment.

INTRODUCTION

Symptomatic hemorrhoids are frequently encountered by gastroenterologists in Western societies. Internal hemorrhoids are the most common cause of recurrent hematochezia in ambulatory adults. Hemorrhoids account for approximately 3.5 million physician visits and 500 million dollars in costs per annum in the USA.¹ The gastroenterologist maintains a unique position between the conservative treatment offered by the primary care physician and the more invasive hemorroidectomy offered by the surgeon.

Anatomy

The dentate line is a key anatomical landmark that divides the insensitive rectal mucosa from the richly innervated anal skin. The dentate line dictates classification, sensation, and therapy of hemorrhoids. Internal hemorrhoids are cushions of fibrovascular tissue located proximal to the dentate line. They may prolapse and bleed, but are usually painless. External hemorrhoids are distal to the dentate line and are covered by richly innervated squamous epithelium of the anus. Any therapy whether it be excision or ligation must respect the dentate line if done without anesthesia. Three major cushions reside in the left lateral, right anterior, and right posterior positions. These positions are relatively fixed and allow banding “blind” in certain office techniques. The blood vessels within these cushions are sinusoids that have direct arteriovenous communications between branches of the superior and middle hemorrhoidal arteries, and the superior, rectal continence. Hemorrhoidal cushions contribute to 15-20% of the resting pressure of the anal verge and provide continence by forming complete closure of the anal canal.²

Etiology & Pathophysiology

Risk factors for symptomatic hemorrhoids include low dietary fiber, chronic straining, excessive time on the commode, constipation, diarrhea, pregnancy, and family history.³ The underlying pathophysiology likely involves a multifactorial process involving venous dilation, arterio-venous distention, protrusion of congested anal cushions downward and progressive stretching and collapse of the support structure of the cushions overtime.⁴ This culminates in prolapse beyond the anal verge. Internal hemorrhoids that remain prolapsed develop ischemia, thrombosis, or gangrene. It is only in this setting that internal hemorrhoids become painful or pruritic. More often encountered is painless bleeding which occurs when the submucosal sinusoids are disrupted. The bleeding is bright red from the pre- sinusoidal arterioles.⁵

Classification

Hemorrhoids are classified, first, as internal or external relative to the dentate line. (Figure 1) Hemorrhoids above the dentate line are internal hemorrhoids. Hemorrhoids below the dentate line are external hemorrhoids. External hemorrhoids are covered by the very sensitive anoderm. Internal hemorrhoids are further classified by the degree or prolapse, which has direct therapeutic and prognostic consequences:

Grade I Do not prolapse below the dentate line; visible only on anoscopy
Grade II Prolapse below the dentate line, but spontaneously reduce
Grade III Prolapse below the dentate line, but require manual reduction
Grade IV Prolapse and stay below the dentate line – not reducible

Nearly all patients with symptomatic Grade I internal hemorrhoids respond well to medical therapy. Grade II and small Grade III internal hemorrhoids respond to non-operative therapy. Large, refractory grade III and grade IV internal hemorrhoids often require surgery.

Clinical Manifestation and Evaluation

Symptoms attributed to hemorrhoids include bleeding, itching and pain.⁶ Internal hemorrhoids may prolapse and bleed, but are only painful when they have thrombosed, owing to their position proximal to the dentate line. Bleeding is bright red, owing to arterial bleeding from disruption of the arterial-venous connections of the sinusoids. Drops of bright red blood at the end of a bowel movement or on tissue paper, or blood that coats solid brown stool are suggestive features, but are not reliably predictive.⁷ External hemorrhoids can become very painful when acutely thrombosed.

Three Myths

Three common misconceptions should be dispelled about hemorrhoids. One is the association with portal hypertension. Portal hypertension produces rectal varicies rather than hemorrhoids. In fact, patients with portal hypertension are no more at risk for hemorrhoids.⁸ The second misconception is that hemorrhoids alone can cause a positive result on stool guaiac tests. This is not true.⁹ Fecal occult blood should not be attributed to hemorrhoids until the full colon is adequately evaluated. Lastly, the misconception that bleeding from hemorrhoids causes anemia.¹⁰ It cannot be overstated that patients who present with anemia require further investigation of the gastrointestinal tract.¹¹ All patients older than 50 years or at high risk for colorectal cancer presenting with rectal bleeding or anorectal symptoms should undergo a full colonoscopy before focusing on internal hemorrhoids as the culprit.

Diagnosis

While many patients with anorectal complaints will ascribe their symptoms to “hemorrhoids”, a careful history and examination will often reveal other anorectal pathology. Careful external inspection of the anus and perirectal area may reveal thrombosed external hemorrhoids, but also anal fissures, fistula, condylomata, rectal abscess, locally advanced rectal cancer, proctitis or pruritus ani. Meanwhile painless bleeding will often require an exam complemented by flexible sigmoidoscopy, colonoscopy or anoscopy. Digital rectal examination is usually unable to detect internal hemorrhoids unless prolapsing or thrombosed. Besides internal hemorrhoids, painless rectal bleeding can also be caused by colorectal carcinoma, polyps, anal fistula, rectal varices, Kaposi sarcoma, and telangiectasias. A painful digital rectal examination should raise the suspicion of concomitant anal fissure which is not uncommon.

Medical Treatment

Conservative treatment begins with the universal recommendation to add fiber to the diet, increased fluid intake, and avoid prolonged straining or sitting on the commode. Prospective studies have shown improvement in pain and bleeding from hemorrhoids with dietary fiber supplementation, though the evidence has not been overwhelming.^12,13,14,15 Over the counter sitz baths, suppositories, and topical analgesics and corticosteroids provide symptomatic relief, but no evidence supports their use in reducing actual hemorrhoid swelling, bleeding or prolapse. Long term use of corticosteroid creams is harmful and should be discouraged. Grade I and II hemorrhoids have a decent chance of resolution with medical therapy alone. Taking all comers, the majority of patients presenting with symptomatic hemorrhoids improve with a bowel management program alone.¹⁶

Non-Surgical Procedures

When these conservative measures fail, non-surgical procedures are recommended, which is particularly the case with Grade II and small grade III hemorrhoids. It is here where the gastroenterologist can offer an alternative before invasive surgical excision. All of these procedures affix the hemorrhoidal tissue back onto the muscle wall and do not require anesthesia. While they are not excisional, they ablate the mucosa of the hemorrhoid by controlled necrosis via various mechanisms. These include:

Sclerotherapy
Bipolar Diathermy
Direct Current Electrotherapy
Infrared Coagulation
Rubber Band Ligation

Sclerotherapy is the oldest method and involves injecting a sclerosant (mixture of phenol in oil, quinine, urea and hypertonic saline) into the submucosa at the base of the hemorrhoid through an anoscope using a spinal needle. The technique is suitable for smaller hemorrhoids. Cure rates are reported to be 90%.¹⁷ Recurrence rates are 30% at 4 years.¹⁸ Significant pain is a limiting factor in this method and has reported in 12-70% of cases.^19,20

Bipolar Diathermy, direct current electrotherapy, and infrared coagulation all involve coagulating, occluding and obliterating the hemorrhoidal vascular pedicle proximal to the dentate line. The area of tissue then sloughs and leaves an ulcer that fibrosis at the treatment site. Bipolar diathermy success rates range from 88-100% in randomized trials with first, second and third degree hemorrhoids.^{21,22,23,24,25} Multiple applications are required at the same site, and 20% of cases require excisional hemorrhoidectomy.^26,27 Direct electrotherapy has fallen out of favor because of lengthy treatment times (10 minutes) required to obtain coagulation despite similar success rates (88%) in first, second and third degree hemorrhoids.^28,29 A new method using a disposable probe that plugs into any electrosurgical generator is now available. (Figure 1)

Infrared photocoagulation (IRC) produces infrared radiation from a tungsten-halogen lamp which is directed via a polymer probe tip. Office based IRC (Figure 2) and through the scope probes (Figure 3) are available. These are required to make contact with the base of the hemorrhoid under direct vision to deliver 0.5 to 2 second pulses.²⁷ These polses are delivered to all 3 hemorrhjoids in positions at base of the hemorrhoid column (Figure 3a) Success rates range from 67-96% in first and second degree hemorrhoids. Multiple hemorrhoids can be treated at once.³⁰ The swiftness and notably rare incidences of pain and bleeding have made this the most commonly used method amongst the three coagulant therapies.

Rubber band ligation involves completely encircling the redundant mucosa, vascular bundle, and connective tissue with a ligating rubber band. The process must be at least 1-2 cm proximal to the dentate line lest severe immediate pain due to the innervation of the anoderm. The pain can only be relieved by cutting the rubber band, which is a technically challenging feat. Although internal hemorrhoids can be banded using an upper endoscope with a variceal ligation kit, there is a move towards office banding. During colonoscopy or sigmoidoscopy, the internal hemorrhoids are seen during retroflexion but may be missed if air is not suctioned from the rectum (see Figure 4). Figure 2 shows where the band should be placed in relation to the dentate line regardless of the method used. Two commonly used in-offices devices are available. One involves an instrument that physically grasps the mucosa and pulls the tissue into the applicator, and other involves an accessory suction device that sucks the mucosa into the applicator itself. (Figure 5) There are no reported differences in efficacy. Success rates are high and durable, ranging from 80-90% of patients with resolved or improved symptoms on 5 year follow-up.^31,32 Complications of rubber band ligation include minor bleeding in less than 5% of cases and severe bleeding in 1-2%. Rare incidences of pelvic sepsis occurring after rubber band ligation have been reported, carrying a high mortality rate (~30%).^33,34 Patients should be instructed to seek medical care urgently if symptoms of fever, increased perianal pain, or new onset of urinary retention following the procedure.

Multiple randomized controlled trials have compared each of the above method with one another, but no single has compared all 5 at once. Two meta- analyses concluded that rubber band ligation and infrared coagulation are the most effective. One meta- analysis reported that ligation was more effective because it had more durable benefits, requiring fewer additional treatments for symptomatic recurrence than did coagulation and sclerotherapy.³⁵ However, in a randomized crossover trial comparing rubber band ligation versus infrared coagulation, rubber band ligation resulted in more pain and minor bleeding than infrared coagulation. The use of both ligation and infrared coagulation in combination was 97% successful. Most notably, despite ligation having more pain and bleeding, there was no preference amongst patients for one over the other.³⁶

Surgical Hemorrhoidectomy

Overall, 5-10% of patients with hemorrhoids will require surgery � the vast majority are grade III-IV. Surgical hemorrhoidectomy is indicated in the following scenarios:

Failure of Medical and Nonoperative Therapy
Symptomatic third-degree, fourth-degree, or significant mixed internal and external hemorrhoids
Symptomatic hemorrhoids in the presence of a concomitant anorectal condition that requires surgery
Patient preference after discussion of the treatment options with the referring physician and surgeon.³⁷

Surgical hemorrhoidectomy is the most effective and durable treatment overall for third and fourth degree hemorrhoids. Recurrence following a hemorrhoidectomy is quite rare. However, morbidity is higher, especially with respect to incontinence, post- operative pain, urinary retention, bleeding, infection, and anal stenosis. Most patients do not return to work for 2-4 weeks.^{38,39,40,41,42} The stapled hemorrhoidectomy, developed in 1998, is equally safe as conventional hemorrhoidectomy (open or closed), but with shorter operating time, convalescence and post-operative disability. Effectiveness long term as of yet cannot determined absolutely but hemorrhoid symptoms can recur years later.⁴³

CONCLUSION

Hemorrhoids are normal fibrovascular structures underlying the rectal mucosa and anal skin. Physiologically they maintain the continence of the anus. Symptomatic hemorrhoids bleed, itch or hurt, and require therapy. The dentate line heralds the presence of pain fibers distally, and any intervention must be well proximal to it if done without anesthesia. The degree of prolapse has significant therapeutic implications. For the majority of patients with symptomatic grade I hemorrhoids, lifestyle and dietary changes are effective. Hemorrhoids that fail conservative measures (grade II, III) may be amenable to non-operative procedures that can be done by the gastroenterologist safely in the office setting without anesthesia, including rubber band ligation and infrared photocoagulation. While rubber band ligation seems more durable than photocoagulation, it appears to come at the expense of more pain and minor bleeding. However, patients do not seem to favor one over the other. Surgical hemorrhoidectomy, including a newer stapled technique, is the last line of therapy, and provide the most durable and effective response for refractory hemorrhoids, but expectedly with higher complication rates and prolonged convalescence.

Conference Digest November 2014

Practical Gastroenterology is excited to bring you this special supplement featuring abstract highlights and author insights from the American College of Gastroenterology’s 2014 Annual Scientific Meeting. Among the inflammatory bowel disease-related abstracts, we include those which focus on pregnancy, quality of life issues, treatment options and the importance of colonoscopy surveillance. The abstract highlights and author insights below first appeared on the ACG Blog and were selected by the ACG Educational Affairs and PR Committees as newsworthy.

A CASE REPORT

Perivascular Epithelioid Cell Neoplasm (PEComa) as a Cause of Abdominal Pain in a Child

Catherine Steingraeber,Eric Scaife,Hong Zhou,

Douglas G. Adler,

John F. Pohl

Author Disclosures: Dr. Pohl has received the following funding: INSPPIRE to Study Acute Recurrent and Chronic Pancreatitis is Children, Grant# 10987759, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases

CASE REPORT

A10 year-old previously healthy male presented with a chief complaint of epigastric pain and gastroesophageal reflux symptoms for two years. The abdominal pain was intermittent in nature. Although he complained of intermittent gastroesophageal reflux, he denied emesis, hematemesis, bilious emesis, dysphagia, or odynophagia. He had no weight loss or early satiety. In the gastroenterology clinic, the patient was noted to have normal vital signs and growth parameters (per CDC growth charts). His physical exam was normal. Laboratory testing revealed a normal complete blood count and hepatic function panel. An erythrocyte sedimentation rate, alpha fetoprotein level, lactate dehydrogenase level, and uric acid level were normal.

An esophagogastroduodenoscopy (EGD) was performed which was normal except for evidence of a raised area consistent with a possible mass near the antrum with no overlying mucosal changes (Figure 1). Forcep biopsies of this area were normal. A subsequent computed tomography scan of the abdomen revealed a round soft tissue mass (measuring 1.8 x 2.4 x 2.2 centimeters) in an anterior aspect of the fissure of the falciform ligament (Figure 2). This lesion was exerting a mass effect on the lesser curvature of the stomach. An endoscopic ultrasound (EUS) was performed which revealed a 2.4 centimeter, heterogeneous lesion that was round to oval in shape and well demarcated. The lesion was densely hypervascular, with extensive Doppler flow activity along the entire periphery of the lesion and, to a lesser extent, within the lesion itself. The lesion did not appear to arise from the gastric wall, and there was no peri-lesional adenopathy (Figure 3).

The patient underwent laprascopic surgical excision, and a spherical mass was found within the falciform ligament without evidence of peritoneal implants. The liver and gallbladder appeared normal. The mass was shown to measure 2.8 x 2.4 x 1.7 centimeters (weight 6.9 grams), and it consisted of an encapsulated purple- tan soft tissue mass with attached fibromembranous tissue (Figure 4). On microscopic examination, the mass demonstrated a fascicular and nested proliferation of spindle to epithelioid cells with clear to eosinophilic cytoplasm and round to oval nuclei (Figure 5). There was no mitotic activity, necrosis, or vascular invasion identified. Immunohistochemically, the cells were positive for smooth muscle marker (smooth muscle actin) and melanocytic markers (HMB-45 and Melan-A) (Figures 6 and 7). Cell staining was negative for AE1/ AE3, EMA, desmin, MYF-4, S-100, ALK-1, CD117, and CD34. These findings were consistent with a PEComa of the falciform ligament.

The pediatric oncology service was consulted on this patient, and it was decided that the risk of tumor recurrence and metastases was low. He has been scheduled for annual abdominal ultrasounds. Interestingly, all symptoms of abdominal pain and gastroesophageal reflux resolved after tumor removal.

DISCUSSION

PEComas neoplasms are characterized by perivascular epithelioid cells with a myomelanocytic immunophenotype that are typically arranged around the vasculature.^1,2,3 These cells have a characteristic epithelioid or spindled appearance, and such cells that express both smooth muscle and melanocytic cell markers using immunohistochemistry staining.⁴ Many PEComas are benign, but they also have been reported to be malignant. The ratio of PEComa formation is equivalent between males and females in prepubertal children, although there is a strong female predominance of PEComa formation in adolescents and adults.⁵ There is a known association of PEComa with tuberous sclerosis.^{6, 7}

The differential diagnosis of submucosal gastric masses is quite large but will include lipomas, duplication cysts, pancreatic rests, PEComas, gastrointestinal stromal tumors, metastatic melanoma, clear cell sarcoma, leiomyosarcoma, carcinoid tumor as seen in multiple endocrine neoplasia type 1, and paragangliomas.⁸ These lesions may be difficult to differentiate based on morphology alone although use of endoscopic ultrasound and tissue immunohistochemistry greatly aids in the diagnosis.^5,9,10 Notably, clear cell sarcomas of a gastrointestinal variant consist of spindle and epithelioid cells with diffuse S100 positivity and variable positivity for other melanocytic markers and can be confused with a PEComa based on immunohistochemistry. Molecular genetic studies to demonstrate the EWS-ATF fusion gene representing t(12;22) q13;12) translocation or the EWS-CREB3L2 fusion gene representing t(2;22) q34;12) translocation can be used to differentiate clear cell sarcoma from PEComa a clear cell sarcoma is often S100 positive and may be difficult to differentiate from PEComa based on immunohistochemistry.^10,11,12

Previously, tumors such as angiomyolipoma, lymphangioleiomyomatosis, primary extrapulmonary sugar tumor, and clear cell myomelanocytic tumor were noted to have similar histologic characteristics although they were present in different anatomic locations. Hence, the term “PEComa” has been developed to broadly characterize this tumor grouping with specific histologic findings regardless of anatomic location or morphology.^{5, 13} As a result, PEComas have been described as occurring at the kidney, liver, lung, nose, bladder, and other locations.^{1, 13, 14, 15, 16} PEComas have been noted to occur in the gastrointestinal tract anywhere from the stomach to the rectum. PEComas of the gastrointestinal either tend to be located in the serosa with involvement of all associated bowel layers or will be polypoid tumors involving the mucosa and submucosa.8, 17 In our described case, the presumed gastric mass was, in actuality, a gastric compression from a PEComa at the falciform ligament which is a known location for such a tumor.¹⁸

PEComas are rare in children, and approximately 40 pediatric cases have been documented as occurring in the pelvis, vagina, eye/orbit, and gastrointestinal tract (including the duodenum, appendix, colon, and rectum).⁵ Abdominal pain may be a presenting symptoms in a patient with an abdominal PEComa.¹ In 2000, a unique pediatric case series of 7 patients described PEComas occurring exclusively at or near the ligamentum teres and falciform ligament. Most of these cases occurred in females, and the cases occurred at an average age of 11 years. Immunohistochemistry staining demonstrated that the tumors were positive for smooth muscle actin, melin-A, and myosin but were negative for desmin, in a pattern similar to our patient. Follow-up data, available in six of seven cases, showed five patients to be free of disease, and one to have a radiographically presumed lung metastases.¹⁹

PEComas have been described as benign or malignant although the criteria for malignancy are not well defined. Most PEComas located at the falciform ligament or ligamentum teres are benign, and the therapy for such tumors is surgical resection.⁵ PEComa features associated with a higher risk of malignancy include large size (greater than 5 centimeters), infiltrative growth, mitoses greater than 1 in 50 per high power field, vascular invasion, and necrosis.²⁰ The PEComa of our described patient had a diameter less than 5 centimeters and had no significant nuclear atypia, necrosis, or mitotic activity likely consistent with a benign lesion.

Surgical resection is the treatment of choice for most PEComas; however, chemotherapy may be indicated for PEComas with histologic features consistent with malignant disease or in the event of associated metastases.^{5, 21} Doxorubicin and ifosfamide have been used as chemotherapy for PEComas, and rapamycin, an inhibitor of m-TOR, also has been used as treatment for malignant PEComa.^{22, 23} Close follow-up with serial imaging (such as ultrasound) is necessary to evaluate for PEComa recurrence.²⁴

NUTRITION ISSUES IN GASTROENTEROLOGY, SERIES #135

Fresh Look at Holiday Foods for those with Dysphagia

Laura Knotts,

Carole Havrila

Dysphagia may result from a disease process, surgery, trauma, dental work, or congenital problem affecting any aspect of the swallowing process from the mouth to the esophagus. In this article, we outline recipes and how-to tips to prepare soft, moist foods, so that patients living with dysphagia will once again be able to experience the joy associated with holiday eating.

INTRODUCTION

For a variety of reasons, many patients experience dysphagia, or difficulty swallowing, and may therefore require liquids or soft consistency foods to meet their nutrition requirements. Dysphagia may result from a disease process, surgery, trauma, dental work, or congenital problem affecting any aspect of the swallowing process from the mouth to the esophagus. Some of the more common reasons include mechanical issues such as cancers in the mouth or throat and neurological issues caused by stroke, Amyotrophic Lateral Sclerosis (ALS), or Parkinson?s disease. In addition, patients who have undergone esophageal surgeries and/or stent placements often require a period of altered food consistency. Softer food choices can make manipulating foods in the mouth easier and shorten meal times, while reducing the risk of aspiration.

Numerous specialists are involved in determining the appropriate course of action for patients with dysphagia. Speech-language pathologists often direct oropharyngeal dysphagia recommendations, while gastroenterologists and radiologists usually manage esophageal dysphagia. The Registered Dietitian?s job is to translate these recommendations into actual foods, taking into consideration the nutritional adequacy of a limited consistency diet.

People with dysphagia go through profound adjustments in their lives. The dietary changes that are imposed, in particular, often create a feeling of loss related to the social aspects of eating that are so important in our lives. This can result in a decrease in intake that leads to significant, and sometimes severe, weight loss. Without appropriate intervention, such a cycle will negatively impact quality of life, and health.

Below is a collection of festive holiday recipes and how-to tips to prepare soft, moist foods, so that people living with dysphagia will once again be able to experience the joy associated with holiday eating. We think these recipes are so tasty, they will delight all of those around the holiday table. Bon Appetit!

Sensational Holiday Recipes

The holidays are associated with the savory entrees, side dishes and soups we all know and love. While any favorite meal can be processed to meet swallowing consistency needs, we offer the following recipes that may become some of your new favorites.

SUMMARY

The holidays bring opportunities to enjoy a variety of festive dishes. With some planning, those living with dysphagia can continue to enjoy the celebrations (and food!) that so often accompany the holiday season. Hopefully, these recipes will allow your patients to feel more included in their traditional family festive food fares and more fully enjoy their holiday season.

See this article’s PDF for complete recipes for side dishes, entrees, smoothies and shakes, snacks and desserts.

GERD IN THE 21st CENTURY, SERIES #24

Circadian Reflux Pattern on PPI Therapy

Mustafa Abdul-Hussein,

Janice Freeman,

Carlos Zapata,

Donald O. Castell

Proton pump inhibitor (PPI) therapy has become the standard treatment for patients with GERD. The normal circadian pattern of intragastric acidity is an increase in acid secretion beginning in the evening until midnight. The circadian pattern for esophageal reflux is largely unknown. The aim of this study was to analyze the circadian pattern of both acid and non-acid reflux and the effect of PPI therapy.

INTRODUCTION

Gastroesophageal reflux disease (GERD) is a common entity, affecting approximately one third of the US population at least once a month and 7% on a daily basis.¹ The presenting symptoms associated with GERD vary from the more typical heartburn and regurgitation to extraesophageal symptoms such as cough and throat clearing. Proton pump inhibitor (PPI) therapy has become the standard treatment for patients with GERD.²

The normal circadian rhythm of intragastric acidity is usually an increase in the acid secretion in the evening until the middle of the night and then a decrease in the morning.³ The circadian rhythm for total esophageal reflux episodes is unknown. Gudmundsson⁴ in 1988 showed that the time pattern of GERD separated into 3 periods when studied with ambulatory 24 hour esophageal pH-monitoring and identified the most acid reflux episodes in the evening.

Ambulatory Impedance-pH technology (Imp-pH) provides the opportunity to monitor all reflux episodes; acid and non-acid in type, both in the upright and recumbent positions.^5,6 This test avoids a false positive study due to acidic meals or drinks and is becoming the gold standard for detection and analysis of GERD and for clarifying its relationship to symptoms.^7,8 Excluding the meal periods from ambulatory pH monitoring improves the diagnosis of esophageal reflux disease.⁹ The goal of this study was to further characterize the daily pattern of reflux and the effect of PPI therapy on it using combined pH and impedance to identify all types of reflux.

MATERIALS AND METHODS

A retrospective review was performed on MII-pH studies from 300 patients (162 females); age range 18-84 years, who underwent testing to evaluate symptoms felt to be possibly due to GERD. These MII-pH studies were performed on 100 patients each on twice a day (BID) PPI therapy, once a day (QD) PPI therapy and off (OFF) PPI therapy. The Imp-pH studies were performed from February 2006 until February 2008.

Indications for the test were heartburn, regurgitation, indigestion, persistent cough, asthma or throat clearing.

Exclusion Criteria:

Patients on more than one reflux therapy agent (H2 blockers, more than 1 PPI, baclofen, sucralfate)
Studies with only upright or only recumbent readings.
History of a previous Nissen fundoplication.

A standard multichannel intraluminal impedance pH (Imp-pH) catheter (Sandhill Scientific Inc., Highlands Ranch, CO) was used. This 2mm diameter catheter has the ability to record esophageal and gastric pH, along with the presence and direction of any liquid flow in the esophagus. The MII-pH catheter was passed through the nasal cavity and into the stomach in the morning 0800-1000 in fasting patients. The catheter was slowly pulled back in a stepwise fashion to locate the LES using the single pressure sensor in the distal portion of the catheter. Dual pH monitoring was located at 5 cm above and 15 cm below the proximal margin of the LES with 6 impedance sensors at 3, 5, 7,9,15 and 17 cm above the LES.

The following parameters were assessed:

Total number of reflux episodes (acid and non- acid) per hour (both upright and recumbent)
Type of reflux; acid or non-acid.
Relation of reflux episodes to the beginning of the recumbent time.

The daily analysis was further divided into 3 segments: morning between 0600- 1400, evening between 1400- 2200, and night between 2200-0600. Meal times indicated by the patient on MII-pH studies were excluded from data analysis. Comparison of the 3 groups was performed using ANOVA.

RESULTS

The mean number of daily reflux episodes per patient off PPI was 24.3, on QD PPI it was 32.00 and on BID PPI it was 28.2 (Table 1). These were not significantly (NS) different, although interestingly numerically greater in the two groups studied on PPI. The mean number of acid reflux episodes per patient was 15.6 off PPI. It was 6.2 on QD PPI and 5.1 on BID PPI (p<0.05). The mean number of non-acid reflux episodes per patient in off PPI was 8.7, on QD PPI it was 25.8 and on BID PPI was 23.1 (p<0.05). There was also no significant difference between total number nor type of reflux seen on PPI therapy comparing QD with BID.

The circadian patterns of reflux episodes in the three patient groups were similar. (Fig. 1-3) The circadian pattern of the off PPI group (Figure 1) is similar to that of the QD PPI group (Figure 2), with peaks around meal and postprandial hours and a drop in the number of episodes after 11pm. The circadian pattern in the BID PPI group (Figure 3) is similar to those of the off PPI and QD PPI groups. There is an increase in the average reflux episodes per hour around lunch time which continuously increases with peaks around 2pm and 7pm related to the post prandial state. Reflux episodes were frequent in the evening postprandial period. There is a sudden drop in the number of reflux episodes after 11pm. During night time recumbency there were less frequent reflux episodes with no significant difference between the OFF and ON therapy groups (< 1 episode/ hour in all groups).

Figure 4 (a,b,c) shows the decrease in the average number of reflux episodes when assuming the recumbent position, shown with episodes aligned to the start of recumbency (R) for all 3 groups.

The total number of reflux episodes seen in the 3 groups was 8455 (Figure 5). Of these, only 862 (10.2%) occurred in the recumbent position. In the off PPI group, 11.0% of the episodes were seen in the recumbent position. In the QD PPI group it was 10.3% and in the BID PPI group it was 9.4% of the total episodes (NS)

When analyzing the mean reflux episodes per hour during each segment of the day, most of the episodes (nearly double) were seen in the evening when compared to the other two segments of the day. (Figure 6) The mean number of reflux episodes in the evening in the off PPI group was 1.9 and was 2.6 and 2.3 in the QD and BID PPI group respectively. These values are not significantly different. Also, there was no significant difference between the mean reflux episodes in the morning. Mean reflux episodes was 1.0 in the off PPI group, 1.3 in the QD PPI group and 0.9 in the BID PPI group. During the night time, mean number of reflux episodes was 0.6 in the off PPI group, 0.7 in the QD PPI group and 0.5 in the BID PPI group (NS).

As expected, when the total reflux episodes were divided into acid and non- acid in type in the off PPI group (Figure 7A), there was more acid type reflux seen. In contrast, in the 2 on therapy groups (QD and BID)), the number of non- acid type reflux predominates. (Figures 7B and 7C respectively).

DISCUSSION

Reflux occurs in all individuals and it is affected by position, feeding, and state of wakefulness. Circadian variations shown by Stein in 1990 found no differences in the frequency of esophageal contractions during upright, supine, and meal periods in normal volunteers or patients with GERD.² Wang in 1996 found that there were more reflux episodes in the upright than in the supine position¹⁰ in 10 healthy subjects.

Reflux episodes are believed to be more common at nighttime than in the rest of the day, especially in the early part of the nighttime¹¹ and are associated with late meals and snacks. Also there is evidence that nighttime GERD has a greater impact on a patient?s life than daytime GERD.^12,13 Moore and Englert found greater rates of acid secretion in the evening with no variation in the serum gastrin levels when compared to the morning levels.^14,15 Sharma et.al.¹⁶ showed that many patients on PPIs still experience GERD symptoms that were frequently associated with non-acid reflux.

An important observation from our data is that circadian patterns of all types of reflux (acid and non- acid) are similar in patients on and off PPI therapy. There is no change in the number of reflux episodes, especially in the upright position, but PPI therapy does change the composition of the refluxate from acid to primarily non-acid in type.

We found that about 90% of reflux episodes occur in the upright position and in the evening related to a meal and the post prandial state. This is not a surprise since transient lower esophageal sphincter relaxations mainly occur in the meal-distended stomach.¹⁷.

There is also a dramatic decrease in episodes with the onset of recumbency, which strengthens the concept of transient lower esophageal sphincter relaxation (TLESR) and suggests that in the recumbent position, the LES contracts to prevent the passage of food and secretions from the stomach to the esophagus.¹⁸

Our study confirmed the results of prior study done by Tamhankar et.al.¹⁹ Also, it extended the analysis to include the hourly circadian pattern of reflux episodes, both acidic and non-acidic in patients taking different types of PPIs.

Dent et al. in a study of mechanisms of GERD in the recumbent position in 10 healthy volunteers, showed that only 20% of reflux episodes defined by pH level <4 occurred after midnight while the majority occurred within a few hours after the evening meal.²⁰ Our data confirm and extend that study to patients with suspected GERD symptoms. In addition, the advanced technology provided by combined MII-pH monitoring allows detection of all types of reflux, both acid and non-acid. The latter is particularly important during the post-prandial periods when buffering from the meals produces a milieu where refluxate is largely pH >4.

The highest percentage of recumbent episodes occur in the off PPI therapy group probably because during recumbency, which is generally associated with sleep, the normal physiological acid secretion response is not altered by meals, drinks (especially carbonated), stress or changes in positions that is more commonly seen in the upright position or awake. In that way, therapy can be more effective. Also, one should consider that while the patient is sleeping, it will be more difficult to record the symptoms on the tracing unless they are so severe that they awaken the patient.

One particular finding from our study was that in the off PPI group, the patients had a trend toward less reflux episodes when compared to the 2 on PPI therapy groups (which was significant between the QD PPI therapy group and the off PPI therapy group in the recumbent position only). This finding was unexpected. Does this mean that PPI therapy increases the number of reflux episodes? One possible explanation for these findings is that the patients taking PPIs had fewer symptoms than patients off PPIs so that they will tend to eat more food and will be more comfortable with having larger meals when compared to the off therapy group.

We did not comment on symptom association with reflux (acidic and nonacidic) in this study as the main goal was to document that PPI therapy does not totally reduce reflux episodes, it only alters the pH of the refluxate. We think that it will be important to consider new therapeutic agents for GERD that could eventually reduce the amount of reflux episodes and not only their composition. n

GASTROINTESTINAL MOTILITY AND FUNCTIONAL BOWEL DISORDERS, #3

Superior Mesenteric Artery Syndrome

Yi Jia,

Omar Sosa,

Richard W. McCallum

The recognition of Superior Mesenteric Artery Syndrom (SMA) as a distinct clinical entity remains controversial because it can be confused with other anatomic or motility- related causes of duodenal obstruction. It was often regarded as a diagnostic dilemma and a diagnosis of exclusion. In this article we will review the pathophysiology, symptoms, diagnosis and treatment of SMA syndrome.

INTRODUCTION

Superior mesenteric artery (SMA) syndrome was first described in 1842 by Rotikansky during an autopsy for duodenum compression.¹ Wilkie in 1921 defined the patho-physiological changes of the third (transverse) portion of the duodenum when obstructed by arteriomesenteric compression and used the term “chronic duodenal ileus”.² He published the first comprehensive SMA syndrome case series of 75 patients in 1927.³ Since then, approximately 500 cases of SMA syndrome have been reported.^4,5

SMA syndrome, also known as Wilkie disease, duodenal arterial mesenteric compression, duodenal ileus, aortomesenteric artery compression, and cast syndrome, is an uncommon and sometime life threatening gastrointestinal-vascular disorder.^6-8 Some studies report the incidence of SMA syndrome to be 0.013-0.3% of the general population.^9,10 Approximately 0.013-0.78% of routine upper GI barium studies identify SMA compression suggesting this diagnosis.¹¹ 75% of the patients reported with SMA syndrome are aged 10-30 years and predominantly females.^4,9,10 A delay in the diagnosis of SMA syndrome can result in malnutrition, electrolyte inbalance, gastric perforation, pneumatosis and hypovolemia, with a reported mortality rate up to 33%.^12-15 In the usual clinical setting its nonspecific presentations can result in under-diagnosis and severe outcomes.^16,17

The recognition of SMA syndrome as a distinct clinical entity remains controversial because it can be confused with other anatomic or motility-related causes of duodenal obstruction.^4,5,18 The left renal vein may also be entrapped and compressed between the aorta and the SMA, a clinical setting referred to as “nutcracker syndrome”.⁵ The symptoms of SMA syndrome do not always correlate well with abnormal anatomic findings on radiologic studies, and may not resolve completely after treatment.^4,18,19 It was often regarded as a diagnostic dilemma and a diagnosis of exclusion. In this article we will review the pathophysiology, symptoms, diagnosis and treatment of SMA syndrome.

Pathophysiology

Anatomically, the third portion of the duodenum typically crosses caudal to the origin of SMA, passing between the aorta and the superior mesenteric artery at the level of 3rd lumbar vertebral body, and being suspended by its attachment to the ligament of Treitz (Figure 1). SMA normally arises from the anterior aspect of the aorta at the level of the L1 vertebral body, and is enveloped in fatty and lymphatic tissue. It forms a takeoff angle of approximately 45° (normal range 38-65°) from the abdominal aorta with the normal mean aorto-mesenteric distance of 10-28 mm.^17,20,21

Any factor that decreases this takeoff angle and narrows the aorto-mesenteric distance can compress the third part of the duodenum as it passes between the SMA anteriorly and the spine posteriorly, resulting in SMA syndrome. In the case of SMA syndrome, this takeoff angle can be sharply narrowed to approximately 6-25° and the aorto-mesenteric distance can be decreased to 2-8 mm, causing entrapment of the third part of the duodenum and mechanical obstruction at the level of the third and fourth parts of the duodenum (Figure 2).^6,18,20,22

Predisposing conditions for SMA syndrome that can change the aorto-mesenteric angle have been categorized into three groups. Group A is a function of weight loss. Significant weight loss leading to loss of the mesenteric fat pad and accompanying decreased body mass index is the most common cause of SMA syndrome. Reduction of the retro-peritoneal fat, which lies between the duodenum and the spine allows the SMA to compress the duodenum against the vertebrae. In the absence of an appropriate fatty support, the angle at which the SMA leaves the aorta promotes more compression of the third portion of the duodenum.^23,24 This weight loss can be a consequence of medical, psychological or surgery disorders, malignancy, malabsorption syndromes, human immunodeficiency viral infection, trauma, anorexia, burns, bariatric surgery, diabetes mellitus and eating disorders.^7,25-30

Group B is attributed to external causes, such as the corrective spinal surgery for scoliosis with instrumentation or body casting resulting in prolonged post operation recovery. This procedure can displace the origin of the superior mesenteric artery by relative lengthening of the spine, reducing the aorto-mesenteric takeoff angle and decreasing the mesenteric artery’s lateral mobility, which is also termed as “cast syndrome”.^14,31-33 It has been demonstrated that the incidence of SMA syndrome after surgical procedures for correction of spinal deformities varies between 0.5 and 4.7%.^14,34-36

Group C is a function of intraabdominal anatomy, either congenital or acquired, such as compression or mesenteric tension (e.g. aortic aneurysm), low origin of the superior mesenteric artery, following esophagectomy where gastric pull up into the chest changes the upper GI anatomy and peritoneal adhesions.^37-39 A congenital short ligament of Treitz may pull the duodenum up towards the insertion of the aorto-mesenteric angle predisposing to SMA and malrotation of the small bowel is another predisposing factor.⁴⁰

Symptoms

SMA syndrome patients may present acutely or more insidiously with progressive nonspecific symptoms, the severity depending on the degree of duodenal obstruction.¹⁹ Acute presentations can be related to post-traumatic surgery and often develop from 6-12 days after surgery, and are explained by hyperextension of the SMA compressing the duodenum, sometimes combined with prolonged periods in surgical casts.^33,34 Another pattern of presentation is in the setting of substantial weight loss that could be related to an eating disorder, limited oral intake or vomiting disorders or cachexia of malignancy or serious depression. A more insidious presentation that may be seen by gastroenterologists involves a long history of abdominal symptoms, where the link to compression of the duodenum is overlooked.⁴¹ Patients with mild obstruction may have only postprandial epigastric pain and early satiety, while those with more advanced obstruction may have severe nausea, postprandial abdominal pain, bilious emesis and weight loss.^6,41 These symptoms are associated with reduced food intake related to delayed gastric emptying from the retropulsion of food from the duodenal compression with the accompanying compensatory reversed peristalsis. A key aspect of the history that needs to be elicited is that the symptoms are definitely exacerbated by a meal and are better when fasting or overnight. Other aspects of the history include worsening if lying on the right side or supine, and relief by the Hayes maneuver (pressure applied below the umbilicus in cephalad and dorsal direction), lying prone, knee-chest, or left lateral decubitus positioning.^4,10,42 These positions of alleviation remove tension from the mesentery and SMA, elevating the root of the SMA and increasing the aorto-mesenteric angle and distance. Findings on physical examination are nonspecific but can include abdominal distension, a succussion splash, and high-pitched bowel sounds.

Because of the nonspecific nature of the symptoms, clinicians need a high degree of suspicion in order to diagnose SMA syndrome. The diagnosis is often delayed and arrived at through the process of excluding other etiologies of intestinal lumen obstruction.⁴¹ The differential diagnosis of SMA syndrome includes other causes of bowel obstruction, duodenal dysmotility and gastroparesis such as diabetic gastroparesis, chronic pancreatitis, systematic autoimmune disease, chronic mesenteric ischemia, idiopathic intestinal pseudo-obstruction, megaduodenum, often caused by connective tissue diseases, especially scleroderma.^10,43 Post prandial pain of unknown origin, irritable bowel syndrome, dyspepsia, including peptic ulcer and H. pylori gastritis may also be the working diagnosis in patients with nonspecific symptoms (Table 1).

Delay in recognition of SMA syndrome can often result in complications, such as severe electrolyte abnormalities, malnutrition, obstructing duodenal bezoar, gastric dilation with a risk of perforation and pneumatosis, which could portend a fatal outcome.^12,15,44 Some patients may be receiving chronic narcotics for pain relief, which will further inhibit duodenal and gastric motility promoting postprandial vomiting.⁴²

Diagnosis

The diagnosis of SMA syndrome is usually established by combinations of abdominal radiographs with barium study, ultrasonography, Computed tomographic (CT) angiography, and magnetic resonance imaging (MRI).^45-47 Although less high technology and more “old school”, we still recommend an upper GI barium study with a small bowel follow through and Doppler blood flow assessment as the initial diagnostic evaluations of SMA syndrome. Another advantage of abdominal radiographs and ultrasound is the low cost. Their findings may demonstrate a dilated proximal duodenum with an abrupt termination of the barium column in the third and fourth part of the duodenum proximal to the ligament of Treitz with a normal jejunal caliber beyond the ligament of Treitz.¹⁰ In addition, this oral contrast study could show significantly prolonged retention of barium proximal to the third portion of duodenum and hold up before entering the jejunum as well as dilation of the proximal duodenum and stomach associated with retrograde flow of contrast from reverse peristalsis (Figure 3).¹⁹ It is key to have an informed radiologist performing the small bowel series and the clinician needs to be involved and interacting to help interpret the findings with the radiologist. Follow up EGD to investigate these radiographic findings can reveal old food in the stomach, duodenal dilation with or without bezoar formation, and obstruction at the end of the third part of duodenum.

Conventional arteriography was traditionally performed simultaneously with the barium study to demonstrate the superior mesenteric artery superimposed upon the barium-filled duodenum with the decreased aorto-mesenteric angle.⁴ Ultrasound is a noninvasive method to evaluate the mesenteric artery anatomy and measure the aorto-mesenteric angle.^45,48 Positional maneuvers, such as having patients in the lateral decubitus or even standing, may identify alterations in the aorto-mesenteric angle. Endoscopic ultrasound has recently been effective in demonstrating in more detail and better definition the anatomic abnormalities associated with SMA syndrome.⁴⁹

More advanced imaging studies, such as CT and MRI angiography, are often ordered in the setting of patients with variable abdominal pain with nausea and vomiting and unclear diagnosis.⁴⁷ These noninvasive images can provide additional anatomic details such as the compressed bowel in relation to vessels, and calculate the aorto-mesenteric angle and the amount of intra-abdominal and retroperitoneal fat tissue.^10,46 Recently three-dimensional reconstruction has revealed increased blood flow velocity through the SMA and may unexpectedly reveal other possible causes of the abdominal pain, such as abdominal aneurysm (Figure 4).⁴⁷

The diagnosis of SMA syndrome should come to mind in patients with clinical features of duodenal obstruction and imaging results revealing duodenal obstruction in the third portion with active retrograde peristalsis. More studies should then be pursued to define if the aorto-mesenteric angle is ≤25° and the aorto-mesenteric distance is ≤8 mm. Other features to note are high fixation of the duodenum by the ligament of Treitz, or abnormally low origin of the superior mesenteric artery or anomalies of the superior mesenteric artery.

Patients with SMA syndrome usually have undergone extensive gastrointestinal evaluation and procedures over a period of time, including upper gastrointestinal endoscopy and colonoscopy, to exclude malabsorptive, ulcerative and inflammatory intestinal conditions. These procedures are expensive, have potential risk for patients, and add to the overall economic burden associated with diagnosing this elusive entity.^5,40 When faced with a diagnostic dilemma of unknown etiology of abdominal pain or nausea and vomiting, the gastroenterologist should take or re-take a thorough history and review the imaging studies keeping in mind some less common entities in the differential diagnosis, such as the possibility of undiagnosed SMA.

Treatment

Initial conservative treatment with reversal of any precipitating factor is recommended in all patients with superior mesenteric artery syndrome.^5,10 Initial gastric decompression by nasogastric tube aspiration can reduce the dilated stomach and proximal duodenum and help to monitor fluid balance.⁵ Patients with acute SMA syndrome usually have electrolyte imbalances, which should be monitored and corrected aggressively. Nutritional support and attempts to increase weight are important initial considerations. These approaches may include an enteral nasojejunal feeding tube if it can be successfully passed through the narrowed duodenum, or a laparoscopic placement of a jejunal feeding tube beyond the ligament of Treitz may be necessary for a short period of 3 to 6 months. Total parenteral nutrition is not recommended because of complications. All these measures are aimed at increasing the fat pad between the spine and duodenum.^49,50

Patients with suspected eating disorders need professional nutritional and psychiatric evaluation to help achieve optimal calorie replacement.⁵ Pro-motility medications, such as metoclopramide, are not appropriate since increasing upper GI motility when duodenal obstruction is present increases the pain component. Antiemetics can be supportive. Younger patients in particular with acute SMA syndrome would benefit from conservative treatment, which should be instituted for at least 3 to 6 months.^5,8

Specific indications for surgery in patients with chronic SMA syndrome include failed conservative treatment, long-standing symptoms, continuing weight loss due to abdominal pain, nausea and vomiting and reduced food intake, and marked duodenal dilatation with stasis.^5,40 (Figure 5) Co-management with dieticians and psychiatry consultation when appropriate are recommended to ensure the best outcomes including concerns for adequacy of wound healing after surgery.

The most common surgical operation for SMA syndrome is duodenojejunostomy, in which the compressed portion of the duodenum is actually bypassed by constructing an anastomosis between the 2nd portion of the duodenum and proximal jejunum anterior to the superior mesenteric artery. Duodenojejunostomy can reestablish the bowel continuity with a success rate > 90%.¹⁶ (Figure 5C, the preferred surgery) Another version of this duodenojejunostomy is division of the 4th part of the duodenum. (Figure 5D) This is not the recommended best option but rather the approach of leaving the duodenum-jejunal continuity intact is the preferred surgery. Surgical complications include bleeding, leakage or stricture at the anatomies site.51 Long term concerns include small bowel bacterial overgrowth in the “blind loop” created in the bypassed 3rd and 4th parts of duodenum.

Another surgical option is gastrojejunostomy (bypass the obstruction by bringing up loop of jejunum to the stomach and anastomosis) (Figure 5B). This may be a potential consideration when adhesions from previous surgeries prevent adequate access to create the Duodeno-jejunal anastomosis. Another surgical approach is the Strong’s procedure (duodenal derotation with lysis of the ligament of Treitz).^8,22,52 However, this procedure is now largely of historic interest and has a higher failure rate to relieve the duodenal obstruction.⁵ Successful laparoscopic techniques for the duodenojejunostomy and Strong procedures have been described.^53,54 Although current literature is limited to case reports and small studies, laparoscopic approaches offer a less invasive surgical option. An important point to address is that all of these surgeries have one thing in common; the SMA itself is never displaced, re-routed or surgically altered. This is a key anatomic feature of the surgeries.

Long term outcomes in SMA syndrome patients after surgery are limited in the literature. One series of 16 patients found significant weight gain, but most symptoms remained unchanged except for decreased vomiting.⁴ The need to address eating disorders, bulimia, and underlying psychiatric issues is a key aspect of post-surgery management. A recent series of 8 patients described improved symptoms but no weight gain.⁵ Surgical morbidity and mortality can be affected by other comorbidities, e.g. diabetes, and end-stage renal disease.⁵⁵

CONCLUSION

Superior mesenteric artery (SMA) syndrome is an uncommon but well recognized clinical entity characterized by compression of the third portion of the duodenum between the aorta and the superior mesenteric artery. This can result in an acute presentation or more commonly chronic nonspecific symptoms explained by duodenal obstruction with decreased aorto-mesenteric angle and distance. The SMA syndrome has a spectrum of symptoms, which can be referred to as “great mimickers” of a GI motility disturbance. The main GI motility conditions that would be encompassed are: delayed gastric empting; dilated duodenum suggesting intestinal pseudo-obstruction; unexplained nausea and vomiting and abdominal pain, suggesting the spectrum of cycle vomiting syndrome on one hand, and irritable bowel syndrome on the other hand. This is a great challenge to the physician in practice.

We recommend that clinicians focus on the following “clinical pearls”: 1) unexplained abdominal pain provoked by eating and accompanied by nausea and vomiting; 2) endoscopic evidence of retained food in the stomach and a dilated proximal duodenum; 3) a slow scintigraphic gastric emptying result (Retention of >60% of isotope at 2hrs and >10% at 4hrs). Armed with these clues, the possibility of SMA syndrome has to come to mind, so the next logical step is to get a “road map” and obtain an upper GI and small bowel series. If the clinical suspicion is borne out by a suggestive contrast study then more sophisticated noninvasive imaging can be pursued to demonstrate the specific features we have summarized in this review. SMA syndrome patients need initial conservative treatment where reversible aspects, particularly nutrition, are the focus. Surgical options are effective for non-responding patients, although complete resolution of all symptoms may not always be achieved. The role of explanation, education and practicing the “art of medicine” are all important features in the total care and outcome of these patients. Making a diagnosis in a patient who has been told there is no explanation for the abdominal pain, nausea and vomiting is very satisfying, particularly when there is a treatable and reversible entity involved. Such is the world of SMA syndrome, an often overlooked entity.

A CASE REPORT

Jejunal Dieulafoy Lesion: A Rare Cause of Lower Gastrointestinal Bleed

October 2014 • Volume XXXVIII, Issue 10

Natalia Khalaf,Francis Felice,Divyesh Sejpal

CASE REPORT

An 83-year-old Caucasian woman with past medical history of hypertension, gastroesophageal reflux disease, diverticular disease, iron deficiency anemia and recurrent occult gastrointestinal (GI) bleeding secondary to jejunal arteriovenous malformations (AVMs), treated with cauterization, presented to the emergency department with a three day history of intermittent, mild epigastric pain and five episodes of black, tarry stool.

On initial examination, the patient was in no acute distress, afebrile, and had a blood pressure of 110/54 with a heart rate of 96. She had diffuse upper abdominal tenderness but normal bowel sounds and no signs of an acute abdomen; her rectal exam was significant for melena without palpable masses or bright red blood. Her laboratory work-up was significant for a hemoglobin of 9.1 g/dL, from a baseline of 14.0, with an MCV of 76 fL, a reticulocyte count of 3.2%, and an elevated blood urea nitrogen to creatinine ratio of >20:1. Complete blood count and metabolic panel were otherwise normal. The patient was fluid resuscitated, started on a proton pump inhibitor drip, and taken for emergent esophagogastroduodenoscopy due to concern for an acute upper GI bleed.

Due to her previous history of jejunal AVMs, a pediatric colonoscope was used with the intent of performing push enteroscopy if needed, as balloon enteroscopy was not available at the facility. In the proximal jejunum, an actively bleeding arterial vessel was seen. The surrounding mucosa appeared normal, without evidence of ulceration. These findings were consistent with a jejunal Dieulafoy lesion (see Figure 1). Three hemoclips were applied to the bleeding vessel, resulting in complete hemostasis (see Figures 2 and 3). Following the procedure, the patient had resolution of her abdominal pain and only one subsequent episode of melena. She was discharged on hospital day number three with stable hemoglobin above 10.0 g/dL.

DISCUSSION

Dieulafoy lesions are becoming increasingly recognized as causes of acute GI bleeds, causing nearly 2% of all GI hemorrhages.1 Greater than 80% of these lesions are found in the upper GI tract.^{1-2, 4} True prevalence is hard to establish, given the difficulty in diagnosing Dieulafoy lesions and their often asymptomatic nature. Originally described as exulceratio simplex, and fatal 80% of the time in the era of surgical treatment,4 Dieulafoy lesions were initially believed to be ulcerations of the GI mucosa with resultant exposure and bleeding from an otherwise normal submucosal artery.⁵ It is now understood that these lesions are histopathologically different than normal vessels. Dieulafoy lesions are aberrant, dilated, submucosal vessels that fail to undergo normal branching and thus carry the potential to subsequently erode the overlying epithelium due to their large size and pulsatile nature.⁶

More prevalent in the elderly and those with comorbid chronic medical conditions,^{1, 3} Dieulafoy lesions are unevenly distributed in the GI tract. The stomach, specifically the lesser curvature due to its highly vascular architecture, is the most common location. Studies suggest that only 1-2% of such lesions are found in the jejunum.^{1, 7} Thus, our patient represents an unusual case of a jejunal Dieulafoy lesion causing acute lower GI bleed.

Endoscopic treatment is the standard of care for bleeding Dieulafoy lesions, with hemostasis approaching 80-100% depending on the method of intervention.^{2, 3, 8} Four methods of achieving endoscopic hemostasis are available: endoscopic hemoclip placement, band ligation, cauterization, and injection of vasoconstrictive compounds. Band ligation and endoscopic hemoclip placement are equally efficacious, with one study suggesting that banding may be preferred due to faster procedure time and shorter length of hospital stay.² For our patient, endoscopic hemoclips were the intervention of choice. Using clips avoids the drawback of needing to remove the endoscope after localization of the bleeding lesion to load the banding kit.

CONCLUSION

This case serves to heighten awareness of the commonly missed diagnosis of Dieulafoy lesion in areas outside of the stomach as a cause of obscure GI bleed. A thoughtful approach to such patients and a trained endoscopist with appropriate and specific equipment should be used to appropriately diagnose and treat this condition. Our patient’s prior history of repeated upper and lower GI bleeds secondary to AVMs serves as a humble reminder that patients often have multiple risk factors and causes for GI bleed. Dieulafoy lesion should be included on the differential diagnosis of potentially life-threatening causes of acute GI bleed, as specialized and time- sensitive management is needed.