Intestinal stenosis following successful medical management of necrotizing enterocolitis

Intestinal stenosis following successful medical management of necrotizing enterocolitis

Intestinal Stenosis Following Successful Medical Management of Necrotizing Enterocolitis By M a r s h a l l Z. S c h w a r t z , C. J o a n Richardson...

4MB Sizes 0 Downloads 0 Views

Intestinal Stenosis Following Successful Medical Management of Necrotizing Enterocolitis By M a r s h a l l Z. S c h w a r t z , C. J o a n Richardson, C. Keith Hayden, L e o n a r d E. S w i s c h u k , and K e n n e t h R.T. T y s o n

Galveston, Texas 9 In t h e p a s t decade, increased clinical a w a r e n e s s and better medical and surgical management of necrotizing e n t e r o c o l i t i s [NEC) has resulted in i m p r o v e d survival. W i t h an increase in t h e n u m b e r of i n f a n t s s u r v i v i n g the acute s t a g e s of NEC t h e sequelae, including i n t e s t i n a l stenosis, h a v e become m o r e a p p a r e n t . In t h e p a s t 5.5 yr, 62 p a t i e n t s w i t h NEC have been treated at o u r i n s t i t u t i o n . Of t h e 28 s u r v i v o r s o f medical management f o r NEC s e v e n patients developed intestinal stenosis. An average o f 23 days elapsed between the r e c o v e r y f r o m NEC and the diagnosis of colonic stenosis. Only three patients manifested s y m p t o m s o f i n t e s t i n a l o b s t r u c tion. T w o patients had blood in their stools and t w o patients w e r e a s y m p t o m a t i c . Five i n f a n t s w e r e managed by p r i m a r y or staged resection of the intestinal stenosis. The remaining two patients w e r e treated n o n o p e r a t i v e l y . Our d a t a s u g g e s t s a high incidence o f i n t e s t i n a l s t e n o s i s (25%) f o l l o w i n g medical m a n a g e m e n t o f NEC. T h e r e is a marked p r e f e r e n c e f o r t h e s t e n o s i s t o o c c u r on t h e l e f t side of the colon. Colon stenoses can exist without s y m p t o m s and r a d i o g r a p h i c a l l y p r o v e n a r e a s o f s t e n o s i s can resolve. W e recommend that all i n f a n t s f o l l o w ing medical management o f NEC have a b a r i u m enema prior to hospital discharge. In selected cases asymptomatic patients with colonic s t e n o s i s m a y not require operative intervention. INDEX WORDS: Necrotizing enterocolitis; i n t e s t i n a l stenosis.

H E clinical and pathological entity now known as necrotizing enterocolitis ( N E C ) was described over a century ago. 1'2 However, only within the past two decades has this devastating process been more clearly defined and characterized in the literature. 3-5 Associated with an increased awareness of N E C has come a

T

marked increase in the number of cases receiving appropriate therapy. Early efforts to improve survival using medical and surgical techniques had little impact on the high mortality rates of 7 0 % - 7 5 % . 4'6 In the past decade earlier diagnosis, better understanding of the pathophysiology, and more effective management have led to a marked decrease in reported mortality rates to 20%-30%. v'8 However, with more neonates surviving the acute phase of NEC, late manifestations or complications are becoming more evident. These include malabsorption, intraabdominal abscess, intestinal fistula and intestinal stricture. The area of narrowing seen on barium enema in patients following N E C has commonly been referred to as a stricture, implying that cicatricial scarring is present. Because fibrosis has not been a universal finding on pathologic examination in the resected specimens, the use of the word "stricture" may be inaccurate. Stenosis is a less specific but more accurate term to describe the area of narrowing seen on barium contrast radiographs. This report presents our experience with the complication of intestinal stenosis following successful management of the acute phase of necrotizing enterocolitis. PATIENT POPULATION AND RESULTS OF THE ACUTE EPISODE OF NECROTIZING ENTEROCOLITIS

Between October, 1974 and March, 1980, 62 patients were treated for N E C at our institution (Fig. 1). The firm diagnosis of N E C was made using the criteria of abdominal distention, hematochezia, and pneumatosis intestinalis. There From the Departments o f Surgery, Pediatrics and Radiolwere 31 infants who required surgical intervenogy, Child Health Center, University o f Texas Medical tion and 18 survivors (Fig. 1 and Table 1). The Branch, Galveston, Texas. Presented before the l lth Annual Meeting o f the Ameri- . . . . . . . remaining 31 infants received medical management consisting of nasogastric suction and broad can Pediatric Surgical Association, Marco Island, Florida, May ~10, 1980. spectrum parenteral antibiotics (usually genAddress reprint requests to Marshall Z. Schwartz, M.D., tamicin 2.5 m g / k g q12 hr and ampicillin 50 Department o f Surgery, University of Texas Medical m g / k g q12 hr). Therapy was initiated as soon as Branch, Galveston, Texas 77550. the diagnosis of N E C was made. In addition, 9 1980 by Grune & Stratton, Inc. 002 2-3468/80/1506~031501.00/0 plain abdominal radiographs and a platelet 990

Journal of Pediatric Surgery, Vol. 15, No. 6 (December), 1980

891

STENOSIS AFTER NECROTIZING ENTEROCOLITIS

.•

Table 2. Data in Patients W i t h Colonic Stenosis

PATIENTS ( 6 2 ) L MEDICAL (31) MANAGEMENT 18.~ , 1 3 c t

i 18~ I STENOSlS (7) 3-~ , 4 o ~

/

\

, 10 WITHOUT STENOSIS (21) I 15~ , 6 o ~ J

Patient A B C D E F G Average

Sex F F M F M M M

Birth Weight (g)

Gestational Age (wk)

2,920 1,260 1,375 2,260 1,890 2,155 1,300

38 34 34 35 32 36 30

1,880

34

Twin Pregnancy No Yes No No Yes No No

SURVIVED (6) I

3~ .30"

I

Fig. 1. Sex distribution and results in patients with necrotizing enterocolitis treated by medical management only or early surgical intervention.

count were obtained every six to eight hours as a means of following the course of the disease process. The infants received support with platelet transfusions if the platelet count was less than 100,000/mm 3, and fresh frozen plasma or 5% albumin to support their intravascular oncotic pressure. There were 28 survivors (90%) in this group. Seven of these 28 infants (25%) developed intestinal stenosis (Fig. 1). Pertinent data from these 7 patients are shown in Tables 2 and 3. An unusually long symptom-free interval (average 23 days; range 0-60 days) was noted between reinstitution of feedings and the diagnosis of intestinal stenosis (Table 3). In fact, two patients (C and F) were discharged from the hospital prior to the diagnosis of colonic stenosis. The indications for barium Study are shown in Table 4. Three patients (C,D, and G) had clinical and radiographic evidence of partial bowel obstruction at the time of the barium study. Two patients (E and F) had blood in their stools and two patients (A and B) had no symptoms suggestive of colonic stenosis. After our experience with Table 1. Results According to Sex Distribution. (Note the high survival rate and low stricture rate in females who only required medical therapy.)

Percentage of total group Survival following surgical management Survival following medical management Stenosis rate following medical management Survival following surgical treatment for colonic stenosis

Males

Females

52%

48%

63%

50%

77%

100%

40%

17%

75%

100%

patient C, who was discharged, readmitted with colonic obstruction, and who ultimately died, we decided to obtain barium studies just prior to discharge on all patients following recovery from NEC. Thus, the two "routine" barium studies on patients A and B (Table 4) were performed following the initiation of this policy. The locations and length of each colonic stenosis are indicated in Table 3. There were 8 stenoses in 7 patients. The descending colon was the site in 5 of 8 stenoses. The midsigmoid colon was the site in two stenoses and the remaining stenosis was located on the left side of the transverse colon. The length of the stenosis was less than 1.5 cm in 6 of the 8 sites. RESULTS OF MANAGEMENT OF COLONIC STENOSES

Results of the management of colonic stenosis are shown in Table 3. Resection with primary anastomosis was utilized in two patients. Patient B had a routine barium enema following resolution of the acute phase of N E C and prior to the initiation of feedings. This study demonstrated a lcm stenosis at the splenic flexure and she underwent elective resection without sequelae. Patient D underwent primary resection after an elective mechanical bowel preparation. An anastomotic leak occurred on the fifth postoperative day necessitating construction of a colostomy and mucous fistula. She subsequently underwent reconstruction without difficulty. Staged resection (i.e., colostomy followed by resection and anastomosis) was attempted in three patients. Two patients had an unremarkable recovery (patients F and G). Patient C was readmitted 60 days following recovery from N E C with evidence of a bowel obstruction. A barium enema demonstrated a previously unknown stenosis in the descending colon (Fig. 2). Imme-

892

S C H W A R T Z ET AL.

Table 3. Patient Data and Results

Patient

Interval from Birth to NEC (days)

Duration of NEC (days)

Interval from Resolution of NEC to Stenosis (days)

A

2

15

10

B

6

15

0

C

7

5

60

D

1.5

16

24

E

24

12

24

F

11

14

20

G

27

25

25

Average

11

15

23

Length of Stenosis (cm)

Symptoms

X-ray Findings

Site(s) of Stenosis*

Procedure

Results

None

Normal

Midsigmoid

1

None

Splenic

1

Primary resection Staged resection Primary resection* None

Alive & well Alive & well Dead

None

Normal

Abdominal distention Abdominal distention Guiac positive stools Hematochezia

Bowel obstruction Bowel obstruction Normal

Abdominal distention

Bowel obstruction

flexure

Normal

Descending colon Descending colon Descending colon Left transverse colon Splenic flexure Descending colon

1 5 5 1 0.5 1.5

Staged resection

Alive & well Alive & well Alive & well

Staged resection

Alive & well

*Primary resection failed resulting in staged procedure.

diately following intestinal resection and a colostomy he developed an obstructive colitis. He had a protracted hospitalization characterized by many complications that unfortunately terminated in death from sepsis. It was this patient that prompted us to do routine barium enemas in infants medically treated for NEC. In two patients with radiographic evidence of colonic stenosis we selected nonoperative management. Patient A was asymptomatic at the time of barium enema (Fig. 3A) that had been performed as a routine procedure. Because her feedings were being tolerated and because she had no evidence of intestinal obstruction on plain abdominal radiographs, it was decided to discharge this patient but maintain frequent outpatient visits. She received follow-up barium enemas with the last study 8 mo after discharge showing partial resolution of the area of stenosis in the midsigmoid colon (Fig. 3A, B, and C). Patient E underwent a barium enema 24 days after the reinstitution of feedings. This infant had developed guiac positive stools and occa-

sional diarrhea thought to be related to milk allergy. The barium enema showed a long area of stenosis involving the descending colon (Fig. 4A). He was changed to a soy protein formula with resolution of his intermittant diarrhea. A barium study performed 2 mo following discharge demonstrated moderate improvement in the stenotic area (Fig. 4B). He was completely asymptomatic at that time. A t h i r d barium enema 2 yr later showed complete resolution of the stenosis (Fig. 4C). DISCUSSION

The etiology of NEC remains obscure. Several comprehensive reviews on this subject have recently been published. 5'7'9 It is generally agreed that the components necessary for the develop-

Table 4. Indication for Barium Enema in Patients Found to Have Colonic Stanosis Patient

Indication for Barium Enema

A

Routine*

B

Routine*

C

Partial b o w e l o b s t r u c t i o n

D

Partial b o w e l o b s t r u c t i o n

E

Guiac p o s i t i v e s t o o l

F

Hematochezia

G

Partial b o w e l o b s t r u c t i o n

* S e e t e x t for e x p l a n a t i o n .

Fig. 2. Patient C demonstrating a previously undiagnosed narrow stenosis in the descending colon ( a r r o w ) 4 1 days following recovery from necrotizing enterocolitis.

STENOSIS AFTER NECROTIZING ENTEROCOLITIS

893

Fig. 3. (A) The initial barium study in patient A demonstrating a narrow stenosis in the midsigmoid colon; (B) Patient A six weeks after discharge showing some improvement in the degree of stenosis (arrows); (C) Barium enema in patient A eight months later showing significant improvement (arrow). She is asymptomatic.

ment of N E C are: (1) ischemia of the intestinal wall or mucosa, (2) presence of intraluminal bacteria, and (3) a substrate (e.g., formula) within the intestinal lumen. 5 That ischemia is a factor in the acute episode is supported by the radiographic finding of pneumatosis intestinalis and the gross appearance of the involved intestine in those patients requiring surgery. Pneumatosis intestinalis is thought to be the result of bowel wall invasion by gas forming bacteria, j~ This phenomenon is presumed to occur because of loss of the protective mucus coating. 1~'~2 Recovery of the bowel may be complete or result in varying degrees of stenosis. This hypothesis was demonstrated by Louw and Barnard ~3 when they produced intestinal atresia following permanent ischemia in the absence of bacteria. More recently, Tibboel et al. ~4showed that transient ischemia for as short as 5 min can result in intestinal stenosis in chick embryos. Under their experimental conditions, hypoxia without vasospasm did not produce intestinal stenosis. 15

The pathologic findings in the resected intestinal segments have been discussed in the recent li'terature) s-~7 In general, the microscopic evaluation of the bowel at the site of stenosis has revealed a loss of normal colonic mucosa with replacement by granulation tissue and fibroblas~ tic proliferation. However, Tonkin et al. ~8 reported three patients in which the pathologic findings differed from this typical pattern. One patient, who has undergone previous diversion, had a barium enema prior to colostomy closure. This study, performed 4 mo after the acute episode, demonstrated a "stricture" of the descending colon which remained unchanged after intravenous glucagon. Histologic examination of the resected specimen showed only focal superficial ulceration of the mucosa, edema of the submucosa but no significant fibrosis. Two additional cases presented similar pathology. It is difficult to explain these histologic findings especially in the patient who was 4 mo beyond the acute illness. DeSa has suggested an "incom-

894

SCHWARTZ ET AL.

Fig. 4. (A) Barium enema in patient E performed 14 days following discharge showing a 5 cm stenosis in the descending colon; (B) Marked improvement in the appearance of the descending colon in patient E 2 mo later; (C) Barium enema in patient E at 2 yr of age showing complete resolution of the long stenosis.

plete phase" of regeneration of the ischemic mucosa that precedes stricture formation but may result in dysfunction. ~9 Whether this intermediate phase could exist for 4 mo is unknown. However, this phenomenon m a y explain complete or partial resolution of the colonic stenosis

in two of our patients. In addition to our two patients, resolution of post-NEC colonic stenosis has been reported by others. 8'~8 Colonic stenosis following N E C presents other features which warrant discussion. Males have a greater tendency to develop intestinal stenosis

STENOSIS AFTER NECROTIZING ENTEROCOLITIS

than females. Among the surviving, medically treated males, 40% (4/10) developed stenosis whereas only 17% (3/18) of the females developed colonic stenosis (Table 1). There is a striking preference for the left colon. All of our

895

patients had stenoses limited to this area. Other investigators have noted this tendency. ~8'2~ Approximately 73 patients with intestinal stenosis following NEC (including our series) have been reported in the literature. 8'9'15'16J8'2~ Over

Fig. 5. (A) Barium study in patient F showing 2 areas of stenosis (1 and 2) that are b e t t a r demonstrated in Fig. 5B and C; (B) Transverse colon stenosis labeled number 1 in Fig. 5A; (C) Splenic flexure stenosis labeled number 2 in Fig. 5A.

896

60% of these stenoses were located in the left colon. Among the medically treated patients, it is common to identify only 1 area of stenosis that is often less than 1-1.5 cm in length. This is in comparison to the more extensive involvement of the colon (and small bowel) during the acute process. This observation has been discussed in other reports, s'2~An explanation for the apparent discrepancy may be that the degree of bowel wall involvement is less extensive in those patients who do not require surgical intervention. Also, without proximal diversion, the fecal stream may act as a dilating mechanism for those areas of the intestinal wall that might have the potential to stricture. However, long and multiple strictures do occur as seen in patients E and F (Figs. 4A and 5A, B, and C). In our patient series several important clinical features of colonic stenosis were identified. The interval between recovery from the acute phase of NEC and the diagnosis of colonic stenosis is prolonged beyond the normal time frame for wound healing and scar formation. In 5 of our 7 patients the diagnosis was made more than 20 days following recovery. Whether the prolonged interval was related to delay in formation of the stenosis or delay in diagnosis is not known. It is most likely related to a delay in diagnosis. Two patients had no symptoms, and thus, the diagnosis could have been further delayed or never made. Because neonates are on a low residue liquid diet, the area of stenosis might have to be severe before symptoms may result. This raises a question concerning the indication for barium studies in all infants who have recovered from the acute phase of NEC. There is general agreement that those infants requiring emergency surgery (which usually includes construction of a stoma) have a contrast study of the defunctionalized bowel prior to reconstruction. On the basis of our data which reveals a 25% incidence of stenosis in those patients following medical therapy, we would strongly suggest that this group of patients undergo a barium enema prior to discharge from the hospital or within the early follow-up period. We support this recommendation on the basis of the significant number of infants in our series with colonic stenosis but without evidence of bowel obstruction (4/7).

SCHWARTZ ET AL.

Two of these patients (F and C) were initially discharged only to return 14 and 41 days later with indications for a contrast study. Patient C ultimately died from complications related to colonic obstruction. It is possible that the incidence of stenosis may even be higher since only 9 of 28 infants surviving medical therapy had a barium study. Two patients in our series with the diagnosis of colonic stenosis were followed in the outpatient clinic because they had minimal symptoms. Patient A had partial resolution of her stenosis 8 mo following diagnosis (Fig. 3C) and patient E had improvement at 2 mo following diagnosis. A barium enema in patient E 2 yr later showed complete resolution (Fig. 4C). This raises a second question, i.e., should asymptomatic infants with colonic stenosis undergo resection? Successful medical management has been employed in at least four other patients. 8'18'21'26In our opinion, each patient must be evaluated individually. To consider nonsurgical management several criteria must be full-filled; (1) no clinical symptoms that could be associated with partial colonic obstruction (i.e., poor feeding, malabsorption or diarrhea), (2) no radiographic evidence of partial intestinal obstruction, (3) minimal to no colonic dilatation proximal to the site of stenosis on barium enema, and (4) the availability of frequent patient follow-up by a physician. If the above criteria can be met, then nonoperative management can be met, then nonoperative management appears valid. However, if signs or symptoms of partial bowel obstruction occur, then proximal diversion or segmental colon resection with primary anastomosis should be undertaken. The identification of intestinal stenosis is likely to increase with a greater appreciation of the frequency of its occurrence following the acute phase of NEC. With the initiation of routine barium enema examinations, the incidence of 25% we recorded may be exceeded. Surgical intervention is recommended for those infants who have symptoms referable to their colonic stenosis. For those infants who are asymptomatic, nonoperative management is more controversial. In selected patients, close observation may be sufficient.

STENOSIS AFTER NECROTIZING ENTEROCOLITIS

897

REFERENCES

I. Simpson JY: Peritonitis in the fetus in uterus. Edinburgh Med Surg J 15:390, 1838 2. Siebold JP: Gerburtshulfe Frauenzimmer und Kinderkrankheiten, Heft I. Leipzig 5:3, 1825 3. Waldhausen JA, Herendeen T, King H: Necrotizing colitis of the newborn: Common cause of perforation of the colon. Surgery 54:365-372, 1963 4. Mizrahi A, Barlow O, Berdon W, et al: Necrotizing enterocolitis in premature infants. J Pediatr 66:697-706, 1965 5. Kosloske AM: Necrotizing enterocolitis in the neonate. Surg Gynecol Obstet 148:259-268, 1979 6. Touloukian R J, Berdon WG, Amoury RA, et al: Surgical experience with necrotizing enterocolitis in the infant. J Pediatr Surg 2:389-401, 1967 7. Touloukian R J: Neonatal necrotizing enterocolitis: Surg Clin North Am 56:281-298, 1976 8. Bell M J, Ternberg JL, Askin FB: Intestinal stricture in necrotizing enterocolitis. J Pediatr Surg 11:319-327, 1976 9. Santulli TV, Schullinger JN, Heird WC, et al: Acute necrotizing enterocolitis in infancy: A review of 64 cases. Pediatrics 55:376-387, 1975 10. Engel RR, Virnig NL, Hunt CE, et al: Origin of mural gas in necrotizing enterocolitis, abstracted. Pediatr Res 7:292, 1973 11. Touloukian R J, Posch JN, Spencer R: The pathogenesis of ischemic gastroenteritis of the neonate: Selective gut mucosal injury in asphyxiated neonatal piglets. J Pediatr Surg 7:194 205, 1972 12. Bounous G, McArdle AH, Hodges DM, et al: Biosynthesis of intestinal mucin in shock. Ann Surg 164:13-22, 1966 13. Louw JH, Barnard CN: Congenital intestinal atresia: Observations on its origin. Lancet 2:1065-1067, 1955 14. Tibboel D, Cornelius J, Molenarr J: The effects of temporary general hypoxia and local ischemia on the development of the intestines: An experimental study. J Pediatr Surg 15:57-62, 1980 15. DeSa DJ, Muclow ES, Gough MH: Neonatal gut infarction. J Pediatr Surg 5:454-459, 1970

16. Woodard BH, Bradford WD, Filston HC: Asymptomatic cecal stenosis after necrotizing enterocolitis. South Med J 72:1613-1615, 1979 17. Joshi VV, Winston YE, Kay S: Neonatal necrotizing enterocolitis. Am J Dis Child 126:113-116, 1973 18. Tonkin IL, BJelland JC, Hunter TB, et al: Spontaneous resolution of colonic strictures caused by necrotizing enterocolitis: Therapeutic implications. Am J Roentgenol 130:1077-1081, 1978 19. DeSa D J: The spectrum of ischemic bowel disease in the newborn. Rosenberg HS, Bolande RP (eds): Perspectives in Pediatric Pathology. Chicago, Yearbook Medical Publishers, 1976, pp 273-309 20. Costin BS, Singleton EB: Bowel stenosis as a late complication of acute necrotizing enterocolitis. Pediatr Radiol 128:435-438, 1978 21. Beck JM, Dinner M, Chappel J: Enterocolitis following exchange transfusion. South Afr J Surg 9:39-43, 1971 22. Stevenson JK, Graham CB, Oliver TK, et al: Neonatal necrotizing enterocolitis: A report of twenty-one cases with fourteen survivors. Am J Surg 118:260 272, 1969 23. Krasna IH, Becker JM, Schneider KM, et al: Colonic stenosis following necrotizing enterocolitis of the newborn. J Pediatr Surg 5:200-206, 1970 24. Kosloske AM, Martin LW: Surgical complications of neonatal necrotizing enterocolitis. Arch Surg 107:223 227, 1973 25. Lloyd DA, Cywes S: Intestinal stenosis and enterocyst formation as late complications of neonatal necrotizing enterocolitis. J Pediatr Surg 4:479-486, 1973 26. Torma M J, Delemos RA, Rogers JR, et al: Necrotizing enterocolitis in infants: Analysis of forty-five consecutive cases. Am J Surg 126:758-761, 1973 27. Cockrill H, Oliphant M, Grossman H: Intestinal strictures in the neonate. A complication of ischemic bowel disease. Am J Roentgenol 123:764-769, 1975 28. Virjee JP, Gill G J, DeSa D, et al: Strictures and other late complications of neonatal necrotizing enterocolitis. Clin Radiol 30:25 31, 1979

Discussion A.M. Kosloske (Albuquerque, New Mexico): R e c e n t l y w e r e v i e w e d o u r e x p e r i e n c e w i t h stricture following necrotizing enterocolitis and a d d r e s s e d t h e s a m e q u e s t i o n s as D r . S c h w a r t z a n d his c o l l e a g u e s . W e m u s t r e e m p h a s i z e a v e r y i m p o r t a n t m e s s a g e : s t r i c t u r e a f t e r N E C is n o t a benign disease; stricture after NEC may be f a t a l . I n b o t h S c h w a r t z ' s s e r i e s a n d in o u r own, a n i n f a n t died. I n B e l l ' s c o l l e c t e d s e r i e s o f strict u r e a f t e r N E C t h e m o r t a l i t y w a s 18%. M o s t o f t h e d e a t h s w e r e d u e to c o m p l i c a t i o n s o f i n t e s t i nal obstruction, such as aspiration pneumonia. In many infants the diagnosis was delayed, and

in a few, t h e o n s e t o f s y m p t o m s w a s a b r u p t , f u l m i n a n t , a n d f a t a l . I n view o f t h i s 1 8 % m o r t a l ity, I t h i n k we m u s t c o n s i d e r s u r g e r y as t h e c o n s e r v a t i v e t r e a t m e n t for c o l o n i c s t e n o s i s . A n y n o n - o p e r a t i v e m a n a g e m e n t , e v e n in a n a s y m p t o m a t i c p a t i e n t , s h o u l d b e c a r r i e d o u t a c c o r d i n g to a c a r e f u l e x p e r i m e n t a l p r o t o c o l , s u c h as t h e p r o t o c o l o f D r . S c h w a r t z . W e a g r e e t h a t all infants deserve a barium enema whether they w e r e t r e a t e d m e d i c a l l y or s u r g i c a l l y . W e o b t a i n the barium enema about 6 wk after resolution of NEC rather than at the time of discharge. That is n o t t o o e a r l y to m i s s t h e e a r l i e s t x - r a y f i n d i n g s ,

898

but not too late to make the diagnosis before complete obstruction occurs. H. Firor (Lubbock, Texas): Stricture to a surgical mind means something which must be resected or dilated. Strictures following N E C may be different. We reviewed four strictures resected following defunctionalizing ileostomies for NEC. One showed obliteration of the lumen; the other three showed mucosal ulceration and submucosal edema without noticeable fibrosis. We have managed five children with colonic strictures nonoperatively. C o n t r a r y to Dr. Schwartz and Tyson's experience, all were symptomatic. Management was by a period of intravenous alimentation, then elemental diets, and finally regular formula. This was reported in the American Journal of Radiology (June, 1978). This 1400-g female infant has the classical findings of N E C that developed at 8 days. At 3 wk, she developed symptomatic abdominal distension. A flat film suggested obstruction, and this is her barium enema. These strictures persist on multiple views. Treated in the manner described at 5 mo this child is asymptomatic. This child developed a single stricture five weeks after conclusion of therapy. She responded to graduated feeding and at 5 mo is thriving. These other children have had similar findings and clinical courses. I present this to raise questions: First, are all strictures after N E C fibrous contraction bands or are some of them lesser entities? Second, is nonoperative management appropriate in some of these strictures? I am not ready to close a stoma, in the face of a distal stricture. I think that would be foolhardy.

R.J. Touloukian (New Haven, Connecticut): Perhaps Dr. Schwartz would comment upon the incidence of undetected colonic strictures developing in babies previously operated upon for severe ileal involvement and who have an: ileal exteriorization. They apparently do not have severe enough colonic involvement to warrant resection o r colostomy at the initial procedure. This is a group that we are particularly concerned about and I think the incidence of undetected colonic obstruction may be higher than in those patients who have not had any surgical therapy. We routinely perform a barium enema before the closure of any ileal exteriorization to avoid the potential complication of an undetected colonic stricture.

SCHWARTZ ET AL.

M. Gauderer (Cleveland, Ohio): We would like to enlarge upon what has been said with two slides illustrating a patient with neonatal necrotizing enterocolitis who had undergone resection 7 mo before referral. The attending surgeon a t another institution failed to obtain a barium enema prior to the "hook up" procedure at 5 mo of age and the anastomosis was done proximal to an area of ascending colon with severe stricture secondary to necrotizing enterocolitis. When the anastomosis broke down, the child was referred to us. The next slide shows a cross section of the subsequently resected segment of colon. There is no lumen making it a complete postinflammatory acquired atresia. After a difficult course, the infant survived. T. Santulli (New York, New York): It is important to emphasize that 20%-25% of patients who have recovered from acute N E C with or without operation will develop a stricture as demonstrated by x-ray. We do not routinely operate on patients who show such a stricture by x-ray unless they are symptomatic. However, as Dr. Touloukian and others have already said, any patient who has a double ostomy and is to have closure must first be investigated. If a stricture does show on x-ray we would rather not challenge such a patient but take care of the stricture before reestablishing continuity of the intestine. D. Shermeta (Baltimore, Maryland): Just a word of caution. We had two infants in the last 3 yr who were successfully nonoperatiVely m a n aged N E C babies who had barium enemas prior to being discharged who returned subsequently with, one, a total atresia of the terminal ileum, and the second one a severe stenosis of the terminal ileum. So the presence of a normal barium enema doesn't get you totally Off the hook. L. Martin (Cincinnati, Ohio): Previous discussants of this well-presented manuscript allude to an experience that appears to be contradictory to that of the authors. This apparent discrepancy of observations can most likely be explained by a simple reminder of the basic principles of wound healing. The same principles of wound healing apply to a wound of the skin, a third-degree burn, caustic burns of the esophagus and intestinal insult secondary to N E C . Epithelial destruction is followed by healing by secondary inten-

STENOSIS AFTER NECROTIZING ENTEROCOLITIS

tion. Granulation tissue fills the defect, which, if it involves most of the circumference of the bowel lumen, will result initially in an "inflammatory stricture." With further healing, the inflammatory stricture may resolve, or it may eventually progress to a cicatricial stricture. A clinically significant stricture 6 mo or more following the initial episode of N E C will be cicatricial and will not resolve; whereas one which occurs earlier may be inflammatory, and may resolve with nonoperative management. In the event of circumferential loss of mucosa and muscularis over a significant distance, the two ends may heal over completely occluding the lumen, in which case, the findings at operation resemble those of congenital atresia. M. Schwartz (closing): It is comforting to have taken a "middle of the road" approach. I would be reluctant to pursue the nonoperative approach to symptomatic patients proposed by Dr. Firor. Severe obstructive colitis could be a sequela. Dr. Koslowske's recommendation of resection in all patients with asymptomatic

899

colonic stenosis is likely to result in unnecessary surgery in a small but significant group. Dr. Martin's comments concerning the healing process are very relevant. I was very careful during the presentation not to refer to the radiographic finding of intestinal narrowing as a stricture. Stricture implies fibrosis which may not be present. Dr. Firor and his colleagues noted that in some of their resected specimens granulation tissue without fibrosis was responsible for the stenosis. It is conceivable that if the granulation tissue is confined to the mucosal layer that resolution of this tissue will result in healing without fibrosis or stricture. We would agree with Dr. Touloukian's comment that a barium study of the decompressed colon be performed prior to stoma closure. I believe that colonoscopy, suggested by Dr. Schuster might differentiate between fibrosis and granulation tissue in those infants with colonic stenosis that will be managed nonoperatively. However, this assumes that the patient is of sufficient size to undergo the procedure.