Bedside peritoneal drainage:A primary treatment for perforated necrotizing enterocolitis

Bedside peritoneal drainage:A primary treatment for perforated necrotizing enterocolitis

T Gobran,MD; M Khalifa,MD; A Waly,MD; Salah M AbdEIAal,MD;

Kamal abd El-elah Aly, MD

Department of Pediatric Surgery, Zagazig and Mansoura Universities, Egypt.

Abstract

Background and purpose: Necrotizing enterocolitis is one of the leading causes of morbidity and mortality in neonatal intensive care units particularly in places without neonatal surgical facilities. The best treatment for perforated necrotizing enterocolitis is uncertain. Bedside peritoneal drainage has been used as primary treatment in the management of perforated necrotizing enterocolitis. This study aimed to compare outcomes of bedside peritoneal drainage done by general surgeons as a primary procedure for the treatment of preterm and/or low birth weight neonates with perforated necrotizing enterocolitis to outcomes of early laparotomy performed by pediatric surgeons as regards effectiveness  of the procedure as a definitive treatment, the need for delayed laparotomy and mortality rate.

Patients and methods: Fifty cases of preterm and/or low birth weight neonates with perforated necrotizing enterocolitis were randomly assigned to one of two groups. Group I was managed by bedside peritoneal drainage done by general surgeon while laparotomy was reserved for non responding cases. Group II was managed by early laparotomy done by pediatric surgeon. Data collected from all cases included birth weight (g), gestational age (weeks), gender, age at operation (bedside peritoneal drainage or early laparotomy). Delayed laparotomy was performed for infants who developed persistent fecal.fistula or suffered late intestinal stricture as well as for closure of enterostomy. Outcomes of bedside peritoneal drainage and early laparotomy were recorded and statistically compared regarding the effectiveness of procedure as a definitive treatment, need for delayed laparotomy and mortality.

Results: 16 cases (64%) showed clinical improvement after bedside peritoneal drainage. Bedside peritoneal drainage in Group I and early laparotomy in Group II were effective as a definitive treatment in 10 cases (40%) and 15 cases (60%) respectively. Delayed laparotomy was indicated in 7 cases(28%) in Group I and 8 cases (32%) in Group II.Mortality was recorded in  8 patients (32%)  in  BPD  group  and  in  9  patients (36%)  in  laparotomy group.

Conclusion: According to this study, outcomes of bedside peritoneal drainage as a primary treatment for low birth weight and/or preterm neonates with perforated necrotizing enterocolitis showed no significant statistical difference as regards the need for delayed laparotomy and mortality rate when compared to the outcomes of early laparotomy as a primary treatment for the same conditions. Bedside peritoneal drainage provides a useful primary procedure for the management of preterm and low birth weight neonates with perforated necrotizing enterocolitis particularly in healthcare facilities without neonatal surgery capacity.

Key words: Necrotizing enterocolitis, peritoneal drainage, neonatal intestinal perforation.

Introduction:

Necrotizing enterocolitis (NEC) is the most common gastrointestinal emergency that affects the newborn. IA total of 90% of NBC cases occur in premature infants.2 NBC is one ofthe

leading causes of morbidity  and mortality in neonatal intensive care units (NICU)_3,4  Over the last two decades  the mortality rate from NBC remained  steady  at 20o/o-50% even in developed countries.5,6

Although pathogenesis of the disease is not fully understood, in severe cases, NEC rapidly progresses from  bacterial invasion of  the intestinal wall to full-thickness bowel necrosis, leading to perforation and  subsequent peritonitis, sepsis, and  possibly death.7

Traditional surgical management of neonatal bowel perforation secondary to NEC has been based  on established surgical principles of laparotomy (LAP), resection of necrotic bowel, debridement, and exteriorization.8 In 1977, Ein et al9 recommended bedside  peritoneal drainage (BPD) as a "temporizing" procedure for perforated NEC in very low birth weight neonates. This  operation was  designed  for patients  considered too unstable to undergo LAP. Since the introduction of BPD, there have been several reports which suggested that BPD may  serve as a definitive therapy. In many centers, BPD has become the routine approach regardless of severity of the underlying illness)O,ll Some reports challenge the role ofLAP as an initial choice of surgery and suggest that bowel perforation secondary to NEC  needs  a far  less  invasive surgical approach for adequate initial management.lO,12

In our  region  (Sharkeya and  Mansoura Governerates- Egypt) there is a single pediatric surgery unit in each governate. Although these units serve a big region with heavy population, they suffer limited resources.Frequently, there is lack ofbeds and unfeasibility to accept cases with perforated NEC. These factors in addition to high mortality rate of cases with perforated NEC and the challenge of optimal  surgical treatment prompted  us to inquire  about the value ofBPD performed by general surgeons as a primary treatment for  neonates with perforated NEC.

Purpose:

This work aimed to compare outcomes of bedside peritoneal  drainage done by general surgeons as a primary procedure for the treatment of preterm and/or low birth weight neonates with perforated necrotizing enterocolitis to outcomes of early laparotomy performed by pediatric surgeons as regards effectiveness of the procedure as a definitive treatment, the need  for delayed  laparotomy and mortality rate.

Patients and methods:

This  randomized clinical study was conducted in  NICU-Faculty of  Medicine­ Zagazig  and  Mansoura Universities in the period  from  January 2007  to June 2010. It included 50 cases ofpreterm and/or low birth weight (LBW) neonates with perforated NEC. Patients were randomly assigned to one of two groups:

Group  I (25 cases): underwent  BPD as a primary treatment done by general  surgeon. Early laparotomy was reserved for patient who showed no  or  limited response to  BPD.

Group II  (25  cases): underwent early laparotomy performed  by pediatric surgeon.

Criteria for exclusion of patients from this study included full term patients, normal birth weight,  associated anomalies and  parent's refusal.

Diagnosis of intestinal perforation was confirmed by plain X-ray of the abdomen that revealed free  air  in  the  peritoneal cavity (pneumoperitoneum).

Parents were counseled  and asked by the attending surgeon to provide written informed consent.

BPD  was performed in NICU under complete sterile conditions by general surgeon. Using local anaesthesia, full thickness incision (O.Scm) was made in the right or left lower quadrant  of the abdomen  and the peritoneal cavity was entered under direct vision. A 1OF catheter was inserted carefully into peritoneal cavity, and peritoneal fluid was collected for microbiologic cultures. The catheter was fixed at the skin and the end left free in a collection bag. The peritoneal cavity was then irrigated with warm saline solution until retrieved fluid became  clear.  If the peritoneal cavity  was believed to be inadequately drained, as evidenced  by re-accumulation of air or fluid in the abdomen, the original drain was manipulated or an additional drain was inserted to help better peritoneal drainage.

Cases  randomly assigned to  early  LAP underwent abdominal exploration through a transverse abdominal incision. Necrotic intestine was resected.  Intestinal  stomas  or primary anastomosis was performed depending on site, extent of intestinal necrosis and general condition of  patients. Evidence of further intestinal necrosis or  perforation was  the

indication for additional early laparotomies.

The patients  were closely  monitored  in

NICU and reassessed by surgeons every 12 to

24 hours. Afterward management of group I depended on the clinical course of the patients as follows:

1.Clinical improvement and stability: It meant haemodynamic stabilization, decrease abdominal distension, decrease abdominal wall cellulitis, no palpable mass, less painful abdominal palpation and no further pneumoperitoneum.Inthese situations, the peritoneal drain was removed when it no longer  discharged fluid  or after the development of a fecal fistula.

2. Clinical deterioration: It was based on the joint clinical judgment of the attending surgeons. It meant increased abdominal distension and/or discoloration, persistent or increasing pneumoperitoneum, palpable mass, signs  of persistent intestinal obstruction and no passage of stools. In these situations, early LAP was performed within 12 to 24 hours from insertion  of peritoneal drain.

Delayed LAP was performed when the patient developed persistent fecal fistula or suffered late intestinal stricture proved clinically and radiologically.

Data Collected from all cases included birth weight (g), gestational age (weeks), gender,

and age at operation  (BPD or early LAP). Routine data recorded at NICU included medications given  at onset  of NEC or perforation, need for mechanical ventilation, Pa02/Fi02, urine output (ml/Kglhour)  and

mean arterial pressure.

The outcome variables in this study were the need for delayed LAP and mortality rate hence to compare the outcome ofBPD versus early  LAP  as  the  primary   treatment for perforated NEC in preterm and/or LBW neonates.

Definitive treatment was defined in this

work as the treatment that resulted in improvement of the patient clinical condition, recovery of normal gastrointestinal function and normal enteral feeding.

Data management:

Data were collected, tabulated and finally analyzed using Epi-6 and Statistical Package for Social Science (SPSS) version 19 statistical

programs. Descriptive statistics were used such as percentage, arithmetic mean and standard deviation.

Statistical tests of significance were used

to compare between the studied groups as Chi­ square test and t-test. P value 0.05 was considered significant difference and P value

0.01 was  considered highly  significant difference.

Results:

This study included 29 boys (58%) and 21 girls (42%). The mean gestational age was

30.6 weeks and 31.4 weeks in group I and group II respectively. The mean birth weight was 1280.9 grams and 1290.3 grams in groups I and II respectively. The mean Apgar score at 1 minute and 5 minutes was 5.4 and 7.2 for group I and 6.3 and 8.5 for group II. The mean age at operation was 13.3 days and 13.7 days in group I and group II respectively Table(l).

In group I the range of peritoneal drain

duration after excluding the cases that required early LAP was 10 -14 days with a mean of

12.6 days.

In group I 16 cases (64 %) showed initial clinical improvement after BPD. Clinical deterioration occurred in 3 cases while clinical improvement was limited in 6 cases. These 9 cases underwent early LAP (36%). Three cases needed intestinal resection and with primary anastomosis and 6 cases  ended with enterostomies.

Eight cases died in group I with mortality

rate of 32%. Five cases died after early LAP because of massive NEC, 2 cases of unknown cause  and  1 case  because of type  IV intraventricular haemorrage. Seven  cases needed delayed LAP (28%) for intestinal stricture in 3 cases, fecal fistula in 3 cases and one case for closure of stoma. BPD was effective as a definitive treatment in 10 cases

{40 %) after excluding 3 cases that died after initial clinical improvement and 3 cases who developed persistent fecal fistula.

In group II early LAP and damaged intestinal resection was performed followed by primary intestinal anastomosis in 17 cases and diverting stoma in 8 cases.  Nine cases died with mortality rate of 36%, 2 of them were intraoperative and 7 cases were during early postoperative course. Eight cases required

delayed LAP (32%), Table(2).It was indicated in 5 cases for closure of stoma, 2 cases due to intestinal stricture and  1 case due  to development of fecal fistula. Early LAP was effective as a definitive treatment in 15 cases (60 %) after excluding one case that developed early postoperative fecal fistula. No cases in this group required further early laparotomies.

In group I early LAP was indicated  for 9 cases (9/25) because of no or limited response to BPD.When we compared that to performing early LAP as the primary treatment in all cases of group II, we found statistically  significant privilege to BPD.

Table (1): Demographic data of the cases included in the study.

Table (1): Treatment outcomes of the study groups.

In this  study  5 cases  died  out of 9 who underwent early LAP because of no or limited response to BPD. When  this result was compared to mortality after BPD only (3/16) we found that the difference was statistically insignificant (X2 = 3.6 and P value= 0.06). When mortality after BPD and early LAP was compared to that after early LAP only (9/25), the difference  was statistically  insignificant, (X2 = 1.04 and P value= 0.3).

Discussion:

Despite decades of clinical advances and research, the management  ofNEC continues to offer a considerable challenge to the pediatric surgeon.13,14 Exploratory  LAP or peritoneal

drainage,  which is the favorable choice as a primary treatment for  perforated NEC  in neonates? The answer  for this question  has been controversial.

The  treatment of  perforated NEC  by

exploratory LAP carries high significant risk in preterm and LBW neonates.8 On the other hand, effective BPD will hypothetically draw off  peritoneal gas, pus,  and  stool, thus minimizing operative stress  and  avoiding intestinal resection. This makes  BPD an attractive alternative to LAP.9 Moreover, BPD has advantages of being bed side procedure, cheap, technically easy, does not need operating room, or general anesthesia, and may not entail stoma  or  second  operation. Nevertheless,

evaluation of disease process and extent of bowel  involvement is  incomplete by  this approach.l5

Several regions in Egypt suffer lack of pediatric surgery facilities. Pediatric Surgery Unit of either Zagazig or Mansoura Universities is the sole neonatal surgical unit in their region with limited resources and inability to accept all referred cases  of perforated NEC from NICUs of other hospitals. It is even difficult sometimes to reach  the  patients at distant hospitals. For this reason this study has emerged as a trial to study if there is a possibility of using BPD performed by general surgeon as a primary  treatment for perforated NEC in these hospitals aiming to save those neonates.

In this study a group of preterm and LBW

neonates who suffered perforated NEC were managed by BPD done by general surgeons at our  hospital as  a primary treatment. Effectiveness of this procedure as a definitive treatment, the  need  for  delayed LAP  and mortality were compared to another group of similar patients with perforated NEC who were managed  primarily  by early LAP performed by pediatric surgeons.

Although many studies have reported  the outcomes of treatment with BPD or LAP, these reports contain significant bias influencing the patient selection  and it is not possible,  even through meta-analysis,  to determine whether BPD or LAP is the better  technique for any size infant.10,16 In the present work, statistical testing  of demographic data of both groups proved no statistically significant  difference between both groups' criteria, which supports unbiased patient selection in  the  study.

Ein and colleagues argue  that LAP with intestinal resection is associated with high rates of morbidity and mortality in very LBW infants. Therefore, they  recommend BPD  as  the preferred initial procedure for neonates with perforated NEC.9

BPD procedure was initially introduced as a method  for pre-operative resuscitation of critically  ill infants with complicated NEC. However, sometimes it was used as the definitive procedure for treatment of neonates with perforated NEC.J7,18

Several  authors  have advised BPD  as a "temporizing procedure", for  all  children weighing less than 1500 grams and for unstable

babies more than 1500  grams, followed  by

LAP in two to three days.19-22

BPD is preferred by some surgeons because they are cautious  against the risk of waiting too long before performing a LAP. Dimmitt and  colleagues suggest  that  death  may  be avoided by  performing a  timely  LAP.24

Takamatsu and colleagues have recommended BPD as definitive strategy for a selected group of extremely LBW babies.25 Inthis study early LAP was performed if the patients did not show clinical improvement within 12-24 hours ofBPD.

In1990 Ein and colleagues19 presented their

13-year experience with BPD in 37 patients with bowel  perforation secondary to NEC. Sixty-five percent of these patients weighted less than 1000 grams and 88% less than 1500 grams.  One third of the reported patients recovered completely by BPD.Inthe remaining patients:nine (24%) died rapidly before LAP, nine (24%) underwent  an early LAP (within

24 hours), and 7 (22%) underwent a delayed

LAP for bowel obstruction or fistula formation.

Dimmitt and colleagues treated 26 patients with perforated NEC. Nine cases were treated with LAP and 17 with BPD. Survival rate was similar between LAP and BPD. Four patients in the BPD group underwent salvage LAP for perceived clinical deterioration. All of these patients  died. The clinical  status of patients who had salvages LAP and died was similar to those who did not and lived.24

By reviewing  previous studies it is noted that there is wide variation  in the results of BPD regarding  survival and outcome. Some reports  suggested that BPD  resulted  in the unexpected survival of  the  infants with perforated NEC, with survival rates approaching or exceeding those with LAP,19,25 while  others  suggested that  LAP  was  the superior treatment.23,15

In this  work  mortality rates showed statistically  insignificant difference  between the group treated primarily with BPD, where LAP was reserved for cases that showed no or limited response, and  the group treated primarily with exploratory LAP. Thus BPD in our  view  provides a reasonable option  for healthcare facilities devoid of pediatric surgery capacity  provided the attending physician is trained  to  perform  BPD   in   neonates.

In this study early LAP was indicated in 9 cases of Group I (9/25) because of no or limited response to BPD.When that was compared to performing LAP as the primary treatment for all cases of Group II, we found  statistically highly significant privilege to BPD.This result supports the role ofBPD as primary procedure for neonates with  perforated NEC  and unfavorable conditions for LAP. LAP  can hence be reserved for the cases not responding to BPD which will help saving neonates lives as well  as healthcare resources. This  is supported by Lessin et al recommendation to use BPD as initial management of all LBW infants with complicated NEC before definitive

LAP.l6,26

However, there is a published literature that suggests that peritoneal drainage can  be a satisfactory definitive procedure, particularly in the  very  LBW infant. Morgan, et al mentioned that BPD alone provided definitive surgical  intervention in 74% of cases  with complicated NEC.22

An international multicenter  randomized controlled trial was performed between 2002 and 2006. Sixty-nine patients were randomized (35 drain, 34 LAP). Early LAP was performed in 26/35 (74%) patients.BPD was effective as a definitive treatment in only  4/35  (11%) surviving neonates.27

Romero et al, on their study of 13 cases of perforated NEC (6 in drainage group and 7 in LAP group) concluded that BPD is a temporary stabilizing procedure and  could not  be considered as a definitive surgical treatment.28

In our study BPD was effective as a defmitive treatment  in 10 cases (40%). The explanation of the wide discrepancy of our result from the previously mentioned studies is  the  difficulty to  differentiate between perforated NEC and isolated bowel perforation based on clinical  and radiological fmdings. There  are studies  that seem to indicate  that BPD is more advantageous in neonates with isolated  perforations of the gastro-intestinal tract not related to NEC. For neonates  with perforation caused by NEC,peritoneal drainage may provide temporary stabilization, but most of these infants require subsequent LAP, and few survive.29

One multicentre prospective study reported the overall incidence of postoperative intestinal

stricture at 10.3% and no difference between the initial LAP versus the initial BPD groups.30

However other studies suggest that strictures occur more frequently in patients  who have undergone BPD.5

In this study, intestinal  stricture occurred in 4 cases (16%)  of BPD group and 2 cases (8%) of LAP group. A possible explanation is that the damaged areas which are prone to stricture formation upon healing would have been resected at the time of LAP.

Conclusion:

According to this study, outcomes ofbedside peritoneal drainage as a primary treatment for low birth weight and/or preterm neonates with perforated necrotizing enterocolitis showed no significant statistical difference as regards the need for delayed laparotomy and mortality rate when  compared to the  outcomes of early laparotomy as a primary treatment for the same conditions. Bedside peritoneal drainage provides a useful primary  procedure  for the management of preterm and low birth weight neonates with perforated necrotizing enterocolitis particularly inhealthcare facilities without  neonatal  surgery  capacity.

Recommendation:

BPD is an easy, cheap and affordable option for stabilizing neonates with perforated NEC. We recommend BPD  training for  general surgeons and pediatricians in NICU to help saving neonates with perforated NEC in health care centers without pediatric surgery facility.

References:

1- Stoll  BJ:  Epidemiology of  necrotizing enterocolitis. ClinPerinato/1994; 21:205-

218.

2- Maayan-Metzger A, Itzchak A, Mazkereth R, Kuint J:Necrotizing enterocolitis in full­ tenn infants: Case-control study and review of the literature.JPerinato/2004; 24:494-

499.

3- Hsueh W, Caplan  MS, Qu XW, et al: Neonatal necrotizing enterocolitis: Clinical considerations and pathogenetic concepts. Pediatr  Dev  Pathol 2003; 6:  6-23.

4- Amoury RA: Necrotizing  enterocolitis- a

continuing problem in the neonate. World

JSurg 1993; 17: 363-373.

5- Horwitz JR, Lally KP, Cheu HW,Vazquez WD,Grosfeld          JL,  Ziegler  MM: Complications after surgical intervention for necrotizing enterocolitis: A multicenter review. J Pediatr Surg 1995; 30(7): 994-

998; discussion 998-999.

6- Bisquera JA,  Cooper TR, Berseth CL: Impact of necrotizing enterocolitis on length of stay and hospital  charges in very low birth  weight  infants. Pediatrics  2002;

109(3): 423-428.

7- Kliegman RM, Fanaroff AA: Necrotizing enterocolitis. N Eng! J Med 1984;  310:

1093-1103.

8- Sharma R, Tepas J J 3m, Mollitt DL, Pieper P, Jacksonville PW: Surgical management of bowel perforations and outcome in very low-birth-weight infants (<1,200 g). J Pediatr Surg  2004; 39(2):  190-194.

9- Bin SH, Marshall DO, Gervan D: Peritoneal drainage under local anesthesia for perforations from necrotizing enterocolitis. J Pediatr Surg   1977;  12:   963-967.

10-Rovin JD, Rodgers BM, Burns RC, et al: The role of peritoneal drainage for intestinal perforation in infants with  and without necrotizing enterocolitis. J Pediatr Surg

1999; 34: 143-147.

11-Bysiek A, Palka  J, Pietrzyk JJ, et al: Peritoneal drainage  as an alternative to laparotomy in  premature infants with complicated necrotizing enterocolitis. Przeg Lek 2002; 59: 67-79.

12-Demestre X, Ginovart G, Figueras-Aloy J, et al: Peritoneal drainage as primary management  in necrotizing enterocolitis: A prospective study. JPediatr Surg 2002;

37: 963-967.

13-K.urscheid T, Holschneider AM: Necrotizing enterocolitis (NBC) - mortality and long­ term results.Eur JPediatr Surg 1993; 3(3):

139-143.

14-Patel JC, Tepas JJ, 3rd, Huffman SD, Evans JS:Neonatal necrotizing enterocolitis: The long-term perspective. Am Surg 1998;

64(6): 575-579.

IS-Ehrlich PF , Sato T T, Short B L, Hartman G E: Outcome  of perforated necrotizing enterocolitis in  very  low  birth  weight neonate my be independent of the type of surgical treatment. Am Surg 2001; 67: 752-

756.

16-Lessin MS, Luks  FI, Wesselhoeft CW: Peritoneal drainage as definitive treatment for intestinal perforation in infants  with extremely  low birth weight (<750 gms). J  Pediatr Surg   1998; 33:   370-372.

17-Albanese CT,  Rowe  MI:  Necrotizing

enterocolitis. Semin Pediatr Surg 1995;

4(4): 200-206.

18-Holman RC, Stehr-Green JK. Zelasky MT:

Necrotizin enterocolitis mortality  in the

United States, 1979-85. Am JPublic Hlth

1989; 79(8): 987-989.

19-Ein  SH, Shandling B, Wesson  D, Filler RM:A 13-year experience with peritoneal drainage under local anesthesia for necrotizing enterocolitis perforation. J Pediatr Surg 1990; 25(10): 1034-1036.

20-Ahmed T, Bin S, Moore  A: The role of peritoneal drains in treatment of perforated necrotizing enterocolitis: Recommendations from  recent  experience. J Pediatr Surg

1998; 33(10): 1468-1470.

21-Cheu HW, Sukarochana K, Lloyd  DA: Peritoneal drainage for  necrotizing enterocolitis.JPediatrSurg 1988; 23: 557-

561.

22-Morgan LJ, Shochat SJ,  Hartman GE: Peritoneal drainage as primary management of perforated  NEC in the very low birth weight infant JPediatr Surg 1994; 29: 30-

34.

23-Azarow KS, Ein SH, Shandling B, Wesson D, Superina R, Filler RM: Laparotomy or drain for perforated necrotizing enterocolitis: Who gets  what  and why? Pediatr  Surg  Int  1997; 12:  137-139.

24-Dimmitt RA, Meier AH, Skarsgard ED, et

al:  Salvage laparotomy for  failure of peritoneal drainage in necrotizing enterocolitis ininfants with extremely low birth weight J Pediatr Surg 2000; 35:856-

859.

25-Takamatsu H, Akiyama H, lbara S, Seki S, Kuraya K,  lkenoue T:  Treatment for necrotizing enterocolitis perforation in the extremely premature infant (weighing less than 1,000  g). J Pediatr Surg 1992; 27:

741-743.

26-Rees  C M, Hall N J, Eaton S, Pierro A: Surgical strategies for  necrotizing enterocolitis: A survey of practice  in the United  Kingdom. Arch Dis Child Fetal