Postobstructive Enteropathy in Infants With Transient Enterostomy: Its Consequences on the Upper Small Intestinal Functions By J.P. Cezard,
Y. Aigrain,
E. Sonsino,
N. Lambert,
J. Macry,
E. Grasset,
G. Weisgerber,
and J. Navarro
Paris, France l Repeated
or prolonged
organic obstruction
of the small
intestine in the neonatal period can lead to severe refeeding problems, despite a transient ostomy. These problems are thought to result from a postobstructive enteropathy (POE) of the apparently normal small intestine segment above the obstruction. Ten infants with a POE, characterized by limited oral caloric and carbohydrate
intakes and increased ostomy
effluent, were compared with 8 controls with an enterostomy and a normal postoperative refeeding pattern. There was no statistical difference in the histomorphometric the mucosa or its digestive border
hydrolases,
or absorptive
glucose
transport)
appearance of capacity
between
(brushthe
two
groups. The effluent and duodenal floras of the two groups were similar. However, abnormal transit
peristalsis
times.
obstruction
all POE patients showed significant
characterized
This suggests
in the neonatal
caused by chronic motricity tine
above
frequent
the
that period
obstruction.
Although
causing prenatal
and carmin or prolonged
could lead to a POE,
abnormalities
after small bowel atresia,
other conditions
by barium repeated
of the small intesthis
POE is more
it may also occur with and postnatal
intestinal
obstruction. Copyright
8 1992 by W.B. Saunders Company
INDEX WORDS: Postobstructive peristaltism;
T
enteropathy;
intestinal dys-
intestinal atresia; intestinal obstruction,
infant.
HE SURVIVAL of infants with small intestine obstructions, especially in the neonatal period, has been considerably enhanced by better surgical management,‘-4 the use of transient enterostomy when indicated,3-5 and improved medical treatment, especially the advent of total parenteral nutrition.3~6,7 Despite this progress, some infants still suffer from persistent severe oral refeeding difficulties, with diarrhea, vomiting, and sometimes bacteriemia of gastrointestinal origin. These clinical disorders can occur in infants having a transient enterostomy but no short bowel,* nonfunctioning anastomosis,2J or dilated proximal bowel.“-” They are presumed to result from a chronic small intestine enteropathy located above the initial obstruction, but the underlying functional defects are not clearly understood. This study was undertaken to investigate the characteristics of the proximal small bowel dysfunction in infants with small intestine diversion, suffering from such a chronic enteropathy. The results for these children were compared to those for a control group of children having a transient enterostomy, a similar length of remaining small bowel, and a normal postoperative refeeding pattern.
MATERIALS
AND
METHODS
Eighteen infants with small intestine diversion as treatment of repeated intestinal obstruction, localized peritonitis, or perforation were studied. They were divided in two groups: a control group (8 infants) and a postobstructive enteropathy (POE) group (10 infants). The characteristics of the two groups are shown in Tables 1 and 2. They were matched for age, sex, mean age of ostomy. and length of proximal small bowel. There was no dilatation or organic obstruction of the proximal small intestine in any case. Obstruction etiologies are given in Table 1. The same pathologies were present in the two groups, with more midsmall intestinal atresia in the POE group. This group also had more organic obstructive relapses before the stomy (8 infants, 1,one in the control group). The main difference between the two groups was the duration of artificial nutritional support. Complete oral feeding were always resumed within 2 months of the transient ostomy, and intestinal continuity was restored within 3 months following the diversion in the control group. In the POE group, exclusive oral feeding, usually by constant rate enteral nutrition, was resumed at the end of the parenteral nutrition period and frequently (7110) after restoration of intestinal continuity (8.5 2 6 months after intestinal diversion). The stoma effluent output, in the two groups, was measured with and without oral feeding and expressed as ml/kg/d. The maximum constant rate enteral feeding tolerance (Kcallkgid) and the maximum daily sugar tolerance (g/l0 cm of small intestine) were calculated for the POE group. Sugar tolerance was measured semiquantitatively in the effluent (Clinistix. Clinitest, Miles Laboratory). These criteria were measured 2 to 6 months after installing the transient stoma and tested with a hypoallergenic infant formula (small peptide casein hydrolysate, Amirige. Nutricia, Malmaison, France) and dextrin maltose (Caloreen, Clinidiet; Roussel Nutrition, Paris, France) plus appropriate amounts of minerals and vitamins. In addition to these criteria defining the POE. the small intestine mucosa was analyzed by suction biopsy via the stoma (at 10 to 15 cm from the opening) and by duodenal biopsies (in 5 controls and 8 POE infants). The biopsy samples were analyzed histomophometrically, including villus and crypt height, number of mitoses per crypt, intraepithelial lymphocyte count, and inflammatory submucosal lesions. Digestive capacity was assessed by measuring brushborder disaccharidases (sucrase, maltase, and lactase) and neutral Glucose active transport was studied aminopeptidase activities. ‘%14 in three of the POE group using an Ussing chamber.r5
From the Services de Gastroenterologie Pediatrique et de Chirurgie Laboratoires d Xnatomo-pathologie et de Bacteriologic, Hopital Robert Debre, Paris, France. Date accepted: September 24, 1991. Supported in part by a grant from the Faculte Xavier Bichat, Paris, France. Address reprint requests to Jean Pierre Cezard, MD, Service de Gastroenterologie et Nutrition Pediatrique, Hopital Roben Debre, 48 Blvd Serurier, Pans 75019. France. Copyright o I992 by W.8. Saunders Company 0022-346819212711-00141$03.00l0
JournatofPediacnc Surgery, Vol27, No 11 (November), 1992: pp 1427-1432
1427
CEZARD ET AL
1428
RESULTS
Table 1. Surgical Course of Children in the Control and Postobstructive
Enteropathy
Groups
Clinical Results
POE
Controls (n = 8)
(n = 10)
Girls
4
4
Boys
4
6
40 + 22
39 f 16
Sex
Age of stoma (d) Organic obstructive
(r = 1:6)
relapses before the
8/10 (r = 2:2)
stomy Length of the proxi6
4
2 (mean; 160 cm;
6 (mean; 145 cm;
range 100 to 240 cm)
range 80 to 220 cm)
1
6
1
1
mal bowel >150cm 80to150cm Etiology Midintestinal or ileal atresia
Histological, Digestion, and Absorption Studies
Total colon Hirschsprung’s disease Necrotizing entero4
colitis Intestinal intussusception
1
1
1
1 -
Intestinal volvulus Meconial peritonitis Duration of transient stoma (mo)
The POE group exhibited the clinical criteria used to separate the two groups (prolonged parenteral nutrition, resumption of complete oral feeding). Two to 6 months after intestinal derivation, they also showed an abnormal significant increase in stoma effluent in the fasting state and especially during refeeding. This increase was due to a severe oral feeding intolerance. Maximum oral feeding tolerance was 34 -+ 14 I&al/kg/d, with a maximum sugar tolerance of 6.2 -+ 2.5 g/kg/d or 2.1 ? 0.8 g/l0 cm small intestine/d. These results were significantly different from the control group (Table 2).
2.6
8.5
(l-24)
(2-21)
Abbreviation: r, relapse.
The stoma effluent was collected via a small sterile canula and analyzed for intestinal microflora. Aeroanaerobic bacteria were identified and counted as previously described.t6 Any bacteriemia was correlated with bacteria found in the small intestine. The duodenal microfloras of in 8 of the 10 POE infants were also examined. Small intestine morphology and transit were studied by barium small intestine transit, and the transit time was measured with an inert marker (carmine test). Data are expressed as means 2 SD. Differences between groups were tested for significance using Student’s t test.
The histomorphometric data for the distal and proximal suction biopsies are shown in Figs 1 and 2. There was no statistical difference between the two groups for villus and crypt heights, the number of mitoses per crypt, or the number of intraepithelial lymphocytes. Edema and an increased submucosal lymphocytes were frequently observed (9/10 POE, 6/8 controls), but were similar in the two groups. The digestive capacity of the intestinal mucosa, as measured by brush-border hydrolases (Table 3) was normal in both groups. Glucose active transport was measured on biopsy fragments from the three children having almost complete intolerance to glucose in vivo. The in vitro values were in the normal range. Small Intestine Flora
The bacteria in the stoma effluent were identified and counted several times for each child. The total
Table 2. Clinical Course of the Children in the Control and Postobstructive
Enteropathy Groups
Controls Fasting stoma output (mL/kg/24
h)
POE
(n = 8)
In = 10)
16 2 5
55 2 lo*
Feeding tolerance Stoma output (mL/kg/24
h)
Maximum calories (Kcal/kg/24
h)
26 2 9
82 ‘- 12*
105 r 12
34 +- 14*
21 ?6
6.2 + 2.5*
Maximum glucose (g/kg/24 h) (g/IO cm intestine/24 h) Parenteral nutritiont (mo)
6.6 * 2.1 1.3 ? 0.6
BacteremiaS
0
2.1 ? 0.8” 19.5 + 9* 15
NOTE. Data given as mean + SD.
MONTHS
*P < .Ol. tOr
total oral feeding
SBacteremia effluent.
restoration.
associated
with the same bacterium
in the stoma
Fig 1. Villus and crypt height in the control [m) and POE (0) groups. Retrograde biopsies were taken 10 to 15 cm from the intestinal stoma.
POSTOBSTRUCTIVE
1429
ENTEROPATHY
I
I I I I I I I I I I I I I I I I I I I I
0 0
0 0
1
2
4
3
6
5
7
I I I I I
MONTHS Fig 2. Villus and crypt height in the control (H) and POE (0) groups (duodenal biopsies).
bacteria counts in the two groups, were not significantly different (lo6 to lo9 bacteria/ml) (Fig 3). The populations were similar and strict anaerobic bacteria were rare (Clostridium, n = 2). Bacteriemia occurred frequently in the POE group (3.1 per patient in the POE v 0.22 per patient in the controls). The duration of parenteral nutrition and catheter-related bacteriemia were probably the main causes, but the same bacterium was also frequently (15 times) found in the small intestine liquid of the POE group (Table 4), suggesting an intestinal origin for these bacteriemia. The bacteria in the duodenum were also identified and counted in 8 of 10 of the POE patients. Bacterial overgrowth (> lo4 bacteria/ml) occurred in two of the POE patients. Intestinal Transit The barium small intestinal transits of the POE group and 7 of the control patients were examined. None had a persistant dilatation of the intestine above the stomy, partial or total organic obstruction, or a blind loop or stenosis of the stoma. However, all the children in the POE group had either accelerated transit times ( 80 cm), still develop obstructive relapse with sepsis, chronic diarrhea with Table 4. Bacteremia in the Postobstructive
Enteropathy Group and
Identification of the Same Bacterium in the Stoma Effluent Table 3. Brush Border Hydrolases (Specific Activities) in the Control and Postobstructive
Enteropathy Groups Controls (n = 8)
Bacteremia POE (n = 10)
Staphylococcus Enterococcus
Sucrase (mU/mg protein)
117 +- 40
146 + 88
Klebsiella
Maltase (mU/mg protein)
430 k 190
431 k 293
Enterobacter
Lactase (mU/mg protein)
92 + 61
80 ? 38
174 + 104
120 2 65
Neutral aminopeptidase
(mU/mg protein)
NOTE. Data given as mean 2 SD.
cloacae
Bacteriumin the Stoma Effluent
15
5
4
4
3
2
2
2
Pseudomonas
3
1
E coli
1
1
Total
28
15
CEZARD ET AL
Fig 4. Barium small intestine transit in a child with a POE secondary to an ileocolonic intussusception with intestinal necrosis and transient stomy. The x-ray taken 7 minutes after barium ingestion showed abnormally rapid transit.
malabsorption, failure to thrive, and contaminated small boweLzO As a result, oral feeding is impossible and prolonged parenteral nutrition necessary in these patients. These pathological events can occur even when a transient intestinal diversion is present, suggesting a POE of the proximal small bowel above the obstruction. From 18 children with a transient enterostomy for the treatment of neonatal intestinal obstruction, we isolated a group of 10 infants who presented these characteristics. A secretory and malabsorptive diarrhea was indicated by a large stoma effluent even with no oral feeding, which increased dramatically when oral feeding was attempted. Sugar malabsorption was readily detected in the effluent liquid and was used to measure maximum glucose tolerance (Table 2). The progress of these cases of POE was slow. Despite a slight improvement in the tolerance of oral feeding during the initial weeks, little improvement occurred thereafter, even when intestinal continuity was restored, and complete oral feeding was not possible before an average of 20 months. This testifies to both the chronic nature of the enteropathy and to
Fig 5. Barium small intestine transit in a child with midintestinal atresia and POE. The rapid transit (5 minutes) was associated with barium dilution and fragmentation.
the slow, progressive adaptation of the gut. In contrast, the control group children were on normal oral feeding before and after transient intestinal diversion (within 3 months). We have examined the changes underlying this POE by studying the histological appearance and digestive and absorptive capacities of the mucosa by serial retrograde and duodenal biopsies. There was no difference between the two groups. Rectal biopTable 5. Small Intestine Transit in the POE and Control Groups by Barium Transit and Carmine Test Controls
POE
Barium transit Rapid < 10 min
117
9/10
o/7
6110
Rapid < 10 min
118
9/10
Dysperistalsis’
l/8
6/10
Dilution fragmentation Carmine test
lDysperistalsis
is the time between appearance and disappearance
of carmin in the stoma effluent > 24 hours.
POSTOBSTRUCTIVE
ENTEROPATHY
1431
sies were also taken from all the POE children; the muscular layer and myenteric plexus were normal (except for th e c ases of Hirschsprung’s disease). The histological appearance of the stomy after intestinal continuity was restored, showed similar lesions in the two groups and it was similar to that usually described.” Transparietal biopsies of the remaining small intestine were not taken. Analysis of the stomy effluent for duodenal flora showed a contaminated small bowe12” in 2 of the 8 POE children examined. This low level of small bowel contamination may be the consequence of transient intestinal diversion. Although gram-negative bacteriemia were frequent in this group, with the same bacterium being present in the stoma fluid, we could find no evidence for intestinal translocation.“’ The abnormalities found in these cases of POE were motility disorders of the proximal bowel above the obstruction, which persisted despite removal of the obstruction. Upper gastrointestinal barium studies eliminated other causes of dysmotricity, such as stenoses of the small intestine or the stomy, or blind or distended loops. The abnormalities found and confirmed by the carmine transit studies were a very rapid transit time (9/10 patients) frequently associated with irregular peristalsis (6/10). This suggests that the main cause of this post obstructive enteropathy is a chronic motility disorder. This could explain the apparent malabsorption observed when oral feeding is attempted. Electromyographic and manometric studies are now being undertaken to describe these abnormalities. Small intestinal atresia (type II and III) was the main etiology; 6 of 10 POE group children presented with this congenital condition. Ineffective peristalsis is well known in infant and experimental intestinal atresia, but it was thought to be restricted to the dilated jejunal or duodenal segment.4,2,8-11All our
cases had only midintestinal or ileal atresia; the dilated portion of the upper small intestine was resectedl,?J and the length of the remaining small intestine (over 80 cm) was adequate. A short jejunoplasty was performed during primary end-to-end anastomosis (5/6) to insure a functioning anastomosis.? This strongly suggests that dysperistalsis may sometimes occur in the apparently normal small intestine above the dilated intestinal segment in midintestinal or ileal atresia. This dysperistalsis may be responsible for the obstructive relapse (5/6) and a POE. Other forms of congenital and postnatal organic bowel obstruction, such as total colon Hirschsprung disease, prenatal and postnatal volvulus, necrotizing enterocolitis, or even severe intestinal intussusception, could give rise to a POE of the small intestine above the organic obstruction. The number of obstructive organic episodes before intestinal diversion also appears to be important. This study demonstrates that prenatal and postnatal organic intestinal obstruction in infants without a short bowel could lead to a chronic POE of the small intestine above the mechanical obstacle. The lifesaving operative procedure used, transient intestinal diversion, allowed identification of the mechanism involved in this enteropathy as a chronic alteration of the intestinal peristalsis, despite a normal appearance and sufficient digestive and absorptive capacities of the intestine above the obstacle. These data require manometric and electromyographic investigations, as well as a histological study of full-thickness specimens of the remaining gut to investigate the muscle and nerve elements and in situ determination of hormone and neuromediators for a better evaluation of these motor disorders. Such studies could improve the medical and surgical management of this enteropathy.
REFERENCES I. Benson CD, Lloyd JR, Smith JD. Resection and primary anastomosis in the management of stenosis and atresia of the jejunum and ileum. Pediatrics 26:265-272, 1960 2. Thomas CC, Carter JM: Small intestinal atresia: The critical role of a functioning anastomosis. Ann Surg 179:663-670, 1974 3. Weber TR. Vane DW. Grosfeld JL: Tapering enteroplasty in infants with bowel atresia and short gut. Arch Surg 117:684-689, 1982 4. Nixon HH: Intestinal obstruction in the newborn. Arch Dis Child 30:13-22, 1955 5. Louw JH: Resection and end-to-end anastomosis in the management of atresic and stenosis of the small bowel. Surgery 62:940-950. 1967 6. Filler RM, Eraklis AJ, Rubin WE, et al: Long term total parenteral nutrition in infants. N Engl J Med 281:589-594, 1969 7. Vargas JH. Ament ME, Berquist WE: Longterm home
parenteral nutrition in pediatrics. Ten years of experience patients. J Pediatr Gastroenterol Nutr 6:24-32, 1987
in 102
8. Grosfeld JL, Reslorca FJ, West KW: Short bowel syndrome infancy and childhood. Am J Surg 200:653-657, 1986
in
9. Cloutier R: Intestinal smooth muscle response to chronic obstruction: Possible applications in jejuno-ileal atresia. J Pediatr Surg 10:3-S, 1975 10. Touloukian mental jejuno-ileal
RJ: Antenatal intestinal adaptation atresia. J Pediatr Surg 13:468-474,
with experi1978
11. Canarelli JP, Leandri J: Longterm modifications in unequal size small intestine anastomosis in the rat. J Chir (Paris) 119:455461,1982 12. Doolin EJ, Ormsbee HS. Hill JL: Motility intestinal atresia. J Pediatr Surg 22:320-324, 1987 13. Cezard
JP. Conklin
abnormality
KA, Das BC, et al: Incomplete
in
intracel-
1432
lular forms of intestinal surface membrane sucrase-isomaltase. J Biol Chem 254:8969+X975,1979 14. Ahnen DJ, Santiago NA, Cezard JP, et al: Intestinal amino-oligo-peptidase. In vivo synthesis or intracellular membrane of rat jejunum. J Biol Chem 257:12129-121351982 15. Evans L, Grasset E, Heyman M: Congenital selective malabsorption of glucose and galactose. J Pediatr Gastroenterol Nutr 4878-886, 1985 16. Bingen E, Lambert-Zechowsky N: Technical aspects of the quantitative and differential analysis of the microbial intestinal ecosystem. Dev Pharmacol Ther 7:134-137,1984 17. Ricour C, Duhamel JF, Amaud-Batandier F, et al: Enteral and parenteral nutrition in short bowel syndrome in children. World J Surg 9:310-315,1985
CEZARD ET AL
18. Ricketts RR: Surgical therapy for necrotizing enterocolitis. Ann Surg 200:653-657, 1984 19. Schullinger JN, Mollitt DL, Vinocur CD: Neonatal necrotizing enterocolitis survival, management and complications. Am J Dis Child 135:612-614,198l 20. Gracey M: The contaminated small bowel syndrome: Pathogenesis, diagnosis and treatment. Am J Clin Nutr 32:234-243, 1979 21. Nilsson LO, Kock NG, Lindgren I, et al: Morphological and histochemical changes in the mucosa of the continent ileostomy reservoir. Stand J Gastroent 15:737-747,198O 22. Wells CL, Maddaus A, Simmons RL: Proposed mechanisms for the translocation of intestinal bacteria. Rev Infect Dis 10:958976,1988