Peritoneal drainage versus laparotomy as initial surgical treatment for perforated necrotizing enterocolitis in preterm and low birth weight infants
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To assess the benefits and risks of primary peritoneal drainage compared to laparotomy as the initial treatment for perforated necrotizing enterocolitis in preterm and LBW infants.
Sub‐group analysis will be planned for the following pre‐specified sub‐categories:
Gestational age (< 30 wks); birthweight (< 1000g).
Background
Necrotizing enterocolitis (NEC) is the most common gastrointestinal emergency in the newborn infant. NEC occurs predominantly in preterm infants, with an incidence of one to three cases per 1000 live births, affecting 1% to 5% of low birth weight infants (Holman 1989; Kliegman 1993). NEC is characterised by variable damage to the intestinal tract ranging from mucosal injury to full‐thickness necrosis and perforation. The areas most commonly affected are the terminal ileum and proximal ascending colon. The pathogenesis of necrotizing enterocolitis (NEC) is multifactorial. Any unifying hypothesis regarding pathogenesis of NEC remains unproven (Kanto 1985; Uauy 1991). Despite several decades of research, the etiology and pathogenesis of NEC remains elusive.
Mortality rates vary across centers and range from 10‐40%, depending on gestational age of the infant (Schullinger 1981; Snyder 1997). Those who survive may have significant complications both in short and long term outcomes including neurodevelopmental impairment (Tobiansky 1995; Sonntag 2000). The most common complication after NEC is intestinal stricture. Other complications include malabsorption, cholestasis and recurrent NEC. Although the incidence of NEC appears to be decreasing, NEC still remains a disease of high mortality and morbidity (Luig 2005; Luig 2005A; Tudehope 2005; Blakely 2005).
Bell staging criteria with the Walsh and Kliegman modification allow uniformity of diagnosis and treatment based on severity of illness. Stage I is defined as suspected NEC, manifesting as feeding intolerance, abdominal distention, and occult or frank blood in the stool; Stage II is definite NEC, presenting with marked abdominal distention, blood in the stool, and pneumatosis intestinalis on radiograph; Stage III is advanced NEC, with impending or proven perforation in addition to pneumatosis intestinalis in a critically ill baby (Walsh 1986).
The mainstay of treatment for patients with stage I or II NEC (Bell 1978; Walsh 1986) is nonoperative management. Medical management includes stopping enteral feedings, performing nasogastric decompression, and initiation of broad spectrum antibiotics. Operative intervention is directed largely at the complications of NEC, particularly intestinal perforation (Kleinhaus 1992). Up to 50% of neonates with NEC develop advanced disease requiring surgical intervention. Current surgical options include primary peritoneal drainage (PPD) or laparotomy. The principal objectives of laparotomy in acute NEC are to control sepsis, remove gangrenous bowel and to preserve as much of intestine as possible. Options include resection of necrotic areas of bowel with enterostomy, resection with primary anastomosis, proximal jejunostomy, 'clip and drop' technique and 'patch, drain and wait' (Pierro 2005).
Peritoneal drainage was introduced in the late 1970s for the management of NEC in extremely low birth weight infants prior to laparotomy in an effort to stabilise and improve the overall status of these critically ill infants who cannot tolerate laparotomy (Ein 1977). The unique features of wound healing described in experimental models may play a role in the healing of gastrointestinal ischemia and perforation in the setting of perforated NEC. Fetal wound healing without inflammation and scarring is well documented for humans and appears to be related to intrinsic differences in fetal cells and the extracellular matrix (Longaker 1989). This type of wound healing begins to diminish at the start of third trimester and does not occur once gestational maturity is attained. These findings would suggest that if the bowel were to heal on its own, it would be less likely to produce scarring and stricture in immature infants. The ability of fetal peritoneal cavity to resorb infarcted bowel is well demonstrated by intestinal atresia (Longaker 1991; Adzick 1992; Roh 2001). Peritoneal drainage can be performed at the bedside under local anaesthesia. Salvage laparotomy is reserved for those who deteriorate after drainage. Over recent years, there is a trend towards peritoneal drainage as the primary surgical procedure (Cheu 1988; Takamatsu 1992 ; Snyder 1997; Demestre 2002; Lee 2003; Rees 2005), especially in very low birth weight infants (Morgan 1994; Dimmitt 2000). In a prospective study of peritoneal drainage as a primary procedure, 86% of the treated infants improved after peritoneal drainage, 64% survived without the need for subsequent laparotomy, and 54% needed delayed surgery after a mean of 11 days (range of 1 ‐ 29 days [Demestre 2002]). In another large prospective cohort study involving 156 ELBW infants with NEC Stage III or intestinal perforation, 80 infants underwent primary peritoneal drainage. There was no difference in the survival rates between the two groups (Blakely 2005). In a retrospective study of 13 infants with NEC, peritoneal drainage helped with initial stabilization of the infants, but the improvement was temporary and subsequent laparotomy was required (Romero 2005). A systematic review of data from observational studies concluded that, with the currently available data, it is not possible to determine whether primary peritoneal drainage or laparotomy is superior and recommended the need for randomised controlled trials (Moss 2001).
There is still no consensus on the optimal surgical management of premature infants with perforated NEC. Surgical practices for NEC vary widely (Rees 2005; Pierro 2003; Pierro 2005) and there is a need to determine the best surgical treatment for perforated NEC.
Objectives
To assess the benefits and risks of primary peritoneal drainage compared to laparotomy as the initial treatment for perforated necrotizing enterocolitis in preterm and LBW infants.
Sub‐group analysis will be planned for the following pre‐specified sub‐categories:
Gestational age (< 30 wks); birthweight (< 1000g).
Methods
Criteria for considering studies for this review
Types of studies
All randomised or quasi‐randomised controlled trials where peritoneal drainage or laparotomy was used as the initial treatment of necrotizing enterocolitis in preterm (< 37 weeks gestation) and LBW (< 2500 g) neonates will be included. Unpublished studies and studies published only as abstracts will be included only if assessment of study quality is possible and if other criteria for inclusion are fulfilled.
Types of participants
Preterm (< 37 weeks gestation) and/or low birth‐weight (< 2500 g) neonates with perforated NEC (defined as presence of clinical signs and radiological evidence consistent with diagnosis of NEC according to the Walsh and Kleigman modification of Bell staging).
Types of interventions
Peritoneal drainage or laparotomy as initial surgical treatment for perforated NEC. Infants allocated to peritoneal drainage may go on to require later laparotomy, and will be included and analyzed in the peritoneal drainage arm.
Types of outcome measures
Primary outcome measures:
1. Mortality at 28 days, within 90 days of surgery, before discharge and, at 1 year corrected age
Secondary outcomes:
Need for chronic parenteral nutrition for greater than three months after initial surgery
Recurrent NEC
Need for laparotomy within seven days of primary surgical intervention
Need for laparotomy >= 7 days of primary surgical intervention
Abdominal wall abscess
Intestinal fistula: intestinal fistulas occurring between two loops of intestine (entero‐enteric fistula) or intestine and skin (entero cutaneous fistula) diagnosed by contrast imaging or operative exploration
Intestinal stricture: a narrow area of intestine that may lead to bowel obstruction diagnosed by contrast imaging or operative exploration
Bladder fistula: abnormal communication between bladder and the rectum or vagina; communication between bladder and the skin diagnosed by contrast imaging or operative exploration
Length of stay in hospital (days)
Hospital re‐admissions within the first year of life
Days receiving parenteral nutrition after surgery (days)
Neurodevelopmental outcome at 18 ‐ 26 months corrected age
Search methods for identification of studies
The standard search strategy of the Cochrane Neonatal Review Group will be used. This includes electronic searches of the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 2, 2006), Oxford Database of Perinatal Trials, MEDLINE (1966 ‐ March 2006), EMBASE (1980 ‐ June 2006) and CINAHL (1982 ‐ June 2006) and previous reviews including cross‐references. Abstracts of Paediatric Academic Societies meetings published in Pediatric Research (1995 ‐ 2006) will be hand searched. MEDLINE and EMBASE will be searched for relevant articles using the following MeSH terms or text words: ( primary peritoneal drainage OR peritoneal drainage OR laparotomy OR abdominal paracentesis OR abdominal drainage) AND (necrotising enterocolitis OR necrotizing enterocolitis OR NEC OR perforated NEC OR perforated necrotising enterocolitis AND (infant, newborn/OR infant, low birth weight/OR infant, very low birth weight/OR infant, premature/OR Infant, Premature, Diseases) OR (neonate: OR prematur*: OR newborn) AND (clinical trial OR Randomised Controlled Trials). Reference lists of published narrative and systematic reviews will be reviewed. No language restrictions will be applied. The authors of all studies will be contacted to clarify reported data or provide additional data and information including details of methodology.
Data collection and analysis
Included and excluded studies
The review will be undertaken by two review authors (SR and LB). SR and LB will screen the title and abstracts of all studies identified by the above search strategy and obtain the full articles for all potentially relevant trials. SR and LB will re‐assess independently the full text of these reports using an eligibility form based on the prespecified inclusion criteria. SR and LB will exclude those studies that do not meet all of the inclusion criteria and will state the reasons for exclusion. SR will request additional information from the trial authors to clarify methodology and to obtain relevant additional information. Where it will not be possible to evaluate the study because of language problems or missing information, SR and LB will classify the study as a 'study awaiting assessment' until a translation or further information can be obtained. SR and LB independently will use a data collection form to aid extraction of relevant information and data from each included study. SR and LB will extract data separately, compare data, and resolve differences by discussion until consensus is achieved.
Study quality
SR and LB will assess the studies independently without blinding to authorship or journal using the checklist developed for the Cochrane Neonatal Review Group. Discrepancies will be resolved by discussion. The trials would be assessed for blinding of randomization, blinding of intervention, completeness of follow‐up and blinding of outcome assessment.
Quality checklist
Allocation concealment
A. Adequate ‐ Randomisation method described that would not allow investigator/participant to know or influence intervention group before eligible participant entered in the study
B. Unclear ‐ Randomisation stated but no information on method used is available
C. Inadequate ‐ Method of randomisation used such as alternate medical record numbers or unsealed envelopes; any information in the study that indicated that investigators or participants could influence intervention group
Blinding
Blinding of investigators: Yes/No/can't tell
Blinding of participants: Yes/No/can't tell
Blinding of outcome assessor: Yes/No/can't tell
Blinding of data analysis: Yes/No/can't tell
The above are considered not blinded if the treatment group can be identified in > 20% of participants because of the side effects of treatment.
Intention‐to‐treat
Yes ‐ Specifically reported by authors that intention‐to‐treat analysis was undertaken and this was confirmed on study assessment
Yes ‐ Not stated but confirmed on study assessment
No ‐ Not reported and lack of intention‐to‐treat analysis confirmed on study assessment. (Patients who were randomised were not included in the analysis because they did not receive the study intervention, they withdrew from the study or were not included because of protocol violation)
No ‐ Stated but not confirmed upon study assessment
Not stated
We would contact the investigators for the data and analyse according to intention to treat.
Completeness of follow‐up
Percent of participants excluded or lost to follow‐up.
Statistical assessment
Statistical analyses will be done according to the recommendations of the Cochrane Neonatal Review Group. We will present outcomes for categorical data as relative risk, risk difference, and number needed to treat, with respective 95% confidence intervals. For continuous data, we will use the weighted mean difference with 95% confidence intervals.
We will estimate the treatment effects of individual trials and examine heterogeneity between trial results by inspecting the forest plots and quantifying the impact of heterogeneity in any meta‐analysis using a measure of the degree of inconsistency in the studies' results (I2). If we detect statistical heterogeneity, we will explore the possible causes. We plan to use a fixed‐effect model for meta‐analyses. We will examine publication bias using a funnel plot, and a regression approach to assess funnel plot asymmetry (Egger 1997).
