Tube Feeding Improves Intestinal Absorption in Short Bowel Syndrome Patients

Transcription

1 GASTROENTEROLOGY 2009;136: Tube Feeding Improves Intestinal Absorption in Short Bowel Syndrome Patients FRANCISCA JOLY,* XAVIER DRAY, OLIVIER CORCOS,* LAURENCE BARBOT, NATHALIE KAPEL, and BERNARD MESSING* *Gastroenterology and Nutrition Support, Center for Intestinal Failure, Adult Approved Home Parenteral Nutrition Center of Paris-Ile de France, Pôle des Maladies de l Appareil Digestif (PMAD), AP-HP, Beaujon Hospital, Clichy, France; Department of Digestive Diseases, AP-HP, Lariboisière Hospital, Paris; and Coprologie fonctionnelle, Pitié Salpétrière Hospital, Paris, France Background & Aims: Tube feeding, recommended for patients with short bowel syndrome in only the postoperative period, has not been compared with oral feeding for absorption. We studied whether tube feeding increased absorption in patients with short bowel syndrome following the postoperative period. Methods: A randomized crossover study compared absorption between isocaloric tube feeding and oral feeding in 15 short bowel syndrome patients more than 3 months after short bowel constitution. An oral feeding period combined with enriched (1000 kcal day 1 ) tube feeding was also tested. We measured the net intestinal absorption rates of proteins, lipids, and total calories using elemental nitrogen, Van de Kamer, and bomb calorimetry methods, respectively. Results: Tube feeding increased the mean ( SD) percent absorption (P <.001) of proteins (72% 13% vs 57% 18%), lipids (69% 25% vs 41% 27%), and energy (82% 12% vs 65% 16%) compared with oral feeding. In the group given the combined feedings (n 9), the total enteral intake and net percent absorption increased (P <.001) for proteins (67% 10%), lipids (59% 19%), and total energy (75% 8%) compared with oral feeding. Absorption (kcal day 1 ) was greater (P <.001) with tube ( ) and combined feedings ( ) than with oral feeding ( ). Conclusions: In patients with short bowel syndrome, continuous tube feeding (exclusively or in conjunction with oral feeding) following the postoperative period significantly increased net absorption of lipids, proteins, and energy compared with oral feeding. Short bowel syndrome (SBS), a devastating clinical condition, was recently defined as follows: SBS results from surgical resection, congenital defect, or disease-associated loss of absorption and is characterized by the inability to maintain protein-energy, fluids, electrolytes, or micronutrient balances when on a conventionally accepted, normal diet. 1 Therefore, improving absorption is the cornerstone of therapy. 2 Indeed, after a period of 2 years in adult patients, the minimal length of postduodenal remnant small bowel required to maintain hydration and/or nutrient balances with a liquid diet and solid macronutrient hyperphagia to avoid home parenteral nutrition dependence was reported to be 115 cm in end jejunostomy (no remaining colon, ie, SBS type I), 65 cm in jejunocolonic anastomosis (ie, SBS type II), and 35 cm in ileocolonic anastomosis (ie, SBS type III). 3,4 According to a recent technical review, 5 dietary macronutrient recommendations after the postoperative period depend on the presence or not of the colon: patients with a jejunocolonic anastomosis should receive kcal/kg per day of complex carbohydrates with soluble fibers, with 20% 30% of caloric intake as fat in the form of medium-chain triglycerides (MCT) and long-chain triglycerides (LCT) with intact proteins ( g kg 1 day 1 ). In patients without a colon, there is no need for fiber supplementation, and LCT lipids alone are recommended. Tube feeding recommendations are for postoperative patients only: oral macronutrients are introduced gradually as tolerated, and, once patients can eat, tube feeding is stopped, and a regular diet should be encouraged. 5 Indeed, studies describing tube feeding in SBS patients are scarce. In one study 6 continuous tube feeding administered through gastrostomy, jejunostomy, or nasogastric tube in the early postoperative period (14 days) with polysaccharides, MCT, and protein hydrolysates mixed with a high viscosity tapioca suspension, was well tolerated and provided a high-calorie intake with a progressive decrease in fecal volume despite a steady increase in the infused volume. In that study, there was no data on absorption and no comparison with oral intake. 6 The only published comparative study between oral nutrition and tube feeding was a series of 5 patients with large ileal resection during the very early adaptive phase, 5 to 12 Abbreviations used in this paper: ETF, exclusive tube feeding; LCT, long chain triglycerides; MCT, medium chain triglycerides; OCEF, oral combined with enteral feeding; OF, oral feeding; SBS, short bowel syndrome; TF, tube feeding by the AGA Institute /09/$36.00 doi: /j.gastro

2 March 2009 TUBE FEEDING AND SHORT BOWEL SYNDROME 825 weeks after resection, where a 35% reduction in steatorrhea was shown during tube feeding with a nonelemental solution. 7 We also found a marked 40% improvement in macronutrient absorption during tube feeding compared with oral nutrition several months after short bowel constitution in a 50-year-old women who had undergone a vagotomy and with SBS (80 cm of jejunum plus 15% of remaining colon and a right colostomy) (B. Messing and F. Joly, unpublished personal data, 2007). In SBS without vagotomy, there are postprandial gastric and fasting and postprandial intestinal motor disturbances in the intestinal remnant, 8 which could worsen malabsorption and diarrhea. Thus, we hypothesized that continuous tube feeding could improve intestinal absorption in SBS compared with a regular oral diet after the early or late postoperative period, ie, more than 3 months after the constitution of SBS. A randomized crossover study was designed to test this hypothesis. Patients and Methods Patients Between January 1998 and June 2003, all consecutive SBS patients from our department, a tertiary care center for intestinal failure, were screened for the study. One to 2 weeks before enrollment, clinical status and biologic parameters were reviewed during screening to assess clinical stability. Routine biologic data included serum albumin, transthyretin (prealbumin), and C-reactive protein (CRP) levels. The urinary output per 24 hours was used to assess the hydration status. Stable clinical conditions were defined as no significant change in body weight; ie, tolerated change of no more than 2 kg per 2 weeks. A patient was not enrolled if she/he had undergone an additional digestive surgical intervention in the last 8 weeks or had had sepsis, whatever the origin, in the past 4 weeks. Criteria for inclusion were (1) extensive small bowel resection, with a remnant small bowel length 150 cm, with or without colonic resection; (2) an SBS constitution time of at least 3 months; and (3) a pattern of eating a Western solid diet ad libitum for more than 2 months that was at least the estimated basal energy expenditure (BEE) according to the Harris Benedict equations 9 using a normalized body weight to obtain a body mass index (BMI) equal to 25. Criteria for exclusion were (1) upper gastrointestinal tract surgery (esophagus, stomach, duodenum, and/or pancreas), (2) a remnant ileum, (3) small bowel lesions in the remnant gut (radiation, active Crohn s disease), (4) organ failure other than digestive, (5) evolutive neoplasia, (6) intestinal fistulae, and (7) treatment by growth hormone or other trophic factors. Rehabilitative surgery of the remnant small bowel, such as a distal reversed jejunal segment anastomosed to the remaining colon, was not exclusion criteria. The length of the postduodenal remnant small bowel was measured along the antimesenteric side of the small intestine by x-ray after a barium meal. The length of the colon, expressed as a percent of the usual length according to the method of Cummings et al, 10 was estimated from per operative data records. The study (No L03) was approved by the Human Investigations Committee of Saint-Lazare Hospital France. All patients gave informed consent. Study Design This prospective study compared intestinal macronutrient absorption in hospitalized patients in the last 3 days of two 7-day controlled regimens, after 4 days of confirming a digestive steady state. In a randomized crossover design, patients received 2 diet protocols: (1) 24-hour exclusive continuous exclusive tube feeding (ETF) alone and (2) free oral feeding (OF). The two 1-week study periods were separated by a washout home period of 1 to 4 weeks. Patients were offered a third optional diet protocol with oral combined with enteral feeding (OCEF) with the same duration washout period. Oral intake. The patients were asked to maintain their usual Western diet intake of carbohydrates, fat, protein, water, and fiber and not to change their drugs. All patients received transit slowing agents and gastric antisecretory drugs, but none received pancreatic enzyme treatment. Oral intake was unrestricted and included normal foods provided in continental meals. The major protein-containing foods included beef, fish, egg, chicken, cheese, and milk. Major fat sources included oil and butter. The dietary regimen was based on free solid food ingested with water recommendations according to the type of intestinal anastomosis. The glucose-polymerbased oral rehydration solution formula recommended by the World Health Organization was used by patients with end-jejunostomy. After screening, patients noted the type, and amount or weight, of foods they ingested at each meal including snacks and nibbling under the supervision of trained dieticians. During the 7-day study, the proposed meals (quantity and types of foods) reproduced each patient s usual intake. A similar oral diet was proposed for both oral periods (oral diet alone and oral supplement tube feeding). The daily intake of total calories, proteins, carbohydrates (total, complex, and simple), lipids, fiber, and alcohol were calculated with Bilnut software. 11 During the last 3 days of each controlled period, the absorption study was performed with ice collected of fecal or stoma output and measurement of intake. Atwater et al s calorie conversion factors were used and confirmed by Southgate and Durnin: 4.2, 9.35, and 5.65 kcal g 1 for carbohydrates, fat, and protein, respectively. 12 Our group has previously reported that intake measured by amount or weight was not significantly different from measurement by duplicated meals and left over in SBS patients. 13

3 826 JOLY ET AL GASTROENTEROLOGY Vol. 136, No. 3 Tube feeding. Tube feeding was administered by a pump device through a nasogastric tube inserted by a nurse. The position of the nasogastric tube was controlled by plain abdominal x-ray before use. Tube feeding was begun with a slow infusion rate (30 ml/h) and increased every 12 hours (30, 45, 60, 75 ml/h) to reach a caloric intake equivalent to the BEE multiplied by 1.5 after 48 hours. When an adequate and welltolerated level of tube feeding was reached, the infusion flow rate was kept constant per 24 hours. Patients were assessed at least 3 times a day by nurses for vomiting, tube dislocation, gastric residual volume, and flow rate provided by the pump. A standard formula of a polymeric solution containing 21% of proteins, 18% of LCT, 13% of MCT, and 48% of carbohydrates and no fiber was administered to all patients (Sondalis iso; Neslé, Vevey, Switzerland). In patients who agreed to participate in the third period (OCEF), tube feeding with the same polymeric diet was infused at a constant infusion rate of 42 ml/h providing 1000 kcal per day (1000 ml day 1 ) to enrich the oral feeding diet. Methods: Intestinal Macronutrient Absorption Stools or stoma output were ice collected and weighed daily during the 3-day metabolic periods as previously described. 13 Briefly, nitrogen, fat, and total energy were determined by elemental nitrogen analysis (N analyser Flash EA1112; Thermo Scientific, Waltham, MA) 14 and the method of Van de Kamer and bomb calorimetry (PARR 1351 bomb calorimeter; Parr Instrument Company, Moline, IL), 15,16 respectively. Quantification of carbohydrate-derived energy was calculated by subtracting the energy associated with the nitrogen and fat components from the total energy. 17 The coefficient of net intestinal absorption, expressed as a percentage of total energy ingested of the 3 main energy sources (fat, nitrogen, and carbohydrates), represented the proportion of ingested energy not recovered in stool output. Statistical Analysis Results are reported as medians, means, range, and SD. Nonparametric 2-tailed Wilcoxon matched-pairs signed-ranks tests were used to assess differences for continuous variables between treatment periods (ie, oral vs tube feeding, oral vs combined oral-tube feeding, and tube feeding vs combined oral-tube feeding) and to calculate the exact level of significance (threshold for significance set at.05). Linear regression equations and correlation coefficients between protein intake and either protein absorption or SBS characteristics were calculated for each metabolic period. Results Patients Eighteen patients were screened to participate in the study. Three patients were excluded: 2 because of active small bowel lesions related to either radiation or Crohn s disease and 1 case because a 30-cm remnant ileum was documented. Prior to enrollment, patients were clinically stable and had normal median range levels of serum albumin, 39 (34 47) g/l; prealbumin, 216 ( ) mg/l; and CRP, 3 (0.2 5) mg/l with a satisfactory urinary output of 1645 ( ) ml day 1. Fifteen patients completed the randomized study, ie, OF first for 7 patients and ETF first for the 8 others, and 9 of 15 accepted the third optional study period. The median time between the first and second study period and between the second and third study periods was 13 (7 28) days and 7 (7 20) days, respectively. Tube feeding was well tolerated in all cases with no infusion rate reduction, nausea, or vomiting. The first study period was performed 7 ( ) months following the last bowel surgery, 11 of 15 patients were studied after 6 (3.2 16) months, and 4 patients were studied after 84 (48 130) months. The clinical characteristics of the 15 patients are shown in Table 1. The BMI and BEE assessed with Harris and Benedict equations 9 using a normalized body weight to obtain a BMI equal to 25 were, respectively, 19.5 (15 29) and 1238 ( ) kcal day 1. Intake and Net Intestinal Absorption Table 2 shows the composition of intake for the 3 periods. Table 3 and Figure 1 show intakes and net absorption of total energy (expressed on kcal day 1 and kcal divided by BEE, fat (g day 1 ), protein (g day 1 ), and carbohydrates (g day 1 ) according to digestive status. Intake of OF, ETF, and OCEF was , , and kcal day 1, respectively. There was no greater variability either in the amount of total calories or in the amount of the 3 macronutrients in the exception of MCT. OCEF intake was significantly greater than in the 2 other study periods (P.001). No difference (P.5) was observed in fecal volume among OF ( ml day 1 ), ETF ( ml day 1 ), and OCEF ( ml day 1 ) metabolic periods. Net macronutrient absorption in percent (Figure 2) was significantly increased in ETF and OCEF vs OF for energy (P.001), lipids (P.001), and proteins (P.001) but not for carbohydrates (P.5). These changes corresponded, respectively, to a net intestinal absorption of and vs kcal day 1 (P.001) with a net protein absorption of and vs g kg 1 day 1 (P.001). No order effect was found between the 2 randomized periods for net absorption results (P.6).

4 March 2009 TUBE FEEDING AND SHORT BOWEL SYNDROME 827 Table 1. Characteristics of the 15 Short Bowel Patients Age(y)/sex Remnant jejunal length a (cm) Digestive type b /APJA c Remaining colon, % d Diagnosis Time elapsed since last surgery (mo) Case 1 f 44/F 25 JC/Y 86 Arterial mesenteric infarction 9 6 Case 2 f 66/F 30 JC/Y 86 Arterial mesenteric infarction 6 5 Case 3 f 53/M 35 JC/Y 58 Crohn s disease Case 4 43/M 65 JC/Y 86 Postoperative complications Case 5 40/M 40 JC/N 58 Aortic dissection 60 5 Case 6 f 66/F 70 JC/N 58 Postoperative complications Case 7 66/F 80 JC/N 58 Small bowel volvulus 48 3 Case 8 56/M 80 JC/N 58 Postoperative complications 6 5 Case 9 f 51/F 100 JC/N 42 Radiation enteritis 4 5 Case 10 66/M 120 JC/N 58 Venous mesenteric infarction 16 3 Case 11 f 66/F 120 JC/N 86 Radiation enteritis 7 7 Case 12 f 49/M 60 EE/N 0 Familial polyposis 4 7 Case 13 22/F 100 EE/N 0 Crohn s disease Case 14 f 65/F 120 EE/N 0 Radiation enteritis 7 7 g Case 15 f 51/F 130 EE/N 0 Arterial mesenteric infarction Median (range) 53 (22 66) 80 (25 130) 58 (0 86) 7.5 ( ) 5 (3 7) HPN e HPN, home parenteral nutrition; JC, jejuno-colonic anastomosis; EE, end-enterostomy; APJA, antiperistaltic jejunal segment anastomosed to the remaining colon; N, no; Y, yes. a Postduodenal small bowel length taken along the antimesenteric side of the small intestine on x-ray films of a barium meal. b The postduodenal remnant small bowel length was longer in patient with end-jejunostomy (P.012) and patients with jejunocolonic anastomosis without antiperistaltic segment (P.05) than in patients with jejuno-colonic anastomosis and antiperistaltic segment with NS difference in the 2 first groups. c APJA: antiperistaltic jejunal segment anastomosed to the remaining colon. d According to Cummings et al. 10 e Number of infusions per week. f Nine patients who accepted the third optional metabolic study. g Hydrolectrolytic infusions. When the anatomic SBS types were taken into account, no correlation was found between the net absorption of the 3 macronutrients and either the length of post duodenal remnant small bowel or the delay since last resection. We found a positive correlation between protein intake and net protein absorption (n 15) during OF (r 0.80, P.001) and ETF (r 0.83, P.0005) but not during OCEF (n 9). In patients studied between 3 and 24 months after the last surgery, the caloric gain in ETF vs OF and in OCEF vs OF was kcal day 1 (n 11) and kcal day 1 (n 8), respectively, which was in the same order of magnitude as in patients studied more than 24 months after the last surgery, ie, kcal day 1 (n 4) and 676 kcal day 1 (n 1), respectively (statistics not performed because of the small number of patients). Discussion This is the first study to demonstrate a significant increase in net macronutrient absorption during continuous tube feeding compared with OF in a heterogeneous representative series of 15 short bowel patients after the postoperative period. The net gains per day were clinically relevant and represented 280 kcal for lipids, 23 g for proteins, and 125 kcal for carbohydrates and a total energy gain of 700 kcal day 1. The latter gain corresponded to 33% of the total energy expenditure for SBS Table 2. Composition of Intake During the 3 Study Periods Exclusive continuous tube feeding Oral feeding Oral feeding combined with continuous tube feeding g/1000 kcal Percent g/1000 kcal Percent g/1000 kcal Percent Proteins Long-chain triglycerides Medium-chain triglycerides Carbohydrates Fibers NOTE. Composition of intake (proteins, LCT, MCT, carbohydrates, and fibers) during ETF, OF, and OCEF expressed in percent and in grams per 1000 kcal. Mean SD.

5 828 JOLY ET AL GASTROENTEROLOGY Vol. 136, No. 3 Table 3. Intake and Net Absorption of Macronutrients During the 3 Study Periods According to the Intestinal Status of Short Bowel Patients Intake Net absorption OF, n 15 ETF, n 15 OCEF, n 9 OF, n 15 ETF, n 15 OCEF, n 9 Remaining colon with APJA (n 4) a Energy (kcal day 1 ) Energy/BEE 1.5 b Lipid (g day 1 ) Proteins (g day 1 ) Carbohydrates (g day 1 ) Remaining colon without APJA (n 7) a Energy (kcal day 1 ) Energy/BEE 1.5 b Lipid (g day 1 ) Proteins (g day 1 ) Carbohydrates (g day 1 ) Sum patients with remaining colon (type II, n 11) Energy (kcal day 1 ) Energy/BEE 1.5 b Lipid (g day 1 ) Proteins (g day 1 ) Carbohydrates (g day 1 ) End-jejunostomy (type I, n 4) Energy (kcal day 1 ) Energy/BEE 1.5 b Lipid (g day 1 ) Proteins (g day 1 ) Carbohydrates (g day 1 ) All patients (type I and II) (n 15) Energy (kcal day 1 ) c d d Energy/BEE 1.5 b c d d Lipid (g day 1 ) c d d Proteins (g day 1 ) c d d Carbohydrates (g day 1 ) c NOTE. Values are mean SD. OF, oral feeding; ETF, exclusive tube feeding; OCEF, oral combined with tube feeding. b BEE: Basal energy expenditure assessed with Harris and Benedict equations using a normalized body weight to obtain a BMI equal to 25. Male: BEE (13.75 weight) (5 height) (6.76 age); female: BEE (9.56 weight) (1.85 height) (4.68 age) (weight in kilograms, height in centimeters, and age in years). a APJA: antiperistaltic jejunal segment anastomosed to the remaining colon. Statistical tests were performed for overall patients (sum of types I and II). Intake was significantly greater c (P.001) with OCEF than in OF and ETF. Macronutrient absorption was significantly increased d in ETF and OCEF (P.001) vs OF for energy, fat, nitrogen but not for carbohydrates. patients, with total energy expenditure estimated as calculated basal energy expenditure 1.5, a figure necessary to obtain an energy balance equilibrium in SBS patients. 13 In the 9 patients who participated in the combined oral and continuous tube feeding protocol, these gains were maintained compared with tube feeding alone, ie, 360 kcal for lipids, 27 g for proteins, and 486 kcal for carbohydrates with a total energy gain of 1036 kcal day 1. The latter gain corresponded to 57% of the total energy expenditure of our patients. Therefore, the increase in net absorption obtained with tube feeding alone was not blunted by combined oral plus tube feeding. The latter gain in digestive balance was due not only to a significantly increased absorption rate but also to a significant increase in total enteral intake. In this study, it is difficult to determine whether the significant absorptive gain of lipids and proteins induced by tube feeding was due to the continuous mode of enteral nutrition and/or to the change in the type of macronutrients used with tube feeding. It is therefore worth noting that the gain in macronutrient absorption was, in percent of intake, of the same order of magnitude during the 2 intervention periods. This suggests that (1) the number of calories administered during tube feeding was not the main factor affecting the increase in net absorption; (2) in the combined period, continuous tube feeding associated with an unchanged oral intake of nutrients improved absorption with no adverse effects; (3) the result in the latter group was an increase in total enteral intake suggesting that supplement tube feeding of 1000 kcal day 1 did not cause any break in oral intake in the very short term; (4) this result was significant because compared with OF alone, the net absorptive gain was equivalent to the number of kilocalories added

6 March 2009 TUBE FEEDING AND SHORT BOWEL SYNDROME 829 Figure 1. Net absorption for total (A) calories, (B) lipids, and (C) proteins during the 3 study periods. In the histograms, intakes (in light grey) and losses (in black) are above and below the zero line, respectively, the dark grey being the net absorption (intake losses). Total calories are expressed in kcal day 1 and lipids and proteins in grams day 1. *Total calories, lipids, and protein intakes, ie, light grey bars, were significantly higher with OCEF than with OF and ETF (P.001). Net absorption for total calories, lipids, and proteins, ie, dark grey bars, were significantly higher with ETF and OCEF than with OF (P.001). Mean SD. OF, oral feeding; ETF, exclusive tube feeding; OCEF, oral combined with tube feeding. through tube feeding. Indeed, even if the aim of this study was not to describe the mechanisms of improved absorption from tube feeding, these results suggest that continuous tube feeding accommodates the gastric and small intestinal motor disturbances related to SBS and clearly improves intestinal absorption. 6,18 20 Therefore, this study suggests that continuous constant rate mode of tube feeding was an important factor in the increased absorption vs oral feeding alone. This might be explained by a better distribution of calories in the remnant small bowel induced by the continuous mode of tube feeding instead of discontinuous OF. 21 Moreover, the increase in lipid absorption during OCEF was more than twice the percentage of MCT infused with tube feeding, eg, the MCT moiety was only 12% of fat intake, and we reported a 20% increase in fat absorption. Indeed, MCT oral supplementation is thought to improve fat absorption in SBS patients. 5 During the ETF study period, MCT intake was 30 (range, 25 40) g day 1 (39% of total lipid intake), a level that can be considered adequate. In a previous study, a similar level of oral intake did not result in an increase in stool loss or a decrease in the absorption of the 2 other macronutrients, especially in jejunostomy patients. 21 However, when the MCT intake was higher (50% oral replacement of LCT with 60 [range, 50 80) g day 1 of MCT), 50% absorption was reported in jejunostomy patients with an accompanying osmotic diarrhea, but an 85% absorption in jejunocolonic patients, explaining their potential benefit in energy balance. 22 Our study showed that, with the same level of protein intake, the net absorption of proteins was significantly higher during tube feeding alone compared with OF. In short bowel patients, protein absorption depends on remnant small bowel length and protein oral intake. 23 In this study, we did not find a significant correlation between small bowel length and protein absorption because of the different types of intestinal anastomosis (type I and type II with or without a jejunal antiperistaltic segment) in this small series of SBS patients. 24 However, we found a significant correlation between protein intake and protein absorption in the whole group. A standard polymeric solution was used because the advantage of providing (semi)elemental peptide-based solutions is offset by their high osmolarity and a poorer systemic metabolic use. 5,25 In addition, in SBS patients, the improvements in absorption with semielemental diets compared with isocaloric polymeric diets during tube feeding were not found during oral ingestion. 26 In the latter study, McIntyre et al found no difference in energy, nitrogen, fat, carbohydrate, or fluid absorption 27 during oral ingestion whatever the diet and concluded that patients with a high jejunostomy output do not need a special oral diet. Ingested whole casein proteins, delivered more slowly to the small bowel than lactoglobulin proteins, were better absorbed in jejunostomy patients. 28 These

7 830 JOLY ET AL GASTROENTEROLOGY Vol. 136, No. 3 Figure 2. Net absorption (in percent) of total calories, lipids, proteins, and carbohydrates during the 3 study periods. The histogram represents the percent net absorption of total calories, lipids, proteins, and carbohydrates, respectively, in black, light grey, dark grey, and white. The coefficient of net intestinal absorption, expressed as a percentage of total calories ingested and of the main energy sources (lipids, proteins, and carbohydrates), represents the proportion of ingested calories and of the main energy sources not recovered in stool output. * Macronutrient absorption in percent was significantly increased in ETF and OCEF vs OF, for total calories (P.001), lipids (P.001), and proteins (P.001). Mean SD. OF, oral feeding; ETF, exclusive tube feeding; OCEF, oral combined with tube feeding. data suggest that the continuous mode of delivery during enteral nutrition through tube feeding played a major role in improving protein and lipid absorption. No significant increase in carbohydrate absorption was found because patients already had a net high carbohydrate absorption with the oral diet, making it difficult to improve absorption compared with other diets. This high net percent absorption of carbohydrates remained unchanged when calorie intake was increased with OCEF inducing a significant increase in energy balance. Indeed, in small bowel patients, unabsorbed carbohydrates are fermented by colonic bacteria to short-chain fatty acids, which are absorbed by an important and well-known salvage energy mechanism Currently, tube feeding is only recommended in postoperative SBS patients. 5 This study clearly showed a significant improvement in net macronutrient absorption in percent after the postoperative period during the adaptive and postadaptive periods. These results were of the same order of magnitude whatever the anatomy of the SBS. No significant correlation was found between net absorption of the 3 macronutrients and the delay since last resection, suggesting that the positive effect on absorption of enteral feeding was not blunted by the adaptive process per se, a phenomenom that has not been directly proved, ie, a net increase in percent absorption, in the human literature. 23,32,33 Short bowel rehabilitation program emphasizes an appropriate diet and growth factors to enhance intestinal absorption and induce pharmacologic adaptation Only 1 of 3 controlled studies was positive with recombinant human growth hormone: there was a significant 15% 5% increase in net intestinal absorption with low-dose growth hormone alone and an unchanged hyperphagic diet in 12 jejunocolonic SBS patients, ie, energy gain was kcal day 1, which represented 19% 8% of the total (BEE 1.5) energy expenditure. 37 To date with glucagon-like peptide-2, a significantly decreased stool wet weight loss of 0.75 L day 1 has been reported (10 of 16 patients had a jejunostomy) with a nonsignificant increase in macronutrient absorption, ie, 154 kcal day If these results are compared with our tube feeding regimen, the energy net gain induced by tube feeding, was highly effective and represented 33% and 57% of the total (BEE 1.5) energy expenditure in ETF and OCEF, respectively, compared with OF. Thus, tube feeding could be suggested as a potentially effective medical therapy in intestinal rehabilitative program in SBS adult patients after the postoperative period. This approach is routinely used in pediatric patients to increase enteral tolerance and autonomy, 35 but, to our knowledge, results with tube feeding and OF have not been reported in this population. In conclusion, after the postoperative period, tube feeding alone or oral combined with supplement tube feeding was well tolerated in SBS patients and induced a significant and relevant increase in energy, lipids, and protein absorption compared with oral feeding. In our experience, enteral autonomy, ie, avoiding severe intestinal failure, 1 can be obtained when absorption is more than 1.5-fold the BEE 13 with an absorbed protein of more than 1 g kg 1 day In this study, 1 of 15 patients met the 2 latter criteria on OF and reached 8 of 15 on ETF and 7 of 9 on OCEF. Therefore, continuous tube feeding can be recommended in patients with a low-level of home parenteral nutrition dependence and in whom the expected gain with tube feeding could allow them to wean off home parenteral nutrition. References 1. O Keefe SJ, Buchman AL, Fishbein TM, et al. Short bowel syndrome and intestinal failure: consensus definitions and overview. Clin Gastroenterol Hepatol 2006;4: Jeejeebhoy KN. Short bowel syndrome: a nutritional and medical approach. CMAJ 2002;166: Messing B, Crenn P, Beau P, et al. Long-term survival and parenteral nutrition dependence in adult patients with the short bowel syndrome. Gastroenterology 1999;117: Carbonnel F, Cosnes J, Chevret S, et al. The role of anatomic factors in nutritional autonomy after extensive small bowel resection. J Parenter Enteral Nutr 1996;20: Buchman AL, Scolapio J, Fryer J. AGA technical review on short bowel syndrome and intestinal transplantation. Gastroenterology 2003;124: Levy E, Frileux P, Sandrucci S, et al. Continuous enteral nutrition during the early adaptive stage of the short bowel syndrome. Br J Surg 1988;75:

8 March 2009 TUBE FEEDING AND SHORT BOWEL SYNDROME Cosnes J, Parquet M, Gendre JP, et al. Continuous enteral feeding to reduce diarrhea and steatorrhea following ileal resection. Gastroenterol Clin Biol 1980;4: Schmidt T, Pfeiffer A, Hackelsberger N, et al. Effect of intestinal resection on human small bowel motility. Gut 1996;38: Harris JA, Benedict FG. A biometric study of human basal metabolism. Proc Natl Acad Sci U S A 1918;4: Cummings JH, James WP, Wiggins HS. Role of the colon in ileal-resection diarrhoea. Lancet 1973;1: Bourgerette P, Rolshansen M. BILNUT 4.0. SCDA Nutrisoft, Le Hallier, Cerellers, France, Southgate DA, Durnin JV. Calorie conversion factors. An experimental reassessment of the factors used in the calculation of the energy value of human diets. Br J Nutr 1970;24: Messing B, Pigot F, Rongier M, et al. Intestinal absorption of free oral hyperalimentation in the very short bowel syndrome. Gastroenterology 1991;100: Kapel N, Matarazzo P, Haouchine D, et al. Fecal tumor necrosis factor alpha, eosinophil cationic protein and IgE levels in infants with cow s milk allergy and gastrointestinal manifestations. Clin Chem Lab Med 1999;37: Miller DS, Payne PR. A ballistic bomb calorimeter. Br J Nutr 1959;13: Van de Kamer JH, Huihink H, Weyers HA. Rapid method for determination of fat in faeces. J Biol Chem 1949;177: Hamaker BR, Rivera K, Morales E, et al. Measurement of fecal carbohydrate in human metabolic balance studies: calculated versus determined. Nutr Res 1995;15: Nightingale JM, Kamm MA, van der Sijp JR, et al. Disturbed gastric emptying in the short bowel syndrome. Evidence for a colonic brake. Gut 1993;34: Remington M, Malagelada JR, Zinsmeister A, et al. Abnormalities in gastrointestinal motor activity in patients with short bowels: effect of a synthetic opiate. Gastroenterology 1983;85: Quigley EM, Thompson JS. The motor response to intestinal resection: motor activity in the canine small intestine following distal resection. Gastroenterology 1993;10: Quigley MM. Gastric motor and sensory function and motor disorders of the stomach. In: Feldman M, Friedman LS, Brandt LJ, eds. Sleisenger and Fordtran s Gastrointestinal and liver disease. Volume 1. 8th ed. Philadelphia: Saunders, 2006: Jeppesen PB, Mortensen PB. The influence of a preserved colon on the absorption of medium chain fat in patients with small bowel resection. Gut 1998;43: Crenn P, Morin MC, Joly F, et al. Net digestive absorption and adaptive hyperphagia in adult short bowel patients. Gut 2004; 53: Panis Y, Messing B, Rivet P, et al. Segmental reversal of the small bowel as an alternative to intestinal transplantation in patients with short bowel syndrome. Ann Surg 199;225: Boirie Y, Dangin M, Gachon P, et al. Slow and fast dietary proteins differently modulate postprandial protein accretion. Proc Natl Acad Sci U S A 1997;94: Cosnes J, Evard D, Beaugerie L, et al. Improvement in protein absorption with a small peptide-base diet in patients with high jejunostomy. Nutrition 1992;8: McIntyre PB, Fitchew M, Lennard-Jones JE. Patients with a high jejunostomy do not need a special diet. Gastroenterology 1986; 91: Mahé S, Messing B, Thuillier F, et al. Digestion of bovine milk proteins in patients with a high jejunostomy. Am J Clin Nutr 1991;54: Nordgaard I, Hansen BS, Mortensen PB. Colon as a digestive organ in patients with short bowel. Lancet 1994;343: Royall D, Wolever TM, Jeejeebhoy KN. Evidence for colonic conservation of malabsorbed carbohydrate in short bowel syndrome. Am J Gastroenterol 1992;87: Briet F, Flourie B, Achour L, et al. Bacterial adaptation in patients with short bowel and colon in continuity. Gastroenterology 1995; 109: Cosnes J, Carbonnel F, Beaugerie L, et al. Functional adaptation after extensive small bowel resection in humans. Eur J Gastroenterol Hepatol 1994;6: Powell TJ. Don t expect too much of intestinal adaptation in short bowel syndrome. Eur J Gastroenterol Hepatol 1994;6: DiBaise JK, Young RJ, Vanderhoof JA. Intestinal rehabilitation and the short bowel syndrome: part 2. Am J Gastroenterol 2004; 99: Torres C, Sudan D, Vanderhoof J, et al. Role of an intestinal rehabilitation program in the treatment of advanced intestinal failure. J Pediatr Gastroenterol Nutr 2007;45: Wilmore DW. Indications for specific therapy in the rehabilitation of patients with the short-bowel syndrome. Best Pract Res Clin Gastroenterol 2003;17: Seguy D, Vahedi K, Kapel N, et al. Low-dose growth hormone in adult home parenteral nutrition-dependent short bowel syndrome patients: a positive study. Gastroenterology 2003;124: Jeppesen PB, Sanguinetti EL, Buchman A, et al. Teduglutide (ALX-0600), a dipeptidyl peptidase IV resistant glucagon-like peptide 2 analogue, improves intestinal function in short bowel syndrome patients. Gut 2005;54: Received Dec 19, Accepted Oct 23, Reprint requests Address requests for reprints to: Francisca Joly, MD, Hôpital Beaujon-Service de Gastro-entérologie et Assistance Nutritive, 100 bd Général Leclerc-Clichy, France. francisca.joly@bjn.aphp.fr. Acknowledgments The authors thank the physicians, especially K. Vahedi and P. Crenn; the dieticians, M. C. Morin and S. Rousseau-Penven; and the nurses, especially I. Pingenot, for their assistance. Conflicts of interest The authors disclose no conflicts.