GASTROENTEROLOGY 2003;124:1111–1134
`
`AGA Technical Review on Short Bowel Syndrome and
`Intestinal Transplantation
`
`This literature review and the recommendations therein were prepared for the American Gastroenterological Association Clinical
`Practice Committee. This paper was approved by the Committee on August 5, 2002, and by the AGA Governing Board on
`November 1, 2002.
`
`The normal human small intestine length is generally
`
`considered to be between 3 and 8 meters, depending
`upon whether radiologic, surgical, or autopsy measure-
`ments are made.1–5 Short bowel syndrome (SBS) occurs
`when there is ⬍200 cm of bowel remaining. This is an
`approximate length as most methods of residual intestine
`measurement (such as radiologic contrast studies, pathol-
`ogy of the resected specimen, and perioperative measure-
`ment of unweighted intestine) are not especially accu-
`rate. Because absorption is related to the amount of
`residual intestine, it is more important to document the
`amount of remaining, viable intestine.
`Those patients at greatest nutritional risk generally
`have a duodenostomy or a jejunoileal anastomosis with
`⬍35 cm of residual small intestine, jejunocolic or ileo-
`coloic anastomosis with ⬍60 cm of residual small intes-
`tine, or an end jejunostomy with ⬍115 cm of residual
`small intestine.6 – 8 It has been suggested that intestinal
`failure is better defined in terms of fecal energy loss
`rather than residual bowel length.9 Given the observa-
`tions that fecal energy loss does not always correlate well
`with residual bowel length,9 and the significant individ-
`ual variability in jejunal absorption efficiency,10 it is
`reasonable to consider a more standardized approach to
`defining intestinal failure and “functional” SBS from a
`clinical standpoint. However, fecal energy loss is a func-
`tion of both energy intake and energy absorption. Pa-
`tients who are unable to increase their oral intake suffi-
`ciently or are unable to absorb sufficient energy despite
`significantly increased intake, are defined as patients
`with intestinal failure and require parenteral nutrition
`support. A standardized diet may be useful for clinically
`defining functional SBS, although there is insufficient
`data with regard to what the composition of such a diet
`optimally should be.
`
`Methods
`Most available data on the treatment of SBS are
`based on retrospective analyses of case series (type II-3 or
`type III data) and are often few in number, because of the
`rareness of the covered diseases, where randomized, con-
`
`trolled trials were undertaken (type 1 and type IIb data),
`and the studies are described in detail. Data and reports
`were obtained from extensive PubMed and Medline
`searches using several key words, including SBS, various
`conditions predisposing to SBS, parenteral nutrition, enteral
`nutrition, relevant specific nutritional deficiencies, intesti-
`nal surgery, and intestinal transplantation. In addition,
`surgical and gastroenterological texts, published national
`and international scientific meeting abstracts, and the ex-
`tensive manuscript/abstract files of the authors were re-
`viewed. Expert opinion was sought for the few areas in
`which no suitable published reports existed (e.g., TPN
`cycling and preparation of the patient for home TPN).
`Human data and reports were reviewed exclusively.
`Patients with functional SBS who have severe malab-
`sorptive processes related to refractory sprue, chronic
`intestinal pseudo-obstruction syndrome, or congenital
`villus hypoplasia are not the specific focus of this tech-
`nical review, although most of the medical and nutri-
`tional management problems and therapies are similar, if
`not identical.
`
`Incidence and Prevalence of Short
`Bowel Syndrome
`It is unclear how many individuals in the USA
`suffer from SBS, but based on the numbers in Europe, the
`incidence may be ⯝2 per million.11 More recent data
`from 1993 indicated the incidence and prevalence of
`home parenteral nutrition, for which SBS was the most
`prevalent indication, increased slightly to 2–3 per year
`
`Abbreviations used in this paper: CMV, cytomegalovirus; CTP, Child-
`Turcotte-Pugh; CVC, central venous catheter; ESLD, end-stage liver
`disease; GLP-I, glucagon-like peptide I; ITR, Intestinal Transplant Reg-
`istry; IVC, inferior venous catheter; LCT, long-chain triglyceride; LILT,
`longitudinal intestinal lengthening and tailoring; MCT, medium-chain
`triglyceride; ORS, oral rehydration solution; SBS, short bowel syn-
`drome; SCFA, short-chain fatty acid; SRSB, segmental reversed small
`bowel; SVC, superior venous catheter; UNOS, United Network of Organ
`Sharing.
`© 2003 by the American Gastroenterological Association
`0016-5085/03/$30.00
`doi:10.1053/gast.2003.50139
`
`Page 1
`
`

`
`1112 AMERICAN GASTROENTEROLOGICAL ASSOCIATION
`
`GASTROENTEROLOGY Vol. 124, No. 4
`
`per million inhabitants and 4 per year per million,
`respectively.12,13 The most recent European survey, in
`1997, indicated the incidence of home TPN increased
`slightly to ⯝3 per million and the prevalence had in-
`creased to 4 per million.14 SBS constituted the largest
`single group of patients who required home TPN (35%).
`In comparison, the most recent data for incidence and
`prevalence in the USA is from 1992. At that time, it was
`estimated based on extrapolated data from the Oley
`Foundation Home TPN Registry that ⯝40,000 patients
`required TPN each year.15 Approximately 26% of the
`patients in the Oley registry had SBS, although some
`patients with a primary TPN indication of malignancy or
`radiation enteritis may have had SBS as well. These
`numbers for either Europe or the USA do not reflect
`patients with SBS who never required TPN or for whom
`TPN could be discontinued successfully. Approximately
`50%–70% of the short bowel patients who initially
`require TPN can be weaned off TPN successfully in
`optimal settings with better outcomes in children.6,16
`Therefore, the number of patients with SBS may be
`substantially greater than previously estimated. A regis-
`try of short bowel patients, including those who require
`TPN permanently, transiently, and not at all, should be
`implemented.
`
`Pathophysiology of Short Bowel
`Syndrome
`SBS may be a congenital or acquired condition.
`Infants born with intestinal atresia (jejunal or ileal) consti-
`tute the congenital forms. Otherwise, SBS results from
`surgical resection of bowel. This is usually related to mul-
`tiple resections for recurrent Crohn’s disease, massive enter-
`ectomy made necessary because of a catastrophic vascular
`event (such as a mesenteric arterial embolism or venous
`thrombosis, volvulus, trauma, or tumor resection in adults,
`and in children, gastroschisis, necrotizing enterocolitis), in-
`testinal atresias, and extensive aganglionosis. Functional
`SBS may also occur in cases of severe malabsorption where
`the bowel length is often intact. Such conditions may
`include chronic intestinal pseudo-obstruction syndrome, re-
`fractory sprue, radiation enteritis, or congenital villus atro-
`phy. Severe nutrient and fluid malabsorption occurs follow-
`ing extensive small intestinal resection. Patients with ⬍100
`cm of jejunum remaining generally have a net secretory
`response to food.17
`Patients can be grouped into 2 distinct subgroups:
`those with intact colon in continuity and those without
`colon in continuity. In patients with SBS, the colon
`becomes an important digestive organ. The colon absorbs
`sodium, water, and energy from short-chain fatty acids
`
`(SCFAs) (see below discussion regarding soluble dietary
`fiber).8,18,19
`
`How Does the Remaining Intestine
`Adapt Following Resection?
`Patients often clinically adapt to the significantly
`reduced energy absorption associated with SBS through
`hyperphagia. However, the intestine adapts as well to
`ensure more efficient absorption per unit length. After
`massive enterectomy, the intestine hypertrophies and
`becomes more efficient in nutrient absorption; there is
`slight lengthening, but more importantly, diameter and
`villus height increase, effectively increasing the absorp-
`tive surface.20 –23 This process may evolve over 1 or 2
`years.6,8,24 Several factors are important determinants in
`the functional adaptation process and clinical out-
`come.6,25,26 These include the presence or absence of the
`colon and ileocecal valve, length of remaining bowel,
`health of the remaining bowel, patient age, and co-
`morbid conditions. Although the length of remaining
`bowel necessary to prevent dependence on TPN is ⯝100
`cm in the absence of an intact and functional colon or 60
`cm in the presence of a completely functional colon,6,8
`the degree of adaptation and TPN dependence may
`be highly individualized. In infants, adaptation to full
`enteral nutrition has been reported with as little as 10 cm
`of residual intestine.24 However, Carbonnel et al. found
`small bowel length, determined radiographically, to be
`an independent risk factor for loss of nutritional auton-
`omy in 103 patients, of which 24 became TPN depen-
`dent.7 In addition, those with a jejunostomy were at
`increased risk for TPN dependence and those with a
`jejunal-ileal anastomosis were at decreased risk. Patients
`with active Crohn’s disease, radiation enteritis, carci-
`noma, or pseudo-obstruction involving their remaining
`bowel will have a blunted adaptation response.
`Animal models of SBS have suggested several gut
`hormones are involved in postresection intestinal adap-
`tation. These include enteroglucagon, glucagon peptide
`II, epidermal growth factor, growth hormone, cholecys-
`tokinin, gastrin, insulin, and neurotensin.27 There is
`little data on the role of either endogenous or exogenous
`hormones on intestinal adaptation in humans.
`Despite the fact that, normally, most nutrients are
`absorbed in the proximal jejunum, the residual ileum is
`able to adapt and to assume the role of macronutrient
`absorption. However, the specialized cells of the terminal
`ileum, where vitamin B12 /intrinsic factor receptors are
`located and where bile salts are reabsorbed, cannot be
`replaced by jejunal hypertrophy.
`
`Page 2
`
`

`
`April 2003
`
`AMERICAN GASTROENTEROLOGICAL ASSOCIATION 1113
`
`Effects of Massive Enterectomy on
`Gastrointestinal Motility and Transit Time
`
`Table 1. Dietary Macronutrient Recommendations for Short
`Bowel Syndrome
`
`Following small intestinal resection, dysmotility
`may develop, which may predispose to bacterial over-
`growth in the residual intestine. In addition, resection of
`the ileocecal valve allows colonic bacteria to enter and
`populate the small intestine.28 Bacterial overgrowth may
`negatively impact on digestion and nutrient assimila-
`tion, as bacteria compete for nutrients with the entero-
`cytes. Diagnosis may be more difficult using breath tests
`because of more rapid intestinal transit in short bowel
`patients. Endoscopically obtained small bowel aspirate
`for culture may be required. Treatment can be under-
`taken with oral metronidazole, tetracycline, or other
`antibiotics.
`Following jejunal resection, gastric emptying of liq-
`uids is more rapid, although intestinal transit may still
`remain normal because of the braking effect of the il-
`eum.29 Gastric emptying is significantly slower in pa-
`tients who have residual colon in continuity and is
`similar to normal controls. The loss of inhibition on
`gastric emptying and intestinal transit in patients with-
`out colon is related to a significant decrease in peptide
`YY (PYY), glucagon-like peptide I (GLP-I), and neuro-
`tensin.30 PYY is normally released from L cells in the
`ileum and colon when stimulated by fat or bile salts.
`Obviously, these cells are missing in patients who had
`distal ileal and colonic resection. Those patients with the
`shortest residual jejunum (⬍100 cm residual) exhibit the
`most rapid liquid gastric emptying.29 Solid emptying
`may also be more rapid in these patients. Rapid gastric
`emptying may contribute to fluid losses in patients with
`SBS.
`
`Medical Therapy of Short Bowel
`Syndrome
`The goal of medical therapy is for the patient to
`resume work and a normal lifestyle, or as normal of one
`as possible. This is undertaken via the use of specific
`measures to gradually decrease the requirement for TPN,
`and at best, to eliminate its need. The most important
`aspects of the medical management of the patient with
`SBS are to provide adequate nutrition, including both
`macro- and micronutrients (to prevent energy malnutri-
`tion and specific nutrient deficiencies), to provide suffi-
`cient fluid (to prevent dehydration), and to correct and
`prevent acid-based disturbances. Most macronutrients,
`including carbohydrate, nitrogen, and fat, are absorbed
`within the first 100 cm, and up to 150 cm of jejunum.31
`
`Colon present
`
`Colon absent
`
`Variable
`30–35 kcal/kg per day
`
`Carbohydrate Complex carbohydrate
`30–35 kcal/kg per day
`Soluble fiber
`LCT
`MCT/LCT
`20%–30% of caloric
`20%–30% of caloric
`intake
`intake
`⫾ low fat/high fat
`⫾ low fat/high fat
`Intact protein
`Intact protein
`1.0–1.5 g/kg per day
`1.0–1.5 g/kg per day
`⫾ peptide-based formula ⫾ peptide-based formula
`
`Fat
`
`Protein
`
`LCT, long-chain triglyceride; MCT, medium-chain triglyceride.
`
`Macronutrient Assimilation and Dietary
`Therapy
`Typically, patients who have undergone massive
`enterectomy require TPN for the first 7–10 days. Nu-
`tritional therapy should not be introduced until the
`patient is hemodynamically stable and fluid management
`issues are relatively stable. The goal is to provide patients
`with ⯝25–35 kcal/kg per day depending upon whether
`nutritional support is for maintenence or correction of
`undernutrition and 1.0 –1.5 kg per day of protein (Table
`1). Additional energy and protein are required by chil-
`dren, especially infants and neonates. Some debate exists
`whether the patient’s actual body weight or ideal body
`weight should be used in this calculation. For the post-
`operative patient, standard enteral formula is recom-
`mended. These should be instituted gradually as toler-
`ated. Once patients are able to eat, they should be
`encouraged to eat a regular diet, but modified as de-
`scribed below. There is no value in separating liquids
`from solids in the diet. Such practices have no effect on
`macronutrient, electrolyte or mineral absorption, fecal
`volume, or fecal weight.18
`Proteins and amino acids. Dietary protein is first
`digested, and then absorbed as dipeptides and tripep-
`tides. Therefore, it was reasoned that dietary protein
`provided in a predigested form would be more readily
`absorbed. However, nitrogen absorption is the macronu-
`trient least affected by the decreased intestinal absorptive
`surface. Therefore, the utility of peptide-based diets in
`such patients is generally without merit. McIntyre et al.
`compared energy, nitrogen, and fat absorption, as well as
`stool weight in 7 patients, all with end-jejunostomy and
`⬍150 cm (range, 60 –150 cm) of remaining small intes-
`tine. These patients were fed with either a peptide-based
`or an essentially isocaloric and isonitrogenous polymeric
`formula. Although the study was small, no differences
`were observed in energy, nitrogen, fat, carbohydrate,
`
`Page 3
`
`

`
`1114 AMERICAN GASTROENTEROLOGICAL ASSOCIATION
`
`GASTROENTEROLOGY Vol. 124, No. 4
`
`electrolyte, mineral, or fluid absorption.32 Uncontrolled
`data from Levy et al. supports these findings.33 However,
`in a small study of 6 patients, all with 90 –150 cm of
`residual jejunum and end-jejunostomy, data from Cosnes
`et al. suggests nitrogen absorption may be improved
`with the use of a peptide-based diet. Energy, other
`macronutrient, electrolyte, mineral, and fluid absorption
`was unaffected.34 Therefore, the clinical effect of the
`modestly increased nitrogen absorption was insignifi-
`cant. It must be recognized that all the studies described
`above were very small, the study populations somewhat
`heterogeneous, the various peptide constituents and their
`concentrations in these various formulas differed signif-
`icantly, and there was variation in the type and amount
`of fat (long-chain triglycerides [LCTs] versus medium-
`chain triglycerides [MCTs]). It is therefore difficult to
`make definitive comparisons between studies.
`The amino acid glutamine, together with glucose, is
`the preferred fuel for the small intestinal enterocyte.35
`Rodent TPN models suggested that both parenteral or
`enteral glutamine supplements could effect more rapid
`and significant bowel adaptation following massive en-
`terectomy.36,37 Therefore, it was thought glutamine sup-
`plementation in humans would have a similar effect.
`Although an early case series of 10 patients suggested
`that glutamine, combined with growth hormone supple-
`mentation and a high complex carbohydrate diet could
`result in decreased stool output and increased absorption
`of energy, protein, carbohydrate, sodium, and water,38 2
`subsequent double-blinded,
`randomized placebo-con-
`trolled trials failed to confirm any of these effects.39,40 In
`addition, Scolapio et al. showed that glutamine and
`growth hormone supplementation did not lead to mor-
`phological changes in the intestine.39 Glutamine-supple-
`mented oral rehydration solution (ORS) (see fluid and
`electrolyte management) was associated with decreased
`Na absorption and a trend toward decreased fluid absorp-
`tion in a small controlled trial in 6 patients.40,41 All of
`the patients studied by Byrne et al. had colon in conti-
`nuity38; it is likely the treatment-associated increase in
`energy absorption was related solely to the increased
`complex carbohydrate diet discussed above. Treatment
`with growth hormone and glutamine in the setting of
`SBS has been associated with significantly increased ex-
`tracellular fluid and peripheral edema.39,42,43 Therefore,
`treatment with glutamine and growth hormone cannot
`be recommended.
`Lipid. Luminal digestion of lipid may be im-
`paired because of impaired bile salt reabsorption related
`to resected ileum (⬎100 cm).44 Therefore, treatment
`with ox bile supplements has been attempted in 3 cases
`
`to increase the duodenal bile salt concentration to a
`concentration greater than the level at which micellar
`solubilization of lipid occurs.45– 47 Unfortunately, this
`therapy has been associated with significantly increased
`fecal volume, at least in those patients with intact colon.
`A preliminary, open-labeled study of 4 patients (2 with
`colon in continuity) indicated treatment with the con-
`jugated bile acid cholylsarcosine (6 g/day) was associated
`with an increase in fat absorption of 17 ⫾ 3 g/day
`without any effect on stool wet weight.48 As a conjugated
`bile acid, cholylsarcosine is resistant to colonic bacterial
`deconjugation, although 1 of the 4 patients did experi-
`ence a significant increase in wet stool output and an-
`other experienced nausea. Cholestyramine is not useful in
`patients with ⬎100 cm of
`ileal resection, and may
`actually worsen steatorrhea because of the binding of
`dietary lipid.49
`Although dietary fat restriction may result in in-
`creased fecal fat losses, there is no difference in the
`percentage of fat absorbed between high fat (75% non-
`protein calories derived from fat)/low carbohydrate and
`low fat/high carbohydrate, isocaloric and isonitrogenous
`diets.50 In addition, stool weight did not differ between
`diets. Because fat is energy-dense (9.0 kcal/g) when
`compared to carbohydrate (4.0 kcal/g), fat restriction
`may ultimately deprive the patient of a necessary source
`of energy. Up to 65% of dietary carbohydrate may be
`malabsorbed and lost in the feces without degradation by
`colonic bacteria.31
`The colon also absorbs MCTs (C8-C10), possibly re-
`lated to the fact that MCTs are water soluble. In a study
`of 10 short bowel patients with colon in continuity and
`9 patients with no residual colon, [randomized in a
`cross-over design to consume an LCT diet based on
`ordinary dietary fat, consisting of 20% carbohydrate,
`24% protein, and 56% fat versus an MCT-LCT diet,
`where 50% of the LCT was replaced with MCT (marga-
`rine, MCT oil)],51 those patients with intact colon ab-
`sorbed 96% ⫾ 3% of C8 and 87% ⫾ 6% of C10, versus
`63% ⫾ 25% for C8 and 57% ⫾ 28% for C10, respec-
`tively, in patients with no residual colon (P ⫽ 0.007 for
`C8 and P ⫽ 0.004 for C10) from the mixed LCT-MCT
`diet. Significantly increased energy absorption (⯝2.1
`MJ/day; 500 kcal/day) was found in patients with colon,
`but the LCT-MCT diet did not result in increased energy
`absorption when compared to the LCT diet in patients
`with an end-jejunostomy or ileostomy whose fecal output
`was also increased. MCT contain 8.3 kcal/g. Some, but
`not all, LCT can be replaced by MCT in the diet. In a
`short bowel patient eating 10.5 MJ/day (2500 kcal/day),
`⯝1.5–3 MJ/day (360 –720 kcal/day; 40 – 80 g) of LCT
`
`Page 4
`
`

`
`April 2003
`
`AMERICAN GASTROENTEROLOGICAL ASSOCIATION 1115
`
`can be replaced with MCT. However, LCTs are still
`necessary to provide essential fatty acids, and primarily
`linoleic fatty acid, which is not found in MCTs. In
`addition, excessive intake of MCT may result in nausea,
`vomiting, and ketosis.
`Carbohydrates. Rarely is the proximal jejunum
`resected in patients who require massive enterectomy.
`Because most intestinal disaccharidases are present in
`highest concentration proximally, it would stand to rea-
`son such patients would be unlikely to benefit from a
`lactose-free diet. Marteau et al. studied 14 short bowel
`patients in whom a lactose-free diet was compared to a
`diet containing 20 g/day containing ⱕ4 g milk.52 Lac-
`tose absorption, breath hydrogen, subjective symptoms
`of flatulence, and diarrhea were similar regardless of
`which diet was consumed. This data confirmed the find-
`ings of an earlier controlled study in 17 short bowel
`patients where it was also reported that lactose absorp-
`tion was enhanced when provided in yogurt rather than
`via milk.53 Regardless, in the absence of significant je-
`junal resection, lactose should not be restricted in the
`diet of the short bowel patient. The amount of lactose
`found in a glass of milk (20 –25 g) is generally well
`tolerated even in patients with an end-jejunostomy.53
`Because most lactose is found in milk-based foodstuff,
`which are also the most important source of dietary
`calcium, dietary lactose restriction will result in insuffi-
`cient dietary calcium intake.
`The role of soluble fiber. Soluble nonstarch poly-
`saccharides and some starches54 are not generally ab-
`sorbed by the small intestine. Soluble fiber is water
`soluble and found primarily in the following (in descend-
`ing order of concentration): oatmeal, oat bran, psyllium
`(Metamucil, Procter and Gamble, Cincinnati, OH; Kon-
`syl, Konsyl Pharmaceuticals, Ft. Worth, TX), barley,
`artichokes, strawberries, legumes, prunes, grapefruit, and
`squash. Soluble fiber and starches pass undigested into
`the colon where colonic bacteria ferment them not only
`to hydrogen and methane, hence patient “gas” com-
`plaints, but also to SCFAs, including butyrate, propri-
`onate, and acetate. SCFAs are the preferred fuel for the
`colonocyte.55 Therefore, in the patient with SBS, the
`colon becomes an important machine for energy absorp-
`tion. Approximately 75 mmol of SCFA are produced
`from 10 g of unabsorbed carbohydrate.56 Patients with
`SBS, but intact colon in continuity were able to decrease
`fecal energy loss by 1.3–3.1 MJ/day (310 –740 kcal)
`when they were fed a diet consisting of 60% carbohy-
`drates.57 Colonic metabolism of unabsorbed carbohydrate
`was indicated by decreased fecal carbohydrate losses in
`the patients with colon in continuity. It is possible for an
`
`Table 2. Vitamin and Mineral Supplements for Patients
`With Short Bowel Syndrome
`
`Vitamin A
`Vitamin B12
`
`Vitamin C
`Vitamin D
`
`Vitamin E
`Vitamin K
`Calcium
`Magnesium
`Iron
`Selenium
`Zinc
`Bicarbonate
`
`10000–50000 units dailya
`300 ␮g subcutaneously monthly for those
`w/ terminal ileal resections or disease
`200–500 mg
`1600 units DHT daily; may require 25-OH-
`or 1,23 (OH2)-D3
`30 IU daily
`10 mg weekly
`See text
`See text
`As needed
`60–100 ␮g daily
`220–440 mg daily (sulfate form)
`As needed
`
`NOTE. The table lists rough guidelines only. Vitamin and mineral
`supplementation must be monitored routinely and tailored to the
`individual patient, because relative absorption and requirements may
`vary.
`aUse cautiously in patients with cholestatic liver disease.
`
`intact colon to absorb up to 2.2– 4.9 MJ (525–1170 kcal)
`daily from dietary fiber.8,9,57 Colonic energy absorption
`may also increase somewhat during the postresection
`adaptation phase, related to increased colonic bacterial
`carbohydrate fermentation.10,58 This may be related to
`increased colonic bacteria in patients with SBS as well as
`an increase in the concentration or activity of various
`enzymes, such as ␤-galactosidase, over time during the
`adaptation period.58 Because SCFAs stimulate sodium
`and water absorption,42 patients might be expected to
`experience decreased fecal fluid and sodium loss, but this
`has not been observed clinically.57
`Vitamins. Micronutrients often require supple-
`mentation (Table 2). Because water-soluble vitamins are
`absorbed in the proximal jejunum, it is unusual for
`deficiencies to develop in short bowel patients (except in
`those who have high jejunostomies or duodenostomies),
`although these patients generally require TPN. Thia-
`mine deficiency has been reported and became an impor-
`tant issue during a recent parenteral vitamin shortage.59
`Patients have presented with Wernicke’s encephalopa-
`thy, beriberi, and severe metabolic alkalosis.60 If thia-
`mine deficiency is suspected, whole blood thiamine con-
`centration is not helpful; this reflects recent nutritional
`intake. Erythrocyte transketolase activity should be de-
`termined and empiric therapy begun with 100 mg of
`parenteral thiamine daily. Biotin deficiency has rarely
`been reported in patients with SBS.61 It is manifested in
`a scaly dermatitis, alopecia, lethargy, hypotonia, and
`lactic acidosis. Therapy consists of parenteral biotin sup-
`plementation of 0.3–1 mg daily, although this is not
`currently commercially available. Vitamin B12 supple-
`mentation is required (300 ␮g/month SQ) in patients
`
`Page 5
`
`

`
`1116 AMERICAN GASTROENTEROLOGICAL ASSOCIATION
`
`GASTROENTEROLOGY Vol. 124, No. 4
`
`who have had a significant portion of their terminal
`ileum resected (⬎60 cm).62 Folic acid is provided as a
`constituent of parenteral multivitamins. However, in
`patients with proximal jejunal resections, folate defi-
`ciency may develop.63 Such patients should receive 1
`mg/day supplement.
`Fat-soluble vitamin deficiency (A, D, and E) is more
`common because of the steatorrhea that occurs in SBS
`and the subsequent decrease in micellar formation and fat
`digestion.64 The use of cholestyramine may also result in
`fat-soluble vitamin deficiency.65 Night blindness and
`xerophthalmia has been described in SBS.66 As vitamin A
`deficiency progresses, corneal ulceration and permanent
`visual loss may ensue, and short bowel patients who do
`not receive parenteral multivitamins should have their
`serum vitamin A concentration monitored. If a low-
`serum vitamin A concentration is detected, therapy is
`10,000 –50,000 units daily, and may be administered
`either orally or parenterally.
`Vitamin D deficiency manifests in osteomalacia. Usu-
`ally dietary intake is a relatively unimportant source of
`vitamin D because the majority is endogenously synthe-
`sized from 7-dehydrocholesterol via ultraviolet light.67,68
`However, because enterohepatic circulation is disrupted
`in patients who have undergone significant ileal resec-
`tions, deficiency may result.69
`Vitamin E deficiency in patients with SBS may manifest
`in hemolysis70 and various neurological deficits.71 Because
`serum vitamin E concentration reflects serum total lipid
`concentration, which may be low in short bowel patients, a
`low serum vitamin E concentration alone may not be in-
`dicative of a deficient state; the ratio of serum vitamin E to
`total lipid should be calculated.72,73
`Vitamin K is synthesized by colonic bacteria (60%),74
`although dietary intake accounts for about 40% of re-
`quirements; deficiency is therefore uncommon in pa-
`tients with intact colon. However, vitamin K deficiency
`is frequent in patients who have no residual colon or have
`been given recent broad-spectrum antibiotics. Require-
`ment is ⯝1 mg daily.74 Vitamin K only recently has
`been a constituent in adult multivitamins for TPN,
`although pediatric multivitamin formulations all contain
`vitamin K.
`Trace metals. Patients with SBS lose a significant
`amount of zinc and selenium in their feces. A significant
`amount of zinc is lost in small bowel effluent (12 mg/L
`small intestinal fluid and 16 mg/L stool).75 Zinc defi-
`ciency has been associated with growth abnormalities,76
`delayed wound healing,77 and cellular immunity dys-
`function.78 Patients in whom zinc deficiency is suspected
`should be treated empirically with oral zinc sulfate
`
`(220 – 440 mg daily) or parenteral zinc if the patient
`requires TPN. Serum and leukocyte measurements of
`zinc concentration, although helpful, may be unreli-
`able.79 Selenium deficiency has been associated with car-
`diomyopathy,80 peripheral neuropathy, proximal muscle
`weakness and pain,81 whitening of the hair, and macro-
`cytosis.82 Serum selenium is a reliable indicator of sele-
`nium status, and if low, oral or parenteral supplementa-
`tion should be provided. Although there are 3 reported
`possible cases of chromium deficiency in patients requir-
`ing long-term TPN,83 deficiency has not been reported
`in short bowel patients who do not require TPN; there-
`fore,
`routine supplementation is not
`recommended.
`Chromium is a necessary cofactor for insulin’s effects in
`peripheral tissue.84 Even as TPN is a concern, available
`evidence suggests there is sufficient chromium present in
`the TPN solutions as a contaminant, and supplemental
`chromium may invite the possibility of nephrotoxicity.85
`Copper deficiency is very rare in the patient with SBS.
`Deficiency may result in microcytic anemia, neuropathy,
`and decreased fertility.86
`
`Medication Absorption
`The provision of medications to the patient with
`SBS can represent a challenge to the practicing clinician.
`Just as fluid and nutrient absorption is impaired, medi-
`cation absorption is often impaired as well. As with
`nutrient absorption, significant interpatient variability
`may be observed. Given that the risk for catheter sepsis
`is greater the more times the line is manipulated, it is
`important to use the oral or enteral route for medication
`delivery whenever possible. The degree to which a med-
`ication is malabsorbed is dependent upon several vari-
`ables. These include the surface area and health of the
`residual intestinal surface area, morphologic and physi-
`ologic factors, including the presence or absence of the
`terminal ileum (B12 and bile salt absorption—necessary
`for cyclosporin absorption), or the presence of an acidic or
`alkaline environment (related to the use of H2 blockers in
`TPN or use of proton pump inhibitors). Many, but not
`all, medications are absorbed in the jejunum; so, for
`many medications, absorption will be minimally im-
`pacted in the absence of decreased intestinal transit time,
`which will decrease mucosal contact time. Most of the
`available data on oral medication absorption in patients
`with SBS is in the form of isolated case reports.87
`
`Fluid and Electrolyte Management
`Massive enterectomy is associated with transient
`gastric hypersecretion. Basal acid secretion is signifi-
`cantly increased up to the first few months following
`resection.88,89 Massive small bowel resection is associated
`
`Page 6
`
`

`
`April 2003
`
`AMERICAN GASTROENTEROLOGICAL ASSOCIATION 1117
`
`with hypergastrinemia during the initial first 6 months
`after surgery.90 The H2 antagonists and proton pump
`inhibitors are useful in reducing gastric fluid secretion,
`and therefore will also reduce fluid losses during this
`period.17,91–94 However, absorption of orally dosed med-
`ications may be impaired, and either large doses, oral
`medication, or intravenous delivery may be required.
`Although fluid losses are decreased, macronutrient and
`electrolyte absorption are not affected by H2 antagonists
`and proton pump inhibitors.
`Fluid losses usually require chronic control with anti-
`motility agents, such as loperamide hydrochloride or
`diphenoxylate. Typical doses are 4 –16 mg/day. If these
`are ineffective, especially in patients with no colon in
`continuity or those who are left with a minimum of
`residual jejunum or duodenum, codeine sulfate or tinc-
`ture of opium may be necessary. The usual dose for
`codeine sulfate is 15– 60 mg two to three times a day.
`Rarely, patients will require treatment with octreotide.
`The mechanism of action is unclear, but octreotide may
`be useful to slow intestinal transit time and increase
`water and sodium absorption.95,96 In one open-labeled
`study of 9 patients with end-jejunostomies, daily jeju-
`nostomy volume was reduced from 8.1 ⫾ 1.8 to 4.8 ⫾
`0.7 L/day using a dose of 100 ␮g SQ, three times daily
`30 minutes before meals.97 Because use of octreotide does
`not lead to the d