GASTROENTEROLOGY
1991;100:1502-1508
Intestinal Absorption of Free Oral Hyperalimentation in the Very Short Bowel Syndrome BERNARD MESSING, FRANCOIS PIGOT, MONIQUE RONGIER, MARIE CHRISTINE MORIN, URBAIN NDEiNDOUM, and JEAN CLAUDE RAMBAUD Institut
National
de la Sante
et de la Recherche
MBdicale
U 290, HBpital
Saint-Lazare,
Paris,
France
Ten adult ambulatory patients with the nonactive digestive disease short bowel syndrome were prospectively studied to quantitatively assess their free oral intake and their net digestive absorption of total calories, fat, protein, and carbohydrate during a &day period at least 6 months after a resection. The remaining portions of small bowel had a mean length of 75 cm (range, O-200 cm); the remaining colon lengths had a mean of 67% of normal (range, O%-100%). The experimental diets were formulated according to a home dietary inquiry. During the study period, pooled intakes and digestive losses were measured for total calories, fat, and protein using the bomb calorimetry, Van de Kamer, and Kjeldahl techniques, respectively. The ingested diet provided 58 + 14 kcal 1 kg-’ . day-’ (mean 2 SD) and consisted of 46% carbohydrate, 31% fat, and 23% protein. Net digestive absorption was 67% + 12% for 52% -+ total calories, 79% + 15% for carbohydrate, 16% for fat, and 61% + 19O/bfor protein. The larger net digestive absorption of carbohydrate (P 5 0.004) compared with fat and protein suggests salvage of colonic cholesterol in short bowel syndrome patients. It is concluded that these patients with the short bowel syndrome adapted to a hypercaloric, hyperprotein diet to compensate for increased fecal losses and that this hyperphagia does not seem to have impaired their net digestive absorption.
I
n patients with the short bowel syndrome (SBS), the optimal oral nutritional support has not yet been fully defined. Moreover, the problem of providing adequate nutritional support to SBS patients in clinical practice is increasingly encountered because improvements in surgical procedures and intensive care have increased the survival rates for patients who have undergone extensive small bowel resection (1).
Oral nutrition is necessary for intestinal adaptation after small bowel resection (2). Recent experiments in patients with SBS have given some new insight into this problem. McIntyre et al. (3) found that a semielemental diet did not absorb better than a polymeric diet in four jejunostomized patients, and Woolf et al. (4) showed that neither a high-fat nor a high-carbohydrate diet influenced the absorptive capacity of eight patients with SBS. Fluid restriction during meals also did not appear to change the absorptive capacity of macronutrients (5), and the replacement of 50%~75% of dietary long-chain triglycerides with mediumchain triglycerides led to a decrease in fecal loss of water and improvement in nutritional status of seven patients with SBS (6). Determination of the optimal oral nutritional support in patients with SBS should be approached through analysis of their absorptive capacity, which in fact has rarely been extensively studied ($5). Therefore, we performed a prospective study to evaluate macronutrient absorption in SBS patients under their usual and spontaneous oral alimentation conditions . Materials
and Methods
Patients All patients had undergone extensive small bowel resection with or without colonic resection. The length of the remaining gut structures was determined after analysis of peroperative and radiographic data, the length of the colon being expressed in terms of percent of the usual length according to Ihe method of Cummings et al. (7). Abbreviations used in this paper: PN, parenteral nutrition; SBS, short bowel syndrome. o 1991 by the American Gastroenterological Association 0018-5085/91/$3.00
June 1991
MACRONUTRIENT
Criteria for inclusion in the study included (a] last surgical intervention at least 6 months before the study; (b) no radiological, endoscopic, and/or histological evidence of structural abnormality on the residual intestine before the study; (c) no change in medical therapy (antibiotics, transitslowing agents, pancreatic enzymes, etc.) or commencement of oral supplementation with medium-chain triglycerides during the 3 months before the study; (d) no surgical procedures performed slowing digestive transit time; (e) no extradigestive organ failure; (f) an ambulatory life-style; and (g) consumption of a diet ad libitum with or without home parenteral nutrition (PN), the two being unchanged for a period of at least 3 months. The need to continue PN was assessed by the degree of deterioration of nutritional status during previous attempts to reduce or stop PN support. The protocol was approved by the Ethical Committee of Saint Lazare Hospital, and patients gave their informed consent. Study Protocol The prospective study lasted for a total of 5-6 days, which included a z-s-day equilibrium period and a s-day metabolic period. Patient evaluation upon admission included clinical examination with determination of the clinical nutritional status [stage 1, normal; stage 3, evidence of malnutrition; stage 2, not 1 or 3 (S)], and determination of the physical activity index during the previous month [0, normal; 1, subnormal; 2, daily diurnal rest ~3 hours; 3, daily diurnal rest 2 3 hours (9)]. Hematological analysis included determination of serum albumin, prealbumin, and C-reactive protein levels.
ABSORPTION
IN SHORT BOWEL SYNDROME
1503
ries. The calorie-conversion factors used were those of Atwater et al. confirmed by Southgate and Durnin (14): 4.20, 9.35, and 5.65 kcal/g for carbohydrate, fat, and protein, respectively. Lactose levels were measured using an enzymatic method (15). The basal energy expenditure was calculated according to the Harris-Benedict equations (16) using actual body weights. The net total caloric absorption was calculated by substracting the calories excreted in feces from the calories ingested. The coefficient of net digestive absorption expressed as a percent of total calories, and for the three main energy sources, i.e., carbohydrate, fat, and protein, represented the proportion of ingested calories nat recovered in fecal output. The metabolizable energy level was calculated by subtracting the number of calories excreted in feces and urine from those ingested and, if applicable, those calories provided by PN. Only stable patients (no body weight variation in the 3 months before the study) were assumed to be in an energybalance equilibrium. The quantity of energy metabolized by these patients was considered equivalent to their total energy expenditure. In this subset of patients, metabolizable energy needs for equilibrium were approximated in relation to their estimated basal energy expenditure (16). Results are expressed as means + SD and were tested for statistical significance using Student’s t test for unpaired and paired data. Coefficients of correlation between data were calculated using the least-squares method.
Results Pa tien ts
Diet During the study each patient’s diet was designed to closely match the usual spontaneous intake of calories, carbohydrate, fat, protein, lactose, and fiber as determined by a s-day dietary intake inquiry performed during the month before the study. Each diet was prepared in duplicate, the actual oral intake determined by the difference between the duplicate and dish-losses of the served meals. If the patient was receiving PN, the same protocol used during the 3 months before the study was used during the study. Methods Digestive transit time was estimated using the transit time of a colored marker (red carmine) (10) given orally with the first meal of the study period. The following 2-3 days, depending on the appearance of the red carmin in the stools, served as an equilibrium period, after which the metabolic period was begun. During this s-day metabolic period both urine and stools were collected daily, the latter being collected on ice and immediately frozen at -20°C. Oral-input and fecal-output analyses were performed on homogenized (Waren-Bolster) aliquots of s-day pooled samples. Fat, nitrogen, and total caloric content were measured by the Van de Kamer (ll), micro-Kjeldahl (12), and bomb calorimetry methods, respectively (13). The number of carbohydrate calories was calculated from the difference between total calories and fat plus protein calo-
The clinical features of the 10 patients (aged 47 + 15 years) selected for the study are given in Table 1. The mean time elapsed since the last resection was 35 2 31 months (range, 7-96 months). All
patients had intact stomachs and duodena; mean jejunal and ileal length was 67 ? 77 cm (n = 9) and 8 ? 9 cm (n = 5), respectively; and mean colonic percent was 67% ? 37% (n = 9). Five patients (cases l-5) were receiving PN at home (17). Therapeutic oral medication included loperamide (12-l 6 mg/day; cases 2, 3, 5-8, and lo), codeine (l-10 cg/day; cases 4 and 8), and mineral salts plus zinc if required. Digestive transit time was not significantly different between patients receiving and those not receiving transitMean digestive transit times slowing medication. were 9 + 7 hours (range, 0.3-18 hours; first red stool) and 26 + 14 hours (range, 1.2-53 hours; first normalcolored stool). Transit times were not correlated with small bowel or colonic length. The 48-hour equilibration period to served meals was therefore sufficient for all patients except patient 9, whose period was lengthened to 72 hours. Nutritional status was either normal (n = 6) or subnormal (n = 4) because no patient had evidence of malnutrition (stage 3). During the 3 months before the study, weight stability was observed only in the 5
1504
MESSING
Table
1.
Patient no.
GASTROENTEROLOGY
ET AL.
Vol. 100, No. 6
Patient Characteristics Time since resection
M/51 F/53 Ml21 M/28 M/71 M/62 Ml47 M/30 F/57 Ml55
S, stomach; D, duodenum; ”+, Present; -, absent. “Given in cm. ‘Given in % (7). “Lateral ostomy. ‘Terminal ostomy.
gut structures”
Diagnosis
(mol
S”
D”
J”
I”
ICV”
C’
RA”
Mesenteric infarction Mesenteric lymphoma Small bowel volvulus Small bowel volvulus Radiation enteritis Crohn’s disease Mesenteric infarction Mesenteric infarction Mesenteric infarction Ulcerative colitis
13 13 7 28 12 8 96 80 32 60
+ + + + + + + + + +
+ + + + + + + + + +
30 0 5 5 200 170 15 25 35 180”
0 0 0 10 Od 30 15 10 15 0
_ _ + + + + -
64 57 100 100 14 35 100 100 100 0
+ + + + + + + + + +
Sex/aget.,vr)
1 2 3 4 5 6 7 8 9 10
Remaining
J, jejunum;
I, ileum; ICV, ileocecal
valvula;
patients (cases 6-10) not receiving PN, whereas a mean percent weight change of - 3% (range, - 17% to 15%) was observed in the 5 patients (cases l-5) receiving oral nutrition plus PN. Physical activity was normal in 7 patients and subnormal in 2; only 1 patient (case 5) had a daily diurnal rest of < 3 hours (stage 2). The mean weight was 89% (range, 74%100%) of the mean ideal body weight. Mean serum albumin and prealbumin levels were 33 & 4 g/L (3.3 ? 0.4 g/dL) and 27 ? 10 mg/dL, respectively. C-reactive protein levels were normal except in patients 1 and 6. Intake Oral intake was unrestricted and composed of normal foods provided in a Continental meal style. The major protein-containing foods included beef, fish, egg, chicken, cheese, and milk. Major fat sources included oil and butter. Carbohydrates were obtained primarily from bread, biscuits, cereal, and vegetables. Fiber intake, estimated by dietary inquiry, was approximately 20-30 g/day. Mean lactose intake was 10 2 5 g/day (range, l-17 g/day). The total daily oral caloric intake measured by bomb calorimetry analysis was 3103 ? 754 kcal. Table 2 compares total oral caloric intake with the caloric intake estimated by dietary inquiry both at home and during the study period when patients were rehospitalized. The PN support (cases 1-5) provided a mean of 1120 ? 246 kcallday (23.3 -c-6.8 kcal . kg-’ . day-‘), and 1.5 -C0.4 g protein . kg-’ . day-‘. During the study, oral intake was not significantly different between the PN and non-PN groups (2847 * 905 vs. 3360 2 432 kcallday or 57 2 19 vs. 58 f 11 kcal . kg-’ . day-‘, respectively).
C, colon; RA, rectum-anus.
output
Mean stool output was 1374 ? 1152 g/day (range, 317-3812 g/day). Mean stool dry weight was 161 2 53 g/day and represented 13% + 7% of the total stool weight. Mean total fecal caloric output was 973 2 315 kcal/day (18.2 + 5.9 kcal . kg-’ . day-‘), the proportions of carbohydrate, lipid, and protein at 28%, 46% and 26%, respectively. Mean urinary caloric output was 258 2 98 kcal/day (4.8 -+ 1.7 kcal . kg-’ . day-‘). Stool output was not different between the groups, with < 70% (cases 1, 2, 5, 6, and 10) or with all [cases 3,4, and 7-9) of the colon left in place (1511 + 1182 vs. 1032 5 957 g/day, respectively), nor between the groups with (cases 4 and 7-9) or without (cases l-3, 5, 6, and 10) an ileocecal valvula (1630 + 582 vs. 1766 + 1258 g/day, respectively). The stool weight did not correlate with either carbohydrate, fat, protein and total caloric output, digestive structure length, or any coefficient of net
Table 2. Oral Intake of Patients With Short Bowel Syndrome < 1 mo before study
kcaliday
kcal . kg-’
day-’ Carbohydrate (%) Fat (%) Protein (%)
During
the 3-day study
3-day dietary inquiry”
Dietary estimation”
3024 57.1 50 36 15
3106 58.9 50 35 15
2 -e 2 +&
753 14.7 7 8’ 2”
2 ? 4 * -c
563 13.3 6 6 2”
Measured” 3103 ? 57.9 4 46 2 31+9 23 !z
754 14.1 11 4
NOTE. Results are expressed as mean 2 SD (n = 10). Comparison between meal composition estimated by dietary inquiry and bomb calorimetry analysis. “Dietary inquiry. bBomb calorimetry. ‘P = 0.028; dP < 0.001; if not stated, P > 0.05.
June
MACRONUTRIENT
1991
digestive absorption Daily fecal weight between the PN and 1898 +-841 g/day);
except protein (0.05> P > 0.02). was not significantly different non-PN groups (1850 + 1223 vs. neither was fecal calorie output (16 -+6 vs.20 I?6 kcal . kg-’ . day-‘). Absorption
The mean total caloric absorption was 2130 t 697 kcallday (39.8 + 14.6kcal . kg-’ . day-‘), and was composed of 55% carbohydrate, 24% fat, and 21% protein (Figure 1). The coefficient of net digestive absorption for total calories, carbohydrate, fat, and protein was 67% 2 12%, 79% k 15%, 52% t IS%, and 61% + 19% respectively (Table 3 and Figure 2). The carbohydrate coefficient of digestive absorption was higher than that for fat (P = 0.002) and protein (P = 0.004), the latter two not being significantly different from each other. Coefficients of digestive absorption for the three different energy sources were neither correlated with each other nor to the absolute amount or proportion of each nutrient in the oral diet. The coefficients of digestive absorption also showed no correlation with the time elapse from the last resection, the length of the small bowel and colon, or the transit time (first red stool). The fat coefficient of digestive absorption was not significantly different between the groups with (n = 5) and without (n = 5) a remnant ileum nor between the group with and without an ileocecal valvula. The coefficients of digestive absorption were not significantly different between the PN and non-PN groups.
ABSORPTION
IN SHORT
1505
BOWEL SYNDROME
Table 3. Coejficients of Digestive Absorption Net digestive
Patient no.
Calories
1
2 3 4 5 6 7 a 9 10 Mean 2 SD
a5 54 41 72 68 64 60 73 a2 72 67 2 12
NOTE. Results
are expressed
Carbohydrates 98 77 47 86 a7 57 a5 88 86 77 79 _' 15 as a percent
absorption Lipids 74 29 44 62 40 65 30 57 73 51 52 _f 16
Proteins a5 56 29 45 42 71 61 59 a4 76 61 2 19
of the total oral intake.
2.5 k 0.4 times their basal energy expenditure. Because this subgroup had a typical ambulatory lifestyle and no body weight variation during the 3 months before the study, their metabolizable energy expenditure would correspond with their total energy expenditure, which was 1.5 2 0.2 times their basal energy expenditure (Table 4). In contrast, energy balance equilibrium was not indicated in the subgroup of 5 patients who had received complementary PN because their weight was not stable before the study. Moreover, because their oral intake and net digestive absorption of the ingested diet was similar to that of the subgroup not receiving PN, they had a higher metabolizable energy (P = 0.048) equal to 2.27 5 0.75 times their basal energy expenditure. Discussion
Energy Balance
The 5 patients exclusively under oral nutrition (cases 6-10) had an oral caloric intake equivalent to
This study of ambulatory patients with SBS, performed under free spontaneous alimentation, showed a high caloric intake and a digestive absorption of about two thirds of the total calories, with net p = 0.004 ‘p = 0.002
INGESTED
I
ABSORBED
Figure 1. Percent of calories (kcal kg-’ day-‘) as protein (top), lipid (middle), and carbohydrates (bottom) ingested and absorbed.
Figure 2. Net absorption expressed as a percent of the total for calories (67% f 12%), carbohydrate (79% t 15%), lipid (52% 2 16%), and protein (61% k 19%).
1506
MESSING
GASTROENTEROLOGY
ET AL.
Vol. 100, No. 6
Table 4. Energy Balance for the Five Stable Patients Without Parenteral Nutrition
Patient no. 6 7 8 9 10 Mean ? SD
Oral intake
Basal energy expenditure
Urinary excretion
Fecal excretion
(BEE1
(OII
OIiBEE
(FE)
WE1
Metabolizable energy [ME)”
1332 1272 1765 1163 1386 1384 2 229
3337 3869 3470 2570 3554 3360 5 432
2.5 3.0 2.0 2.2 2.6 2.5 2 0.4
1216 1538 943 456 1008 1032 + 397
477 233 222 303 305 308 2 102
1644 2098 2305 1811 2241 2020 ? 283
NOTE. Results are expressed “ME = 01 - (FE + UE).
as kcaliday.
Basal energy
expenditure
carbohydrate absorption significantly greater than that of fat or protein. These findings cannot be extrapolated to patients in the initial course following intestinal resection because all patients selected for this study had completed a 6-month postoperative course, received oral alimentation for > 3 months, and probably completed the period of digestive adaptation. In this respect, although few data on the duration of digestive adaptation in humans are available (18), it has been shown that fecal output (volume, nitrogen, fat, sodium, and potassium) after small bowel resection becomes stable 2-3 months after resection when patients have achieved an adequate nutritional status with oral nutrition (19). In our patients, digestive adaptation seemed to have been accomplished because coefficients of digestive absorption did not correlate with the time that had elapsed since resection. Our 10 patients also had no active digestive disease, a criterion that if not met could considerably deteriorate the absorptive capacity and nutritional status of SBS patients. In addition, as determined by the 3month survey performed before the study, the 10 patients had a near-normal nutritional status, an ambulatory life-style, and no dietary restrictions. To ensure that the food intake during the study simulated the patients’ usual intake, dish losses were allowed. Mean estimations of the total amount of calories and carbohydrates consumed at home and during the study were not significantly different from the mean actual intakes measured by bomb calorimetry (Table 2). Moreover, dietary inquiries appeared to mostly overestimate fat intake and underestimate protein intake compared with bomb calorimetry intake measurements. The daily total oral caloric intake for the 10 patients (3103 + 754 kcal/day) was far greater than that described in previous studies by Woolf et al., 1650 kcal/day (4) and 1869 kcal/day (5). Interestingly, our patients’ total caloric coefficients of digestive absorption were similar to those of the patients of Woolf et al., 67% vs. 65% (4) and 62% (5), respectively. Because the results of the present study and those of
was calculated
according
to the Harris
and Benedict
equations
ME/BEE 1.2 1.6 1.3 1.6 1.6 1.5 + 0.2 (16).
Woolf et al. were all based on patients in the sequelae phase of the very SBS, it could be concluded that in such patients hyperalimentation has no deleterious effect on the digestive absorptive capacity of at least the three main energy sources. Moreover, although the coefficients of digestive absorption for the total caloric intake were similar in these three studies, the value for protein absorption in the study of Woolf et al. was greater than that for carbohydrate and fat absorption (81% for protein vs. 51% and 54% for carbohydrate and fat, respectively), the latter two being not significantly different from each other (5), whereas in the present study carbohydrate absorption was significantly greater than fat and protein absorption (Figure 2). It was previously shown in SBS patients that the proportion of fat and carbohydrate present in the oral alimentation did not influence the respective coefficients of digestive absorption (3,4,20). However, these studies were carried out at a low daily total caloric level (1650, 1515, and 1900 kcal/day, respectively) compared with that in the present study (3103 kcal/day). It is therefore possible that this finding does not hold true at high levels of oral energy intake. Another possible explanation for the greater intestinal carbohydrate utilization in our patients is that 9 of 10 of the patients had colonic structures remaining compared with 3 of 8 of the patients in the study by Woolf et al. (5). This point is relevant in light of the increasingly evident role of colonic flora adaptation upon carbohydrate salvage. It was previously shown that ingestion of a glucosidase inhibitor during a a-week period induced a net increase in fecal bacterial mass (21) and that intake of an amylase inhibitor over a T-day period by six non-insulin-dependent diabetics resulted in a decrease in the quantity of malabsorbed carbohydrates without any variation in volatile fatty acid and lactic acid stool output (22). The same findings were also obtained during an 8-day period of lactulose ingestion by eight normal patients in which breath-test analysis indicated an increase in the colonic digestion of lactulose after adaptation (23). These results suggest that small bowel nutrient malabsorption induces an
June 1991
MACRONUTRIENT
adaptation in colonic bacterial metabolism, leading to an increased absorption of carbohydrate products by the colonic mucosa. The capacity of colonic flora to metabolize significant concentrations of carbohydrates during chronic malabsorption has been shown in four patients with a jejunoileal bypass (24).It should be noted, however, that these studies included no SBS patients, and thus further studies are needed to assess the colonic salvage mechanism in such patients. As in our study, the absence of correlation between coefficients of digestive absorption and remaining gut structures was observed in a study by Woolf et al. (5). This finding is not surprising because these studies included few patients with heterogeneous gut structures remaining. A study of 40 jejunostomized patients suggests that a parallel between digestive absorptive capacity and digestive structure length may in fact exist, because a difference in the protein coefficient of digestive utilization was shown between patients with more and those with less than 150 cm of small bowel resected (25).Similarly, a correlation between total caloric absorption and jejunal length was found after a single liquid test meal in three ileostomized subjects without any small bowel resection and in seven patients with a jejunostomy (26). In an attempt to determine the composition of the spontaneous ingested diet of ambulatory patients with SBS, we focused on the five patients receiving oral alimentation only (cases 6-10). Because these patients showed no body weight variation during the 3 months before the study, they could be considered in energy-balance equilibrium. Their mean total energy expenditure level would thus be equivalent to 1.5 times their basal metabolic rate (Table 4), a value very close to that described previously for normal subjects (27,28). Moreover, to compensate for their malabsorption syndrome and to achieve a steady state for energy balance, these five patients ingested a total calorie level equivalent to 2.5 times their basal energy expenditure (table 4). We already noted that the spontaneous oral caloric intake was far greater than those used in several previous SBS studies (3,4,20,26). Indeed, a high calorie level for the spontaneous ingested diet of ambulatory short bowel syndrome patients has also recently been observed (29). It can therefore be postulated that patients with very SBS may achieve a satisfactory energy balance if their oral intake is 2 2.5 times their basal metabolic energy expenditure, taking into account their obligatory increased fecal losses. Compared with patients not receiving PN, patients receiving PN supplementation at a caloric level equal to 0.9 4 0.2 times the basal energy expenditure did not significantly decrease oral intake. However, we noted that the subgroup receiving PN had a slightly higher metabolizable energy level compared with
ABSORPTION
IN SHORT
BOWEL SYNDROME
1507
those without PN. Moreover, the subgroup receiving PN showed a large body weight change (from - 17% to 15% of the actual body weight) during the 3 months before the study. It is not possible, therefore, to make any valid assumptions on energy balance in this subgroup of patients. Conclusion
This prospective study of 10 ambulatory SBS patients on free oral alimentation showed a high caloric intake (3103 & 754 kcal/day) with a net digestive absorption of 67% + 12% of the total ingested calories, a percentage similar to the level reached at lower caloric intakes in previous studies (3-5,25,26). The presence of colonic structures in 9 of the 10 patients likely played a role in the greater absorption of carbohydrates vs. fats and proteins. Because parenteral nutritional support did not influence either the oral intake or the absorptive capacity of SBS patients, we suggest parenteral administration at a caloric level equivalent to the basal energy expenditure and permission for free oral alimentation (30). Absorption of the latter would easily fulfill the requirement for total metabolic expenditure. Finally, the SBS ambulatory patients receiving only oral free alimentation in this study were in apparent energy-balance equilibrium with a spontaneously ingested hypercaloric level equivalent to 2.5 times their basal energy expenditure. References 1. Pigot F, Chaussade
2.
3.
4.
5.
6.
S, Pfeiffer A, Pouliquen X, Jian R, Messing B. Severe short bowel syndrome with a surgically reversed small bowel segment. Dig Dis Sci 1990;35:137-144. Levine GM, Deren JJ, Yezdimiz E. Small bowel resection: oral intake is the stimulus for hyperplasia. Am J Dig Dis 1976;Zl: 542-546. McIntyre PB, Fitchew M. Lennard-Jonnes JE. Patients with a high jejunostomy do not need a special diet. Gastroenterology 1986;91:25-33. Woolf GM, Miller C, Kurian R, Jeejeebhoy KN. Diet for patients with a short bowel: high fat or high carbohydrate? Gastroenterology 1983;84:823-828, Woolf GM, Miller C, Kurian R, Jeejeebhoy KN. Nutritional absorption in short bowel syndrome. Evaluation of fluid, calorie, and divalent cation requirements. Dig Dis Sci 1987;32: 8-15. Bochenek W, Rodgers JB, Balint JA. Effects of changes in dietary lipids on intestinal fluid loss in the short bowel syndrome. Ann Intern Med 1970;72:205-213. Cummings JH, James WPT, Wiggins HS. Role of the colon in ileal resection diarrhoea. Lancet 1973:1:344-347. Baker JP, Detsky AS, Wesson DE, Wolman SC, Stewart S, Whitewell J, Langer B, Jeejeebhoy KN. Nutritional assessment. A comparison of clinical judgment and objective measurements. N Engl J Med 1982;306:969-972. Dewys WD, Begg C, Lavin PT. Band PR, Bennet JM, Bertino JR, Cohen MH, Douglas HO, Ergstrom PF, Erdinli EZ, Horton J, Johnson GJ, Moertel CG, Oken MM, Perlia C, Rosenbaum C,
1508
MESSING
ET AL.
Silverstein MN, Skeel RT, Sponzo RW, Torney DC. Prognostic effect of weight loss prior to chemotherapy in cancer patients. Eastern Cooperative Oncology Group. Am J Med 1980;69:491497. 10. Bernier JJ. Etude du temps de transit intestinal par la courbe de dilution du Cl3 51 Cr. Comparaison avec le test au carmin. Biol Gastroenterol 1969;58:171-180. 11. Van de Kamer JH, Huihink H, Weyers HA. Rapid method for determination of fat in faeces. J Biol Chem 1949;177:347-355. 12. Wootton IDP. Micro-Kjeldahl method. In: Churchill JA, ed. Micro-analysis in medical biochemistry. 4th ed. London: 1964: 140-143. Br J Nutr 13. Miller DS, Payne PR. A ballistic bomb calorimeter. 1959;13:501-508. 14. Southgate DAT, Durnin JVGA. Calorie conversion factors. An experimental reassessment of the factors used in the calculation of the energy value of human diet. Br J Nutr 1970;24:517535. 15. Kurz G, Wallenfels K. II Galactose UV test with galactose dehydrogenase in methods of enzymatic analysis. In: Bergmeyer HU, ed. Volume 3. Weinheim: Verlag Chemie, 1974: 1180-1184,1279-1282. 16. Harris JA, Benedict FG. Biometric studies of basal metabolism in man. Publication number 279. Washington, D.C.: Carnegie Institution, 1919. 17. Messing B, Landais P, Goldfarb B, Irving M. Home parenteral nutrition in adults: a multicentre survey in Europe. Clin Nutr 1989;8:3-9. 18. Hugues CA, Ducker DA. Adaptation of the small intestine. Does it occur in man? Stand J Gastroenterol 1982;17:149-158. 19. Cosnes J, Gendre JP, Lacaine F, Naveau S, Le Quintrec Y. Roles compensateurs de l’ileon et du colon restants apres resection etendue de l’intestin grele. Gastroenterol Clin Biol 1982;6:159165. 20. Ovesen L, Chu R, Howard L. The influence of dietary fat on jejunostomy output in patients with severe short bowel syndrome. Am J Clin Nutr 1983;38:270-277. 21. Scheppa CHW, Fabian C, Ahrens F, Spengler M, Kasper H. Effect of starch malabsorption on colonic function and metabolism in humans. Gastroenterology 1988;95:1549-1555.
GASTROENTEROLOGY
Vol. 100, No. 6
22. Boivin M, Flourie B, Rizza RA, Go VLW, DiMagno EP. Gastrointestinal and metabolic effects of amylase inhibition in diabetics. Gastroenterology 1988;94:387-394. 23. Florent C, Flour% B, Leblond A, Rautureau M, Bernier JJ, Rambaud JC. Influence of chronic lactulose ingestion on the colonic metabolism of lactulose in man. J Clin Invest 1985;75: 608-613. 24. Bond JH, Currier BE, Buchwald H, Levitt MD. Colonic conservations of malabsorbed carbohydrate. Gastroenterology 19808: 445-447. 25. Hylander E, Ladefoged K, Jarnum S. Nitrogen absorption following small intestinal resection. Stand J Gastroenterol 1980;15:853-858. 26. Rodriguez CA, Lennard-Jones SE, Thompson DG, Farthing NG. Energy absorption as a measure of intestinal failure in the short bowel syndrome. Gut 1989;30:176-183. 27. Ravussin E, Burnand B, Schutz Y, Jequier E. Twenty-four hour energy expenditure and resting metabolite rate in obese, moderately obese and control subjects. Am J Clin Nutr 1982;35:566573. 28. Prentice AM, Black AE, Coward WA, Davies HL, Goldberg GR. Murgatroyd PR, Ashford J, Sawyer M, Whitehead KG. High levels of energy expenditure in obese women. BMJ 1986;292: 983-987. 29. Cosnes J, Lamy Ph, Beaugerie L, Le Quintrec M, Gendre JP, Le Quintrec Y. Adaptative hyperphagia in patients with postsurgical malabsorption. Gastroenterology 1990;99:1814-1819. 30. Just B, Morin MC, Brouard A, Pigot F, Thuillier F, Darmaun D, Messing B. Protein and energy intakes during non exclusive home cyclic parenteral nutrition (H.Cy.PN). Clin Nutr 1989; 8(Suppl):72A.
Received January 24, 1990. Accepted October 29, 1990. Address requests for reprints to: B. Messing, M.D., INSERM U 290, Hopital Saint-Lazare, 107bis, rue du Faubourg Saint-Denis, 75010 Paris, France. The authors thank the nurse team of Hopital Saint-Lazare, F. Berthet for expert typing of the manuscript, and L. Trebler Winters for helping with grammatical editing of the manuscript.
1991;100:1502-1508
Intestinal Absorption of Free Oral Hyperalimentation in the Very Short Bowel Syndrome BERNARD MESSING, FRANCOIS PIGOT, MONIQUE RONGIER, MARIE CHRISTINE MORIN, URBAIN NDEiNDOUM, and JEAN CLAUDE RAMBAUD Institut
National
de la Sante
et de la Recherche
MBdicale
U 290, HBpital
Saint-Lazare,
Paris,
France
Ten adult ambulatory patients with the nonactive digestive disease short bowel syndrome were prospectively studied to quantitatively assess their free oral intake and their net digestive absorption of total calories, fat, protein, and carbohydrate during a &day period at least 6 months after a resection. The remaining portions of small bowel had a mean length of 75 cm (range, O-200 cm); the remaining colon lengths had a mean of 67% of normal (range, O%-100%). The experimental diets were formulated according to a home dietary inquiry. During the study period, pooled intakes and digestive losses were measured for total calories, fat, and protein using the bomb calorimetry, Van de Kamer, and Kjeldahl techniques, respectively. The ingested diet provided 58 + 14 kcal 1 kg-’ . day-’ (mean 2 SD) and consisted of 46% carbohydrate, 31% fat, and 23% protein. Net digestive absorption was 67% + 12% for 52% -+ total calories, 79% + 15% for carbohydrate, 16% for fat, and 61% + 19O/bfor protein. The larger net digestive absorption of carbohydrate (P 5 0.004) compared with fat and protein suggests salvage of colonic cholesterol in short bowel syndrome patients. It is concluded that these patients with the short bowel syndrome adapted to a hypercaloric, hyperprotein diet to compensate for increased fecal losses and that this hyperphagia does not seem to have impaired their net digestive absorption.
I
n patients with the short bowel syndrome (SBS), the optimal oral nutritional support has not yet been fully defined. Moreover, the problem of providing adequate nutritional support to SBS patients in clinical practice is increasingly encountered because improvements in surgical procedures and intensive care have increased the survival rates for patients who have undergone extensive small bowel resection (1).
Oral nutrition is necessary for intestinal adaptation after small bowel resection (2). Recent experiments in patients with SBS have given some new insight into this problem. McIntyre et al. (3) found that a semielemental diet did not absorb better than a polymeric diet in four jejunostomized patients, and Woolf et al. (4) showed that neither a high-fat nor a high-carbohydrate diet influenced the absorptive capacity of eight patients with SBS. Fluid restriction during meals also did not appear to change the absorptive capacity of macronutrients (5), and the replacement of 50%~75% of dietary long-chain triglycerides with mediumchain triglycerides led to a decrease in fecal loss of water and improvement in nutritional status of seven patients with SBS (6). Determination of the optimal oral nutritional support in patients with SBS should be approached through analysis of their absorptive capacity, which in fact has rarely been extensively studied ($5). Therefore, we performed a prospective study to evaluate macronutrient absorption in SBS patients under their usual and spontaneous oral alimentation conditions . Materials
and Methods
Patients All patients had undergone extensive small bowel resection with or without colonic resection. The length of the remaining gut structures was determined after analysis of peroperative and radiographic data, the length of the colon being expressed in terms of percent of the usual length according to Ihe method of Cummings et al. (7). Abbreviations used in this paper: PN, parenteral nutrition; SBS, short bowel syndrome. o 1991 by the American Gastroenterological Association 0018-5085/91/$3.00
June 1991
MACRONUTRIENT
Criteria for inclusion in the study included (a] last surgical intervention at least 6 months before the study; (b) no radiological, endoscopic, and/or histological evidence of structural abnormality on the residual intestine before the study; (c) no change in medical therapy (antibiotics, transitslowing agents, pancreatic enzymes, etc.) or commencement of oral supplementation with medium-chain triglycerides during the 3 months before the study; (d) no surgical procedures performed slowing digestive transit time; (e) no extradigestive organ failure; (f) an ambulatory life-style; and (g) consumption of a diet ad libitum with or without home parenteral nutrition (PN), the two being unchanged for a period of at least 3 months. The need to continue PN was assessed by the degree of deterioration of nutritional status during previous attempts to reduce or stop PN support. The protocol was approved by the Ethical Committee of Saint Lazare Hospital, and patients gave their informed consent. Study Protocol The prospective study lasted for a total of 5-6 days, which included a z-s-day equilibrium period and a s-day metabolic period. Patient evaluation upon admission included clinical examination with determination of the clinical nutritional status [stage 1, normal; stage 3, evidence of malnutrition; stage 2, not 1 or 3 (S)], and determination of the physical activity index during the previous month [0, normal; 1, subnormal; 2, daily diurnal rest ~3 hours; 3, daily diurnal rest 2 3 hours (9)]. Hematological analysis included determination of serum albumin, prealbumin, and C-reactive protein levels.
ABSORPTION
IN SHORT BOWEL SYNDROME
1503
ries. The calorie-conversion factors used were those of Atwater et al. confirmed by Southgate and Durnin (14): 4.20, 9.35, and 5.65 kcal/g for carbohydrate, fat, and protein, respectively. Lactose levels were measured using an enzymatic method (15). The basal energy expenditure was calculated according to the Harris-Benedict equations (16) using actual body weights. The net total caloric absorption was calculated by substracting the calories excreted in feces from the calories ingested. The coefficient of net digestive absorption expressed as a percent of total calories, and for the three main energy sources, i.e., carbohydrate, fat, and protein, represented the proportion of ingested calories nat recovered in fecal output. The metabolizable energy level was calculated by subtracting the number of calories excreted in feces and urine from those ingested and, if applicable, those calories provided by PN. Only stable patients (no body weight variation in the 3 months before the study) were assumed to be in an energybalance equilibrium. The quantity of energy metabolized by these patients was considered equivalent to their total energy expenditure. In this subset of patients, metabolizable energy needs for equilibrium were approximated in relation to their estimated basal energy expenditure (16). Results are expressed as means + SD and were tested for statistical significance using Student’s t test for unpaired and paired data. Coefficients of correlation between data were calculated using the least-squares method.
Results Pa tien ts
Diet During the study each patient’s diet was designed to closely match the usual spontaneous intake of calories, carbohydrate, fat, protein, lactose, and fiber as determined by a s-day dietary intake inquiry performed during the month before the study. Each diet was prepared in duplicate, the actual oral intake determined by the difference between the duplicate and dish-losses of the served meals. If the patient was receiving PN, the same protocol used during the 3 months before the study was used during the study. Methods Digestive transit time was estimated using the transit time of a colored marker (red carmine) (10) given orally with the first meal of the study period. The following 2-3 days, depending on the appearance of the red carmin in the stools, served as an equilibrium period, after which the metabolic period was begun. During this s-day metabolic period both urine and stools were collected daily, the latter being collected on ice and immediately frozen at -20°C. Oral-input and fecal-output analyses were performed on homogenized (Waren-Bolster) aliquots of s-day pooled samples. Fat, nitrogen, and total caloric content were measured by the Van de Kamer (ll), micro-Kjeldahl (12), and bomb calorimetry methods, respectively (13). The number of carbohydrate calories was calculated from the difference between total calories and fat plus protein calo-
The clinical features of the 10 patients (aged 47 + 15 years) selected for the study are given in Table 1. The mean time elapsed since the last resection was 35 2 31 months (range, 7-96 months). All
patients had intact stomachs and duodena; mean jejunal and ileal length was 67 ? 77 cm (n = 9) and 8 ? 9 cm (n = 5), respectively; and mean colonic percent was 67% ? 37% (n = 9). Five patients (cases l-5) were receiving PN at home (17). Therapeutic oral medication included loperamide (12-l 6 mg/day; cases 2, 3, 5-8, and lo), codeine (l-10 cg/day; cases 4 and 8), and mineral salts plus zinc if required. Digestive transit time was not significantly different between patients receiving and those not receiving transitMean digestive transit times slowing medication. were 9 + 7 hours (range, 0.3-18 hours; first red stool) and 26 + 14 hours (range, 1.2-53 hours; first normalcolored stool). Transit times were not correlated with small bowel or colonic length. The 48-hour equilibration period to served meals was therefore sufficient for all patients except patient 9, whose period was lengthened to 72 hours. Nutritional status was either normal (n = 6) or subnormal (n = 4) because no patient had evidence of malnutrition (stage 3). During the 3 months before the study, weight stability was observed only in the 5
1504
MESSING
Table
1.
Patient no.
GASTROENTEROLOGY
ET AL.
Vol. 100, No. 6
Patient Characteristics Time since resection
M/51 F/53 Ml21 M/28 M/71 M/62 Ml47 M/30 F/57 Ml55
S, stomach; D, duodenum; ”+, Present; -, absent. “Given in cm. ‘Given in % (7). “Lateral ostomy. ‘Terminal ostomy.
gut structures”
Diagnosis
(mol
S”
D”
J”
I”
ICV”
C’
RA”
Mesenteric infarction Mesenteric lymphoma Small bowel volvulus Small bowel volvulus Radiation enteritis Crohn’s disease Mesenteric infarction Mesenteric infarction Mesenteric infarction Ulcerative colitis
13 13 7 28 12 8 96 80 32 60
+ + + + + + + + + +
+ + + + + + + + + +
30 0 5 5 200 170 15 25 35 180”
0 0 0 10 Od 30 15 10 15 0
_ _ + + + + -
64 57 100 100 14 35 100 100 100 0
+ + + + + + + + + +
Sex/aget.,vr)
1 2 3 4 5 6 7 8 9 10
Remaining
J, jejunum;
I, ileum; ICV, ileocecal
valvula;
patients (cases 6-10) not receiving PN, whereas a mean percent weight change of - 3% (range, - 17% to 15%) was observed in the 5 patients (cases l-5) receiving oral nutrition plus PN. Physical activity was normal in 7 patients and subnormal in 2; only 1 patient (case 5) had a daily diurnal rest of < 3 hours (stage 2). The mean weight was 89% (range, 74%100%) of the mean ideal body weight. Mean serum albumin and prealbumin levels were 33 & 4 g/L (3.3 ? 0.4 g/dL) and 27 ? 10 mg/dL, respectively. C-reactive protein levels were normal except in patients 1 and 6. Intake Oral intake was unrestricted and composed of normal foods provided in a Continental meal style. The major protein-containing foods included beef, fish, egg, chicken, cheese, and milk. Major fat sources included oil and butter. Carbohydrates were obtained primarily from bread, biscuits, cereal, and vegetables. Fiber intake, estimated by dietary inquiry, was approximately 20-30 g/day. Mean lactose intake was 10 2 5 g/day (range, l-17 g/day). The total daily oral caloric intake measured by bomb calorimetry analysis was 3103 ? 754 kcal. Table 2 compares total oral caloric intake with the caloric intake estimated by dietary inquiry both at home and during the study period when patients were rehospitalized. The PN support (cases 1-5) provided a mean of 1120 ? 246 kcallday (23.3 -c-6.8 kcal . kg-’ . day-‘), and 1.5 -C0.4 g protein . kg-’ . day-‘. During the study, oral intake was not significantly different between the PN and non-PN groups (2847 * 905 vs. 3360 2 432 kcallday or 57 2 19 vs. 58 f 11 kcal . kg-’ . day-‘, respectively).
C, colon; RA, rectum-anus.
output
Mean stool output was 1374 ? 1152 g/day (range, 317-3812 g/day). Mean stool dry weight was 161 2 53 g/day and represented 13% + 7% of the total stool weight. Mean total fecal caloric output was 973 2 315 kcal/day (18.2 + 5.9 kcal . kg-’ . day-‘), the proportions of carbohydrate, lipid, and protein at 28%, 46% and 26%, respectively. Mean urinary caloric output was 258 2 98 kcal/day (4.8 -+ 1.7 kcal . kg-’ . day-‘). Stool output was not different between the groups, with < 70% (cases 1, 2, 5, 6, and 10) or with all [cases 3,4, and 7-9) of the colon left in place (1511 + 1182 vs. 1032 5 957 g/day, respectively), nor between the groups with (cases 4 and 7-9) or without (cases l-3, 5, 6, and 10) an ileocecal valvula (1630 + 582 vs. 1766 + 1258 g/day, respectively). The stool weight did not correlate with either carbohydrate, fat, protein and total caloric output, digestive structure length, or any coefficient of net
Table 2. Oral Intake of Patients With Short Bowel Syndrome < 1 mo before study
kcaliday
kcal . kg-’
day-’ Carbohydrate (%) Fat (%) Protein (%)
During
the 3-day study
3-day dietary inquiry”
Dietary estimation”
3024 57.1 50 36 15
3106 58.9 50 35 15
2 -e 2 +&
753 14.7 7 8’ 2”
2 ? 4 * -c
563 13.3 6 6 2”
Measured” 3103 ? 57.9 4 46 2 31+9 23 !z
754 14.1 11 4
NOTE. Results are expressed as mean 2 SD (n = 10). Comparison between meal composition estimated by dietary inquiry and bomb calorimetry analysis. “Dietary inquiry. bBomb calorimetry. ‘P = 0.028; dP < 0.001; if not stated, P > 0.05.
June
MACRONUTRIENT
1991
digestive absorption Daily fecal weight between the PN and 1898 +-841 g/day);
except protein (0.05> P > 0.02). was not significantly different non-PN groups (1850 + 1223 vs. neither was fecal calorie output (16 -+6 vs.20 I?6 kcal . kg-’ . day-‘). Absorption
The mean total caloric absorption was 2130 t 697 kcallday (39.8 + 14.6kcal . kg-’ . day-‘), and was composed of 55% carbohydrate, 24% fat, and 21% protein (Figure 1). The coefficient of net digestive absorption for total calories, carbohydrate, fat, and protein was 67% 2 12%, 79% k 15%, 52% t IS%, and 61% + 19% respectively (Table 3 and Figure 2). The carbohydrate coefficient of digestive absorption was higher than that for fat (P = 0.002) and protein (P = 0.004), the latter two not being significantly different from each other. Coefficients of digestive absorption for the three different energy sources were neither correlated with each other nor to the absolute amount or proportion of each nutrient in the oral diet. The coefficients of digestive absorption also showed no correlation with the time elapse from the last resection, the length of the small bowel and colon, or the transit time (first red stool). The fat coefficient of digestive absorption was not significantly different between the groups with (n = 5) and without (n = 5) a remnant ileum nor between the group with and without an ileocecal valvula. The coefficients of digestive absorption were not significantly different between the PN and non-PN groups.
ABSORPTION
IN SHORT
1505
BOWEL SYNDROME
Table 3. Coejficients of Digestive Absorption Net digestive
Patient no.
Calories
1
2 3 4 5 6 7 a 9 10 Mean 2 SD
a5 54 41 72 68 64 60 73 a2 72 67 2 12
NOTE. Results
are expressed
Carbohydrates 98 77 47 86 a7 57 a5 88 86 77 79 _' 15 as a percent
absorption Lipids 74 29 44 62 40 65 30 57 73 51 52 _f 16
Proteins a5 56 29 45 42 71 61 59 a4 76 61 2 19
of the total oral intake.
2.5 k 0.4 times their basal energy expenditure. Because this subgroup had a typical ambulatory lifestyle and no body weight variation during the 3 months before the study, their metabolizable energy expenditure would correspond with their total energy expenditure, which was 1.5 2 0.2 times their basal energy expenditure (Table 4). In contrast, energy balance equilibrium was not indicated in the subgroup of 5 patients who had received complementary PN because their weight was not stable before the study. Moreover, because their oral intake and net digestive absorption of the ingested diet was similar to that of the subgroup not receiving PN, they had a higher metabolizable energy (P = 0.048) equal to 2.27 5 0.75 times their basal energy expenditure. Discussion
Energy Balance
The 5 patients exclusively under oral nutrition (cases 6-10) had an oral caloric intake equivalent to
This study of ambulatory patients with SBS, performed under free spontaneous alimentation, showed a high caloric intake and a digestive absorption of about two thirds of the total calories, with net p = 0.004 ‘p = 0.002
INGESTED
I
ABSORBED
Figure 1. Percent of calories (kcal kg-’ day-‘) as protein (top), lipid (middle), and carbohydrates (bottom) ingested and absorbed.
Figure 2. Net absorption expressed as a percent of the total for calories (67% f 12%), carbohydrate (79% t 15%), lipid (52% 2 16%), and protein (61% k 19%).
1506
MESSING
GASTROENTEROLOGY
ET AL.
Vol. 100, No. 6
Table 4. Energy Balance for the Five Stable Patients Without Parenteral Nutrition
Patient no. 6 7 8 9 10 Mean ? SD
Oral intake
Basal energy expenditure
Urinary excretion
Fecal excretion
(BEE1
(OII
OIiBEE
(FE)
WE1
Metabolizable energy [ME)”
1332 1272 1765 1163 1386 1384 2 229
3337 3869 3470 2570 3554 3360 5 432
2.5 3.0 2.0 2.2 2.6 2.5 2 0.4
1216 1538 943 456 1008 1032 + 397
477 233 222 303 305 308 2 102
1644 2098 2305 1811 2241 2020 ? 283
NOTE. Results are expressed “ME = 01 - (FE + UE).
as kcaliday.
Basal energy
expenditure
carbohydrate absorption significantly greater than that of fat or protein. These findings cannot be extrapolated to patients in the initial course following intestinal resection because all patients selected for this study had completed a 6-month postoperative course, received oral alimentation for > 3 months, and probably completed the period of digestive adaptation. In this respect, although few data on the duration of digestive adaptation in humans are available (18), it has been shown that fecal output (volume, nitrogen, fat, sodium, and potassium) after small bowel resection becomes stable 2-3 months after resection when patients have achieved an adequate nutritional status with oral nutrition (19). In our patients, digestive adaptation seemed to have been accomplished because coefficients of digestive absorption did not correlate with the time that had elapsed since resection. Our 10 patients also had no active digestive disease, a criterion that if not met could considerably deteriorate the absorptive capacity and nutritional status of SBS patients. In addition, as determined by the 3month survey performed before the study, the 10 patients had a near-normal nutritional status, an ambulatory life-style, and no dietary restrictions. To ensure that the food intake during the study simulated the patients’ usual intake, dish losses were allowed. Mean estimations of the total amount of calories and carbohydrates consumed at home and during the study were not significantly different from the mean actual intakes measured by bomb calorimetry (Table 2). Moreover, dietary inquiries appeared to mostly overestimate fat intake and underestimate protein intake compared with bomb calorimetry intake measurements. The daily total oral caloric intake for the 10 patients (3103 + 754 kcal/day) was far greater than that described in previous studies by Woolf et al., 1650 kcal/day (4) and 1869 kcal/day (5). Interestingly, our patients’ total caloric coefficients of digestive absorption were similar to those of the patients of Woolf et al., 67% vs. 65% (4) and 62% (5), respectively. Because the results of the present study and those of
was calculated
according
to the Harris
and Benedict
equations
ME/BEE 1.2 1.6 1.3 1.6 1.6 1.5 + 0.2 (16).
Woolf et al. were all based on patients in the sequelae phase of the very SBS, it could be concluded that in such patients hyperalimentation has no deleterious effect on the digestive absorptive capacity of at least the three main energy sources. Moreover, although the coefficients of digestive absorption for the total caloric intake were similar in these three studies, the value for protein absorption in the study of Woolf et al. was greater than that for carbohydrate and fat absorption (81% for protein vs. 51% and 54% for carbohydrate and fat, respectively), the latter two being not significantly different from each other (5), whereas in the present study carbohydrate absorption was significantly greater than fat and protein absorption (Figure 2). It was previously shown in SBS patients that the proportion of fat and carbohydrate present in the oral alimentation did not influence the respective coefficients of digestive absorption (3,4,20). However, these studies were carried out at a low daily total caloric level (1650, 1515, and 1900 kcal/day, respectively) compared with that in the present study (3103 kcal/day). It is therefore possible that this finding does not hold true at high levels of oral energy intake. Another possible explanation for the greater intestinal carbohydrate utilization in our patients is that 9 of 10 of the patients had colonic structures remaining compared with 3 of 8 of the patients in the study by Woolf et al. (5). This point is relevant in light of the increasingly evident role of colonic flora adaptation upon carbohydrate salvage. It was previously shown that ingestion of a glucosidase inhibitor during a a-week period induced a net increase in fecal bacterial mass (21) and that intake of an amylase inhibitor over a T-day period by six non-insulin-dependent diabetics resulted in a decrease in the quantity of malabsorbed carbohydrates without any variation in volatile fatty acid and lactic acid stool output (22). The same findings were also obtained during an 8-day period of lactulose ingestion by eight normal patients in which breath-test analysis indicated an increase in the colonic digestion of lactulose after adaptation (23). These results suggest that small bowel nutrient malabsorption induces an
June 1991
MACRONUTRIENT
adaptation in colonic bacterial metabolism, leading to an increased absorption of carbohydrate products by the colonic mucosa. The capacity of colonic flora to metabolize significant concentrations of carbohydrates during chronic malabsorption has been shown in four patients with a jejunoileal bypass (24).It should be noted, however, that these studies included no SBS patients, and thus further studies are needed to assess the colonic salvage mechanism in such patients. As in our study, the absence of correlation between coefficients of digestive absorption and remaining gut structures was observed in a study by Woolf et al. (5). This finding is not surprising because these studies included few patients with heterogeneous gut structures remaining. A study of 40 jejunostomized patients suggests that a parallel between digestive absorptive capacity and digestive structure length may in fact exist, because a difference in the protein coefficient of digestive utilization was shown between patients with more and those with less than 150 cm of small bowel resected (25).Similarly, a correlation between total caloric absorption and jejunal length was found after a single liquid test meal in three ileostomized subjects without any small bowel resection and in seven patients with a jejunostomy (26). In an attempt to determine the composition of the spontaneous ingested diet of ambulatory patients with SBS, we focused on the five patients receiving oral alimentation only (cases 6-10). Because these patients showed no body weight variation during the 3 months before the study, they could be considered in energy-balance equilibrium. Their mean total energy expenditure level would thus be equivalent to 1.5 times their basal metabolic rate (Table 4), a value very close to that described previously for normal subjects (27,28). Moreover, to compensate for their malabsorption syndrome and to achieve a steady state for energy balance, these five patients ingested a total calorie level equivalent to 2.5 times their basal energy expenditure (table 4). We already noted that the spontaneous oral caloric intake was far greater than those used in several previous SBS studies (3,4,20,26). Indeed, a high calorie level for the spontaneous ingested diet of ambulatory short bowel syndrome patients has also recently been observed (29). It can therefore be postulated that patients with very SBS may achieve a satisfactory energy balance if their oral intake is 2 2.5 times their basal metabolic energy expenditure, taking into account their obligatory increased fecal losses. Compared with patients not receiving PN, patients receiving PN supplementation at a caloric level equal to 0.9 4 0.2 times the basal energy expenditure did not significantly decrease oral intake. However, we noted that the subgroup receiving PN had a slightly higher metabolizable energy level compared with
ABSORPTION
IN SHORT
BOWEL SYNDROME
1507
those without PN. Moreover, the subgroup receiving PN showed a large body weight change (from - 17% to 15% of the actual body weight) during the 3 months before the study. It is not possible, therefore, to make any valid assumptions on energy balance in this subgroup of patients. Conclusion
This prospective study of 10 ambulatory SBS patients on free oral alimentation showed a high caloric intake (3103 & 754 kcal/day) with a net digestive absorption of 67% + 12% of the total ingested calories, a percentage similar to the level reached at lower caloric intakes in previous studies (3-5,25,26). The presence of colonic structures in 9 of the 10 patients likely played a role in the greater absorption of carbohydrates vs. fats and proteins. Because parenteral nutritional support did not influence either the oral intake or the absorptive capacity of SBS patients, we suggest parenteral administration at a caloric level equivalent to the basal energy expenditure and permission for free oral alimentation (30). Absorption of the latter would easily fulfill the requirement for total metabolic expenditure. Finally, the SBS ambulatory patients receiving only oral free alimentation in this study were in apparent energy-balance equilibrium with a spontaneously ingested hypercaloric level equivalent to 2.5 times their basal energy expenditure. References 1. Pigot F, Chaussade
2.
3.
4.
5.
6.
S, Pfeiffer A, Pouliquen X, Jian R, Messing B. Severe short bowel syndrome with a surgically reversed small bowel segment. Dig Dis Sci 1990;35:137-144. Levine GM, Deren JJ, Yezdimiz E. Small bowel resection: oral intake is the stimulus for hyperplasia. Am J Dig Dis 1976;Zl: 542-546. McIntyre PB, Fitchew M. Lennard-Jonnes JE. Patients with a high jejunostomy do not need a special diet. Gastroenterology 1986;91:25-33. Woolf GM, Miller C, Kurian R, Jeejeebhoy KN. Diet for patients with a short bowel: high fat or high carbohydrate? Gastroenterology 1983;84:823-828, Woolf GM, Miller C, Kurian R, Jeejeebhoy KN. Nutritional absorption in short bowel syndrome. Evaluation of fluid, calorie, and divalent cation requirements. Dig Dis Sci 1987;32: 8-15. Bochenek W, Rodgers JB, Balint JA. Effects of changes in dietary lipids on intestinal fluid loss in the short bowel syndrome. Ann Intern Med 1970;72:205-213. Cummings JH, James WPT, Wiggins HS. Role of the colon in ileal resection diarrhoea. Lancet 1973:1:344-347. Baker JP, Detsky AS, Wesson DE, Wolman SC, Stewart S, Whitewell J, Langer B, Jeejeebhoy KN. Nutritional assessment. A comparison of clinical judgment and objective measurements. N Engl J Med 1982;306:969-972. Dewys WD, Begg C, Lavin PT. Band PR, Bennet JM, Bertino JR, Cohen MH, Douglas HO, Ergstrom PF, Erdinli EZ, Horton J, Johnson GJ, Moertel CG, Oken MM, Perlia C, Rosenbaum C,
1508
MESSING
ET AL.
Silverstein MN, Skeel RT, Sponzo RW, Torney DC. Prognostic effect of weight loss prior to chemotherapy in cancer patients. Eastern Cooperative Oncology Group. Am J Med 1980;69:491497. 10. Bernier JJ. Etude du temps de transit intestinal par la courbe de dilution du Cl3 51 Cr. Comparaison avec le test au carmin. Biol Gastroenterol 1969;58:171-180. 11. Van de Kamer JH, Huihink H, Weyers HA. Rapid method for determination of fat in faeces. J Biol Chem 1949;177:347-355. 12. Wootton IDP. Micro-Kjeldahl method. In: Churchill JA, ed. Micro-analysis in medical biochemistry. 4th ed. London: 1964: 140-143. Br J Nutr 13. Miller DS, Payne PR. A ballistic bomb calorimeter. 1959;13:501-508. 14. Southgate DAT, Durnin JVGA. Calorie conversion factors. An experimental reassessment of the factors used in the calculation of the energy value of human diet. Br J Nutr 1970;24:517535. 15. Kurz G, Wallenfels K. II Galactose UV test with galactose dehydrogenase in methods of enzymatic analysis. In: Bergmeyer HU, ed. Volume 3. Weinheim: Verlag Chemie, 1974: 1180-1184,1279-1282. 16. Harris JA, Benedict FG. Biometric studies of basal metabolism in man. Publication number 279. Washington, D.C.: Carnegie Institution, 1919. 17. Messing B, Landais P, Goldfarb B, Irving M. Home parenteral nutrition in adults: a multicentre survey in Europe. Clin Nutr 1989;8:3-9. 18. Hugues CA, Ducker DA. Adaptation of the small intestine. Does it occur in man? Stand J Gastroenterol 1982;17:149-158. 19. Cosnes J, Gendre JP, Lacaine F, Naveau S, Le Quintrec Y. Roles compensateurs de l’ileon et du colon restants apres resection etendue de l’intestin grele. Gastroenterol Clin Biol 1982;6:159165. 20. Ovesen L, Chu R, Howard L. The influence of dietary fat on jejunostomy output in patients with severe short bowel syndrome. Am J Clin Nutr 1983;38:270-277. 21. Scheppa CHW, Fabian C, Ahrens F, Spengler M, Kasper H. Effect of starch malabsorption on colonic function and metabolism in humans. Gastroenterology 1988;95:1549-1555.
GASTROENTEROLOGY
Vol. 100, No. 6
22. Boivin M, Flourie B, Rizza RA, Go VLW, DiMagno EP. Gastrointestinal and metabolic effects of amylase inhibition in diabetics. Gastroenterology 1988;94:387-394. 23. Florent C, Flour% B, Leblond A, Rautureau M, Bernier JJ, Rambaud JC. Influence of chronic lactulose ingestion on the colonic metabolism of lactulose in man. J Clin Invest 1985;75: 608-613. 24. Bond JH, Currier BE, Buchwald H, Levitt MD. Colonic conservations of malabsorbed carbohydrate. Gastroenterology 19808: 445-447. 25. Hylander E, Ladefoged K, Jarnum S. Nitrogen absorption following small intestinal resection. Stand J Gastroenterol 1980;15:853-858. 26. Rodriguez CA, Lennard-Jones SE, Thompson DG, Farthing NG. Energy absorption as a measure of intestinal failure in the short bowel syndrome. Gut 1989;30:176-183. 27. Ravussin E, Burnand B, Schutz Y, Jequier E. Twenty-four hour energy expenditure and resting metabolite rate in obese, moderately obese and control subjects. Am J Clin Nutr 1982;35:566573. 28. Prentice AM, Black AE, Coward WA, Davies HL, Goldberg GR. Murgatroyd PR, Ashford J, Sawyer M, Whitehead KG. High levels of energy expenditure in obese women. BMJ 1986;292: 983-987. 29. Cosnes J, Lamy Ph, Beaugerie L, Le Quintrec M, Gendre JP, Le Quintrec Y. Adaptative hyperphagia in patients with postsurgical malabsorption. Gastroenterology 1990;99:1814-1819. 30. Just B, Morin MC, Brouard A, Pigot F, Thuillier F, Darmaun D, Messing B. Protein and energy intakes during non exclusive home cyclic parenteral nutrition (H.Cy.PN). Clin Nutr 1989; 8(Suppl):72A.
Received January 24, 1990. Accepted October 29, 1990. Address requests for reprints to: B. Messing, M.D., INSERM U 290, Hopital Saint-Lazare, 107bis, rue du Faubourg Saint-Denis, 75010 Paris, France. The authors thank the nurse team of Hopital Saint-Lazare, F. Berthet for expert typing of the manuscript, and L. Trebler Winters for helping with grammatical editing of the manuscript.
