Effects of Oral Versus Intravenous Nutrition on Intestinal Adaptation After Small Bowel Resection in the Dog

70:712-719, 1976 Copyright © 1976 by The Williams & Wilkins Co.

GASTROENTEROLOGY

Vol. 70, No.5

Printed in U.S.A.

EFFECTS OF ORAL VERSUS INTRAVENOUS NUTRITION ON INTESTINAL ADAPTATION AFTER SMALL BOWEL RESECTION IN THE DOG E. J.

FELDMAN,

R. H.

DOWLING,

J.

MCNAUGHTON, AND

T. J.

PETERS

Departments of Medicine and Surgery, Royal Postgraduate Medical School, London, England

To study the influence of luminal nutrition on the structural and functional adaptive changes which are seen in the residual intestine after partial small bowel resection, quantitative histology, in vitro uptake of ue i-leucine, mucosal enzyme activities, and in vivo absorption of glucose were studied before and 6 weeks after 50% proximal small bowel resection in 10 greyhound dogs, 5 of which were nourished exclusively by the intravenous route while 5 were pair-fed by mouth. In the orally fed jejunectomized dogs, the ileum became dilated with mucosal hyperplasia, the villus height increased from 796 ± SEM 26 Jlm to 1102 ± 28 Jlm (P < 0.001), and there was a corresponding increase in glucose absorption in vivo (milligrams· centimeter of intestine -1min - ') from 5.3 ± 1.2 to 10.1 ± 1.6. The increased absorption seems mainly caused by the dilation and villus hyperplasia, since there was no significant change in in vitro absorption and mucosal enzyme activity when expressed per unit weight of intestine. In the absence of exogenous luminal nutrition, the well nourished intravenously fed dogs showed no evidence of functional adaptation and a significant fall in mean ileal villus height from 823 ± 48 Jlm at the initial operation to 732 ± 57 Jlm 6 weeks after jejunectomy. These results provide the most direct evidence to date that luminal nutrition is essential for the development of intestinal adaptation after resection. They also suggest that luminal contents are necessary to maintain the structural and functional integrity of the normal small bowel. In many animal species including man, after resection of one part of the small bowel, the residual intestine develops both structuraP-6 and functionaP-11 adaptive changes, which include small bowel dilation, villus enlargement, epithelial cell hyperplasia, more rapid cell migration, and enhanced absorption (when expressed per unit length of intestine)_ There are several possible mechanisms for this intestiReceived August 19, 1975. Accepted October 31, 1975. This paper was presented in part at the British Society of Gastroenterology, September, 1973 and at the American Gastroenterology Association meeting in San Francisco, May, 1974. Preliminary reports of this work have been published in abstract form.'·2 Address requests for reprints to Dr. Dowling at: Gastroenterology Unit, Department of Medicine, Guy's Hospital Medical School, London, SEI 9RT England Dr. Dowling's work was supported by the Medical Research Council (Intestinal Malabsorption Group). Dr. Feldman's present address is: Division of Gastroenterology, Department of Medicine, U.C.L.A. School of Medicine, Los Angeles, California 90024. The authors are grateful to Professor C. C. Booth and Mr. John Spencer, F.R.C.S., for their advice and criticism, Miss Gillian Wells and Mr. Brian Chalk for valuable technical assistance, Messrs. Dennis Wilson and Ken Stevenson for their help with animal care, and Mrs. Hazel Creed, who typed the manuscript. They also wish to thank Messrs. Vitrum A. B., Stockholm, Sweden for their generous supplies of Intralipid and Yamin for the parenteral feeding. 712

nal adaptation, including changes in luminal nutrition, 12 the trophic effects of pancreaticobiliary secretions, 13 and the influence of hormonal factors. 14, 15 (Luminal nutrition is defined as the presence of nutrient material in the intestinal lumen; it does not necessarily mean that the nutrients must be absorbed to exert their effect.) Although the evidence that luminal nutrition plays a major role in intestinal adaptation is strong, it is, nonetheless, circumstantial. Therefore, an experimental model was designed to study this in the dog more directly by comparing both the structural and functional adaptive changes in the ileum from two groups of jejunectomized dogs, one group being nourished exclusively by vein and the second being pair-fed by mouth. The results obtained in these studies show not only that exogenous luminal nutrition is essential for ileal adaptation after jejunal resection, but also that luminal nutrients are necessary for maintaining the structural and functional integrity of the normal small intestine.

Methods Experimental Design Experimental animals. Pedigree female greyhounds weighing

20 to 28 kg were used throughout. Previous studies have shown that after small bowel resection in the dog the adaptive changes in the residual intestine develop within 1 to 2 months. ,. Therefore, it was decided to examine the structural and functional changes in the residual

May 1976

ORAL VS IV NUTRITION AFTER SMALL BOWEL RESECTION

ileum 6 weeks after jejunectomy in both orally (5 dogs) and intravenously (5 dogs) nourished animals, each dog acting as its own control. Initial operation. At the initial operation, the proximal 50% of the small intestine was resected and the mucosa from a 3to 4-cm segment at the distal end of the excised bowel was then rapidly dissected from the underlying muscle and serosa for in vitro studies (see below) of HC-leucine " absorption " (The term " absorption " is used to describe tissue uptake of labeled substrates in vitro), mucosal enzyme activities, and histology. The residual ileum was then prepared for in vivo perfusionabsorption studies by cannulation of the transected proximal end and by insertion of a second cannula through a small elliptical stab wound 40 to 60 cm downstream, with 16 French gauge latex rubber Pezzar catheters which were held in place with purse string sutures. After the ileal perfusion, the catheters were removed, the stab wound was closed, intestinal continuity was reestablished with an end-to-end anastomosis, and the abdomen was closed . In dogs to be nourished parenterally, an indwelling silastic catheter was then inserted between the scapulae through a skin tunnel to the external jugular vein at the side of the neck and thence to the superior vena cava. Final operation. At the final operation, the macroscopic appearance of the residual intestine was noted and a further 3 to 4-cm segment was removed from the proximal end of the ileal remnant for repeat studies of histology, mucosal enzymes, and in vitro absorption of HC-l-leucine. (This biopsy segment was taken below the localized scar tissue of the anastomotic site but was otherwise adjacent to the segment removed at the initial operation from the distal end of the resected jejunum.) The in vivo glucose absorption studies were then repeated in the same segment of ileum which had been studied at the initial operation .

713

Intestinal Structure Dissecting microscopy. Since variations in villus shape, other than changes in villus height, may influence the potential absorptive surface area, the mucosal architecture in segments of intestine at the initial and final operation was examined under the dissecting microscope. Light microscopy . Full thickness segments of intestine measuring approximately 1 by 2 cm were pinned flat on cork, fixed in 10% formol saline , processed, and embedded in paraffin. Serial 4 to 5 ~m thick sections were then cut and stained with hematoxylin and eosin for histological measurements. Villus height was then measured in the 10 tallest, well oriented villi and crypt depth was recorded using a calibrated eyepiece graticule. The epithelial cell size and density were also measured in 200-~m lengths of mid villus and again the mean reading from 10 measurements was recorded. Intestinal Function

Mucosal enzyme activity. Tared mucosal samples were stored at -20°C in 0.25 M sucrose containing 1 mM Na2 EDTA, pH 7.4, and 0.1 %ethanol. The thawed tissue was homogenized at O°C with a Polytron homogenizer and aliquots were taken for mucosal protein measurements IS and for enzyme assay. Brush border enzymes (a-glucosidase and leucyl-{j-naphthylamidase) were assayed as described by Peters et al. 19 and the peroxisomal marker enzyme, catalase, was assayed by the method of Baudhuin et al. 20 The results of mucosal protein content were expressed as milligrams of protein per milligram of DNA, while the mucosal enzymes were expressed as milliunits of enzyme per milligram of DN A. 21 Paired " before and after" results were available in 5 dogs nourished intravenously and in 4 dogs from the orally fed group. In vitro absorption. In vitro amino acid absorption was studied using a tissue accumulation method modified from that described by Robinson and Mirkovitch. 22 At the time of Nutrition operation, small, approximately equal sized pieces of mucosa Parenteral. For their intravenous nutrition, the dogs were (weight range 47 to 101 mg; mean 75.1 mg) were quickly placed in a specially constructed cage for approximately 6 hr dissected from the underlying tissue,23 weighed on a_torsion each day. Each animal received 75 kcals · kg of body weight-I balance, and temporarily stored in oxygenated (95% O2 + 5% day- I in the form of an amino acid mixture (Vamin , Vitrum, CO 2 ) Krebs Ringer Tris buffer 2 • (pH 7.4). Duplicate pieces of Sweden), a stabilized synthetic triglyceride emulsion (20% tissue were incubated for 2, 5, 10, 15, and 20 min in 10 ml of an Intralipid, Vitrum, Sweden), and 20% dextrose in a v/v ratio of oxygenated 1 mM solution of l-leucine in Krebs Ringer Tris 3:1:2. The intravenous fluid lines from the infusion bottles were buffer containing I·C-Ieucine (Radiochemical Centre, Amerconnected to the indwelling venous catheter through a flexible sham , U. K. specific activity 10 ~c per mM). At the end of the steel hose which was attached at one end to a canvas and metal incubation period, tissue uptake was stopped by adding harness on the animal 's back 17 while the other end was crushed ice to the incubation medium and by placing the counterbalanced over a pulley system which enabled the dog to incubation flasks in iced water. The tissue was then washed stand, sit or lie at will, during the infusion. In addition, the gently with several rinses of ice-cold 0.15 M NaCl , placed in dogs received a multivitamin preparation (U.S .V. Phar- scintillation vials, solubilized with a commercial alkaline maceutical Corp.) once per week . tissue digestant (Protosol, New England Nuclear, Boston, After the period of intravenous infusion, the dogs were Mass .) and the "c radioactivity was counted in a Beckman LS housed in individual kennels and were allowed free access to 250 liquid scintillation counter with internal quench correctap water but received no food by mouth . The animals were tion. Results were expressed as nanomoles of leucine uptake . milligram of wet weight mucosa -I minute-I. This method does groomed and exercised daily. The response to intravenous feeding was monitored by daily not discriminate between intracellular and extracellular leutemperature recordings, twice weekly measurements of body cine uptake, but pilot studies with labeled inulin as a marker weight, and weekly blood counts, serum electrolytes, liver for tissue water show that the pattern of results is similar "function" tests, and preinfusion serum lipids (cholesterol and whether or not inulin is present in the incubation mixture (J. W. L. Robinson, personal communication). Furthermore, these triglycerides) . Oral. The orally fed dogs received the same amounts of experiments were not designed to study the subsequent fate of carbohydrate, fat, and protein per kilogram of body weight the leucine taken up by the tissues, although with the short (based on the calculated quantities in commercial dog food incubation periods used it is unlikely that the leucine would be preparations) as did the intravenously nourished animals and, incorporated into tissue protein . In vivo absorption . Segmental ileal glucose absorption apart from the infusions, they were handled and housed in the (defined here as disappearance of substrate from the perfusion same way .

714

FELDMAN ET AL.

medium) was measured using a continuous single pass infusion technique in anaesthetized dogs at both the initial and final operation. The test segment (approximately 30 to 50 cm in length) was flushed with 0.15 MNaCI before perfusion with a 21 mM glucose solution made isotonic with NaCI and containing 3 g of polyethylene glycol per liter (mol wt 4(00) as a non absorbable marker. An accurately measured flow rate which ranged between 10 and 11 ml per min was used and after a 30-min equilibration period the effluent fluid was collected over nine 1O-min periods. Net fluid transfer was calculated from change between the initial and final polyethylene glycol concentrations. At the end of the absorption study during the second operation, the perfused ileal segment was dissected from its mesentery and its length was measured against a vertical scale using a standard stretch with a 20 g weight. Glucose absorption, corrected for fluid transfer, was then calculated and expressed as micromoles of glucose absorbed· centimeter of intestine -, hour- '.

Results Nutritional state of the dogs during intravenous feeding. The dogs tolerated the intravenous feeding well with no evidence of discomfort during the period of infusion. Throughout the 6 weeks of parenteral nutrition their general condition was excellent, and comparable to that of the controls. The dogs showed no obvious signs of hunger, but precautions had to be taken to avoid coprophagia. After the first week, the animals passed only small amounts of whitish, mucus "stools." During the first few days of parenteral feeding, the rate of intravenous infusion was kept slow to avoid hyperpnea and pyrexia from the Vamin and Intralipid, respectively. As a result, the dogs received only 50% of their subsequent calorie input on the first day, but this was increased gradually to reach 75 kcal per day within 4 days. In general, the hematological and biochemical screen-

Vol.70,No.5

ing tests were satisfactory, although in some animals the hemoglobin levels fell by up to 3 g per 100 ml during the 6-week period of intravenous feeding. Three of the 5 dogs had raised serum alkaline phosphatase levels, and isoenzyme studies showed that this was of hepatic/bone origin (D. Moss, personal communication). The results of body weight recordings are shown in figure 1. In the first 7 to 14 days after laparotomy and jejunal resection, both the orally and intravenously fed dogs lost approximately 10% of their preoperative body weight, but thereafter both groups gained weight steadily and although the mean weight gain in the orally fed dogs was slightly greater than that in the intravenously fed animals, this difference was not statistically significant.

Intestinal Structure Macroscopic findings. At the second operation, the residual ileum of the orally fed dogs showed obvious gross macroscopic changes, with dilation and enlargement. In contrast, the external appearance of the bowel in the intravenous group was either normal or slightly narrowed and contracted. Dissecting microscopy. The topographical appearance of the ileal mucosa in both the orally and intravenously fed dogs remained unchanged throughout the study, with the normal dog pattern of tall, cylindrical villi. Light microscopy. The results of histological measurements of villus height are shown in figure 2. The measurements of crypt depth and of the number of epithelial cells per 200-~m length of mid villus are given in table 1. In the orally fed dogs, jejunal resection resulted in a marked increase in villus height from a mean of 796 (± SEM 26) ~m to 1102 (± 28) ~m, and this 38% difference was statistically significant (t = 15.9; P < 0.001). There was also a 22 % increase in the mean crypt depth after

120

no - - ----------------------

100 %OF PREOPERATIVE WEIGHT

90

80

o

14

21

28

35

42

TIME AFTER RESECTION (days) FIG. 1. The effect of jejunal resection on body weight expressed as a percentage of the preoperative weight (mean intravenously. There was no significant difference between the two groups at any time interval.

± SEM)

in dogs fed orally or

May 1976

715

ORAL VS IV NUTRITION AFTER SMALL BOWEL RESECTION 1200

1200

1000

1000

~

E-~ -:J

E-

800

800

600

600

II

Villous height

400

400

200

200

(microns)

0

0

BEFORE

I. V. FED

ORAL FED

FIG. 2. The effect of jejunal resection on histological measurements of ileal villus height (I'm) before and 6 weeks after operation in both the orally fed (left panel) and intravenously fed (right panel) dogs. The E-shaped symbols show mean values ± SEM. TABLE 1. Histological measurements (mean

±

SEM) of crypt depth and density of epithelial cells in orally and intravenously fed dogs before and 6 weeks after .50% proximal small bowel resection Epithelial cell density (no./200 I'm length mid villus)

Crypt depth (I'm)

Dogs

After

Before

Orally fed ivfed Statistical significance

549 ± 10 499 ± 12 (NS)

(NS)" (NS)

669 ± 62 498 ± 26

t P

=

2.55;

< 0.05

Before

64.6 63.6

± ±

1.2 3.6

(NS)

After

(NS) (NS)

70.5 ± 2.1 56.3 ± 2.3

t P

=

4.14;

< 0.005

" NS, not significant.

jejunectomy. This increased mucosal thickness was mainly due to hyperplasia with a greater number of cells, rather than to hypertrophy of individual cells-in fact if anything the epithelial cells were smaller after resection than at the original operation, so that the number of cells needed to "fill" a unit length of mid villus was slightly greater after resection, although this difference was not statistically significant (table 1). In contrast to the obvious adaptive changes in the orally fed dogs, there was no evidence of structural adaptation in the intravenously nourished group. Conversely, there was a significant drop in the mean ileal villus height from 823 ± 48 at the initial operation to 732 ± 57 ~m 6 weeks after jejunectomy (t = 2.79; P < 0.05). This change in villus size was again mainly due to a change in the total number of cells, although unlike the results in the orally fed animals, the epithelial cells were slightly larger in the hypoplastic mucosa of the intravenously nourished dogs, with the result that the number of cells per 200-~m length of mid villus was less after jejunal resection than at the first operation. Whereas the "before and after" difference in cell number was not

significant in the intravenous group, there was a significant difference in cell density between the orally and intravenously nourished dogs after jejunectomy (table 1). There was no change in the mean crypt depth.

Intestinal Function Mucosal enzyme activity. The results of mucosal protein content and of enzyme specific activities are shown in table 2. The cell protein content (milligrams of protein per milligram of DNA) of the mucosal homogenates was essentially the same at the initial operation and after resection in both the orally and intravenously nourished dogs. Similarly, there was no significant difference in the mean enzyme activities before and after jejunectomy in the intravenously fed animals, but there was an increase in all three enzyme activities after resection in the orally fed dogs, the results for a-glucosidase and catalase being significantly greater (P < 0.05) than for those in the base line control group (table 2). In vitro absorption. The pattern of results in the HC-l-Ieucine tissue uptake studies was comparable at all

716

Vol.70,No . .5

FELDMAN ET AL.

TABLE 2. Protein content and enzyme activities in mucosal homogenates from normal ileum and ileal remnant after jejunal resection in orally and intravenously nourished dogs Enzyme activity Dogs (no.)

Protein content

Leucyl-i3naphthalamidase

a-Glucosidase

0.91

±

0.11

4.5

±

0.71

10.2

±

1.23

22.4

±

1.13

1.15 1.08

±

0.30 0.31

7.3 3.9

±

0.96 a 0.57

17.7 11.9

±

3.02" 1.72

20.1 23.2

±

4.73 0.70

Catalase

mp./mg protein

Preresection (n Postresection oral (n ~ 4) iv (n ~ 5)

~

7)

±

a Significantly greater than in the preresection group (t "Significantly greater than in the preresection group (t

1.25

~ ~

±

2.34; P 2.30; P

< <

mg/mgDNA

±

±

0.05). 0.05). 1.25

5 MINS

2 MINS

1.00

1.00

lfUC I"" 0.75 ACCUMUlATION

75

MUCOSAL

In moles/mg

wet wt tiss.) 0.50

50

0.25

25

0

ORAL BEFORE

"-

"-AFTER /

)V

/

RfSECTION

ORAL BEFORf

\

AFTER

/

IV

"-RfSECTION/

FIG. 3. The effect of jejunal resection on ileal L-Ieucine absorption measured in vitro from a 1 mM solution before and 6 weeks after jejunectomy in both the orally and intravenously fed dogs. The bars indicate mean values and vertical lines indicate SEM'S after 2 (left panel) and 5 (right panel) min incubation, respectively.

TABLE 3. Ileal glucose absorption measured i'n vivo before and 6 weeks after jejunal resection in orally and intravenously fed dogs (mean ± SEM) Ileal glucose absorption (p.moles·cm intestine-' hr ')

Dogs Before Orally fed iv fed a

29.4 50.6

± ±

6.9 8.4

After 56.1 50.0

± ±

8.8 4.9

Statistical significance

t ~ 2.40; P < 0.05 Nsa

NS, not significant.

time intervals studied. Representative results, at the 2- and 5-min incubation times, are shown in figure 3. As with the mucosal enzyme results, there was again a slight, but statistically insignificant, increase in the amount of leucine absorbed per milligram wet weight of ileal mucosa in the orally fed dogs and a corresponding decrease in the intravenously fed group. It seems, therefore, that there is little evidence of change in function when results are expressed per unit weight of tissue. In vivo absorption. The results of glucose absorption from the ileal perfusion studies are given in table 3. Unlike the results of leucine uptake measured in vitro and expressed per milligram of mucosa, the in vivo

results of glucose absorption when expressed per unit length of intestine show a pattern of functional adaptation similar to the pattern of the histological studies. In orally fed dogs, ileal glucose absorption increased approximately 2-fold 6 weeks after jejunectomy. In contrast, there was no evidence of functional adaptation after jejunal resection in the intravenously nourished animals (table 3).

Discussion The results of this study provide the most direct evidence so far that the presence of luminal nutrition is essential for the development of both structural and functional adaptation in the ileum after jejunectomy. Furthermore, with an intact intestine, the absence of exogenous luminal nutrition in dogs fed parenterally led to a degree of hypoplasia in the ileum comparable to that seen in the present study and to an even greater degree of hypoplasia in the jejunum. 25 The present results confirm that with nothing but parenteral nutrition, dogs can be maintained in excellent health with edema-free weight gain. The mechanism for the low hemoglobin and raised serum alkaline phosphatase levels in the present study is not clear, but

May 1976

ORAL VS IV NUTRITION AFTER SMALL BOWEL RESECTION

it seems unlikely that these changes could have influenced the results of intestinal structure and function. Total parenteral nutrition as a model for studying the consequences of withdrawing exogenous luminal nutrition from the intestine has also been used by others. Johnson et al. 26 showed that after only 6 days of intravenous feeding in the rat there was marked intestinal mucosal hypoplasia, whereas Levine et al. 27 found similar changes with reductions in intestinal mucosal enzymes after 7 days of intravenous nutrition in the rat. It could be argued that with continued intestinal epithelial cell shedding and the secretion of salivary, gastric, and pancreaticobiliary secretions, the intestine is not totally deprived of luminal nutrition during intravenous feeding. However, the contribution to luminal nutrients from these sources is probably relatively small. Furthermore, it seems likely that both the epithelial cell proliferation rate and the volume of digestive secretions will be markedly reduced in the absence of a food stimulus. Evidence in support of this hypothesis comes from another experimental model. When food is excluded from bypassed Thiry-Vella loops in the rat, there is a marked slowing in the epithelial cell migration rate. 28 In most situations when intestinal mucosal hyperplasia occurs as part of an adaptive response, there is an associated increase in absorptive function per unit length of bowel, but paradoxically, there is either no change or even a decrease in the function of individual epithelial cells. 10,29-31 In general, the present findings confirm this pattern of results. The in vitro studies of leucine uptake were expressed per unit mass of tissue (per milligram of protein and per milligram of DNA). Having established that the density of epithelial cells on the side of the villi showed only very minor changes in both the orally and intravenously fed groups, and assuming that the function of individual cells does not change, then even with a 2- or 3-fold increase in mucosal mass per unit of intestinal length after resection, the leucine uptake per unit weight of tissue and the enzymespecific activity would not change. Indeed, it has been shown previously that in the rat in hyper- and hypoplastic mucosa, even though there were corresponding increases and decreases in enzyme activity per unit length of intestine, there was little change in the function of individual cells. 29 In contrast, the in vivo results of glucose absorption in the present study were based on the length of the perfused intestine and showed a pattern similar to the results of the structural studies-namely increased absorption per centimeter of ileum in the orally fed dogs, but no such functional adaptation in the intravenously nourished animals. The results of glucose absorption per unit length of ileum before resection were different in the orally fed and intravenously fed dogs. The reason for this difference is not clear but it may be due to the fact that the average length of the perfused intestine was greater in the orally fed dogs (35 to 50 cm), in which absorption was measured early in the study, than in animals which were nourished intravenously (25 to 35 cm). The choice of

717

shorter test segments in the later experiments avoided mechanical problems such as kinking of the intestinal loops during perfusion in situ, but while these results suggest that the shorter the test loop, the better the absorption per centimeter of length, this hypothesis has not yet been tested systematically. It seems likely that the increased glucose absorption in the orally fed dogs was due to the greater absorptive surface resulting from both intestinal dilation and villus hyperplasia. In relatively large animals such as the dog, it would be difficult, if not impossible, to measure the absorptive surface area with any degree of accuracy. However, the one-dimensional measurement of villus height from histological sections probably does provide a useful index of absorptive surface area, particularly in the dog. Unlike the rat, which has broad, leafy villi and mucosal ridges, 28, 32 and man, where the potential surface area cannot be assumed from measurements of villus height alone,33 the topographical appearance of the dog's cylindrical, finger-like villi remained unchanged after resection in both groups of animals . The changes in villus size were mainly caused by hyperplasia in the orally fed dogs, and by hypoplasia in the animals nourished intravenously. However, there were also minor changes in cell size, with the result that hyperplasia was combined with hypotrophic cells while the epithelial cells on the shorter villi of the intravenous group were hypertrophic. No cell kinetic studies were done in these dogs, but these slight changes in cell size would be consistent with the concept that in hyperplastic mucosa there is more rapid cell proliferation and migration, resulting in villi populated by small, relatively immature cells. Conversely, in hypoplasia, with a slower rate of cell turnover, the cells are, if anything, large and "hypermature" when they migrate from the crypts to the villus side. 34 . 35 The mechanism whereby the absence of exogenous luminal nutrition leads to mucosal hypoplasia is unknown, but there are at least four possible explanations. First, the absence of food could have changed the intestinal bacterial flora, theoretically with either increased or reduced numbers of bacteria resulting in secondary changes in the small bowel mucosa. Bacterial investigations were not carried out in the present study, but the pattern of mucosal change with preservation of the normal villus to crypt size ratio makes this explanation unlikely. In germ-free rats, the normal villus to crypt ratio gives way to a pattern of tall, slender villi with short crypts and decreased cell turnover,36 which contrasts with the present findings. Second, the ingested food could have been utilized directly by the intestinal mucosa during transport, either as an energy source or as substrates for cell synthesis, and there is evidence that both carbohydrates and proteins may be used directly by the intestinal epithelium during absorption. 37 , 38 Third, the absence of luminal contents could have led to smaller than normal amounts of gastrointestinal polypeptide hormones, which may in themselves be trophic to the intestine. Support for this theory comes from results of studies of enteroglucagon 39-41 and the synthetic

718

FELDMAN ET AL.

analogue of gastrin-pentagastrin,42, 43 both of which may stimulate intestinal growth. Finally, the absence of food may have failed to trigger cholecystokinin-pancreozymin and secretin release from the intestinal mucosa and in turn, the resultant exocrine pancreatic hyposecretion could lead to mucosal hypoplasia. There is certainly convincing evidence that the intestinal "trigger" hormones are trophic to the pancreas" and that when these physiological stimulae are absent, as occurs during intravenous feeding, the pancreas becomes hypoplastic- with , diminished funttion;45 If, as has been suggested, pancreatic secretions are trophic to the intestine, 13 the corollary that absence of pancreatic juice should cause intestinal hypoplasia may also be true. We are currently investigating this possibility in dogs maintained on long term intravenous nutrition ± exocrine pancreatic stimulation with parenteral secretin and/or cholecystokinin -pancreozymin. If the present results may be extrapolated to man, it suggests that patients maintained on long term intravenous nutrition should also have, if possible, food by mouth to prevent intestinal mucosal hypoplasia and an associated reduction in absorptive function. REFERENCES 1. Feldman EJ, McNaughton J, Dowling RH: Comparative effect of

2. 3. 4.

5.

6.

7.

8. 9.

10. 11.

12.

13. 14.

intravenous and oral nutrition on intestinal adaptation (abstr). Gut 14:831, 1973 Feldman EJ, Peters TJ, McNaughton J, et al: Adaptation after small bowel resection : comparison of oral versus intravenous nutrition (abstr). Gastroenterology 66:691, 1974 Flint JM: The effect of extensive resection of the small intestine. Bull Johns Hopkins Hosp 23:127-144, 1912 Booth CC , Evans KT, Menzies DF, et al: Intestinal hypertrophy following partial resection of the small bowel in the rat. Br J Surg 46:403-410, 1959 Nygaard K: Resection of the small intestine in rats. Ill. Morphological changes in the intest inal tract. Acta Chir Scand 133:233- 248, 1967 Loran MR, Althausen TL: Cellular proliferation of intestinal epithelia in the rat two months after partial resection of the ileum. J Biophys Biochem Cytol 7:667-671, 1960 Nygaard K: Resection of the small intestine in rats. 1. Nutritional status and adaptation of fat and protein absorption. Acta Chir Scand 132:731-742, 1966 Dowling RH, Booth CC : Structural and functional changes following small intestinal resection in the rat. Clin Sic 32:139-149,1967 Weinsten DL, Shoemaker CP, Hersh T, et al: Enhanced intestinal absorption after small bowel resection in man. Arch Surg 99:560- 562, 1969 Weser E, Hernandez MH: Studies of small bowel adaptation after intestinal resection in the rat. Gastroenterology 60:69-75, 1971 Bury KD: Carbohydrate digestion and absorption after massive resection of the small intestine . Surg Gynecol Obstet 135: 177-187, 1972 Dowling RH: The influence of luminal nutrition on intestinal adaptation after small bowel resection and by-pass. In Intestinal Adaptat ion . Edited by Dowling RH and Riecken E-O . Stuttgart, Schattauer-Verlag, 1974, p 35- 46 Altmann GG: Influence of bile and pancreatic secretions on the size of the intestinal villi in the rat. Am J Anat 132: 167 - 178, 1971 Loran MR, Carbone JV : The humoral effect of intestinal resection on cellular proliferation and maturation in parabiotic rats. In

Vol. 70, No.lj

Sullivan MF: Gastrointestinal Radiation Injury. Amsterdam, Excerpta Medica, 1968, p 127-139 15. Laplace JP: Effet humoral de la resection intestinale chez des porcs en circulation sanguine croisee chronique (abstr). Bioi Gastroenterol (Paris) 6:359, 1973 16. Bochkov NP: Morphological and physiological changes in the small intestine of the dog after its partial resection. Bull Exp Bioi Med 46:1261- 1265, 1958 17. Dudrick SJ, Steiger E, Wilmore DW, et al: Continuous long-term intravenous infusion in unrestrained animals. Lab Anim Care 20:521-529, 1970 18. Lowry OH,' Rosebrough NJ, C ' arr AL, et al : Protein measurement with the folin phenol reagent. J bioi Chern 193:2.65-275, 1951 19. Peters TJ, Heath JR, Wansbrough-Jones MH, et al: Enzyme activities and properties of Iysosomes and brush borders in jejunal biopsies from control subjects and patients with coeliac disease. Clin Sci Mol Med 48:259-267, 1975 20. Baudhuin P, Beaufay H, Rahman-Li YO, et al: Intracellular distribution of monoamine oxidase, aspartate amino ·transferase, alanine aminotransferase, d-amino acid oxidase and catalase in rat-liver tissue. Biochem J 92 :179-184, 1964 21. Kissane JM, Robins E: The fluorometric measurement of deoxyribonucleic acid in animal tissue with special reference to the central nervous system. J Bioi Chern 233:184-188, 1958 22. Robinson JWL, Mirkovitch V: The recovery of function and microcirculation in small intestinal loops following ischaemia. Gut 13:784- 789, 1972 . 23. Hakins A, Lester RG, Tilson N: Absorption by an in vitro preparation of dog intestinal mucosa. J Appl Physiol 18:409- 413, 1963 24. Lazarrow PB, deDuve C: Intermediates in the biosynthesis of peroxisomal catalase in rat liver. Biochem Biophys Res Commun 45:1198- 1203, 1971 25. Dowling RH, Feldman EJ, MacNaughton J, et al: Importance of luminal nutrition for intestinal adaptation following small bowel resection (abstr) . Digestion 10:216-217, 1974 26. Johnson LR, Castro GA, Lichtenberger LM, et al : The significance of the trophic action of gastrin (abstr). Gastroenterology 66:718, 1974 27. Levine G, Steiger E, Deren JJ : The importance of oral intake in maintenance of rat small intestinal mass, proximal-distal gradient (PDG) , and disaccharidase activity (abstr). Gastroenterology 66:850, 1974 28. Gleeson MH, Cullen J , Dowling RH: Intestinal structure and function after small bowel by-pass in the rat. Clin Sci 43:731- 742, 1972 29. Gleeson, MH, Dowling RH , Peters TJ : Biochemical changes in intestinal mucosa after experimental small bowel by-pass in the rat. Clin Sci 43:743-757 , 1972 30. McCarthy DM, Kim YS: Changes in sucrase, enterokinase and peptide hydrolase after intestinal resection. J Clin Invest 52:942- 951, 1973 31. Rodgers JB Jr., Bochenek W: Localization of lipid reesterifying enzymes of the rat small intestine. Effects of jejunal removal on ileal enzyme activities. Biochem Biophys Acta 202:426-435, 1970 32. Chacko CJG, Mathan, VI, Baker SJ : Changes in the mucosal pattern of isolated loops of jejunum in albino rats. A dissection microscope study . Br J Exp Pathol 49 :40-43, 1968 33. Clark RM: Morphological description of intestinal adaptation: Measurements and their meaning. In Intestinal Adaptation. Edited by Dowling RH and Riecken E-O. Stuttgart, SchattauerVerlag, 1974, p 11-19 34. McManus JPA, Isselbacher KJ: Effect of fasting versus feeding on the rat small intestine . Gastroenterology 59:214- 221, 1970 35. Dowling RH, Gleeson MH : Cell turnover following small bowel resection and by-pass. Digestion 8:176- 190,1973

May 1976

ORAL VS IV NUTRITION AFTER SMALL BOWEL RESECTION

36. Abrams GD, Bauer H, Sprinz H: Influence of the normal flora on mucosal morphology and cellular renewal in the ileum. Lab Invest 12:355-364, 1963 37. Kinter WB, Wilson TH: Autoradiographic study of sugar and amino acid absorption by everted sacs of hamster intestine. J Cell Bioi 25:19-39, 1965 38. Smyth DH: Glucose absorption. Gastroenterology 42:76, 1962 39. Gleeson MH, Bloom SR, Polak JM, et al: Endocrine tumour in kidney affecting small bowel structure, motility and absorptive function. Gut 12:773-782, 1971 40. Rudo ND, Rosenberg IH: Chronic glucagon administration enhances intestinal transport in the rat. Proc Soc Exp Bioi Med 142:521-525, 1973 41. Jacobs LR, Polak J, Bloom S, et al: Does enteroglucagon playa

42. 43.

44.

45.

719

trophic role in intestinal adaptation (abstr) Clin Sci Mol Med 1976 (in press) Johnson LR: The trophic action of gastrointestinal hormones. Gastroenterology 70:278-288, 1976 Mayston, PD, Barrowman JA, Dowling RH: Effect of Pentagastrin on small bowel structure and function in the rat. Digestion 12:7884, 1975 Mayston P, Barrowman JA: The influence of chronic administration of pentagastrin on the rat pancreas. Q J Exp Physiol 56:113-122, 1971 Johnson LR, Copeland EM, Dudrick SJ, et al: Structural and hormonal alterations in the gastrointestinal tract of parenterally fed rats. Gastroenterology 67:1177-1183, 1975