Journal of Perinatology (2008) 28, S14–S17 r 2008 Nature Publishing Group All rights reserved. 0743-8346/08 $30 www.nature.com/jp ORIGINAL ARTICLE Evidence vs experience in the surgical management of necrotizing enterocolitis and focal intestinal perforation

CJ Hunter1,2, N Chokshi1,2 and HR Ford1,2 1Department of Surgery, University of Southern California Keck School of Medicine, Los Angeles, CA, USA and 2Department of Surgery, Childrens Hospital Los Angeles, Los Angeles, CA, USA

bacterial colonization and prematurity.4 There is a subset of low Introduction: Necrotizing enterocolitis (NEC) and focal intestinal birth weight infants, however, that sustain focal intestinal perforation (FIP) are neonatal intestinal emergencies that affect premature perforation (FIP) without classic clinical, radiographic, or infants. Although most cases of early NEC can be successfully managed with histological evidence of NEC.5 FIP appears to be a distinct clinical medical therapy, prompt surgical intervention is often required for advanced entity that occurs in 3% of very low birth weight (VLBW) infants or perforated NEC and FIP. and accounts for 44% of gastrointestinal perforations in this 6 Method: The surgical management and treatment of FIP and NEC are population. Optimal surgical management of severe NEC and discussed on the basis of literature review and our personal experience. FIP has been the subject of ongoing controversy for many years.

Result: Surgical options are diverse, and include peritoneal drainage, laparotomy with diverting ostomy alone, laparotomy with intestinal Presentation of NEC and FIP resection and primary anastomosis or stoma creation, with or without Infants with NEC typically present with feeding intolerance and second-look procedures. bloody stools in the second or third week of life. In the more Conclusion: The optimal surgical therapy for FIP and NEC begins with advanced cases, they often demonstrate signs of sepsis and prompt diagnosis and adequate fluid resuscitation. It appears that there is cardiovascular collapse.7 The severity of NEC is classified according no significant difference in patient outcome based on surgical to the staging criteria established by Bell et al.8 The mortality of management alone. However, the infant’s weight, comorbidities, surgeon patients with stage III NEC approaches 40 to 100%. Radiographic preference and timing of intervention should be taken into account before evidence of NEC is characterized by the pathognomonic finding of operative intervention. pneumatosis intestinalis. Other findings include fixed intestinal Journal of Perinatology (2008) 28, S14–S17; doi:10.1038/jp.2008.44 loops, gas within the portal system or pneumoperitoneum.9 Patients with FIP usually present around the first week of life and Keywords: necrotizing enterocolitis; surgery; intestinal perforation do not exhibit the prodromal clinical symptoms or radiographic findings associated with NEC.5 However, patients with FIP may have radiographic evidence of pneumoperitoneum. Etiologic factors Introduction implicated in the pathogenesis of FIP include the use of umbilical Necrotizing enterocolitis (NEC) is the most common catheters, administration of indomethacin or steroid and 1 gastrointestinal emergency that affects newborn infants. A total of congenital defects of the intestinal wall.10,11 In general, the 90% of NEC cases occur in premature infants, with only 10% outcome for patients with FIP is more favorable than for those with occurring in full-term infants; most of whom also have NEC. However, despite appropriate and timely medical comorbidities predisposing them to decreased mesenteric management, over one-third of patients with NEC, and virtually all 2 perfusion. Despite major advances in neonatal medicine that have patients with FIP will require surgical intervention.12,13 resulted in improved outcomes over the past 20 years, the morbidity and mortality associated with NEC remain largely unchanged, and the incidence of NEC continues to increase.3 Although the precise Surgical options in patients >1500 g pathophysiology of NEC is poorly defined, three key contributing In infants with a birth weight >1500 g, laparotomy and resection factors have been identified; these include: early enteral feeding, of the necrotic intestine is generally the preferred approach. In some cases necrosis is limited to an isolated segment of intestine; Correspondence: Dr HR Ford, Childrens Hospital Los Angeles, 4650 Sunset Boulevard, Mail Stop no. 35, Los Angeles, CA 90027, USA. while in others, the disease pattern is more diffuse, involving E-mail: [email protected] multiple segments of small and/or large bowel, with intervening Optimal surgical management of NEC and FIP CJ Hunter et al S15

strategy often resulted in the killing of potentially viable intestine. Hence, some surgeons advocated the ‘patch, drain and wait’ approach.18 In this technique each perforation is debrided and closed, penrose drains are placed and parenteral nutritional support is instituted. In cases of pan-necrosis, outcome is often extremely poor despite aggressive surgical intervention. Our recommendation for diffuse disease is initial laparotomy with proximal diversion alone. If the patient is highly unstable, peritoneal drainage (PD) may be utilized while resuscitation is ongoing. PD should be followed by delayed laparotomy. Laparotomy has the added benefit of allowing confirmation of the severity and extent of intestinal involvement. Focal intestinal perforation is unusual in patients weighing >1500 g as they more commonly develop NEC. Due to the of the limited anatomical area involved, some surgeons advocate primary anastomosis in this setting, or in the setting of focal segmental necrosis, as mentioned earlier.14,19 These small retrospective studies Figure 1 Segments of intestine affected by necrotizing enterocolitis (NEC). The argue that primary anastomosis is safe and effective. However, a classic findings of diffuse NEC are seen in this image. Patchy areas of necrosis are larger retrospective study by Cooper et al.20 reported a 24% lower seen with questionable intestinal viability. survival rate for patients who underwent primary anastomosis compared to ostomy creation. On the basis of the current available areas of questionable viability (Figure 1). The goal is to limit the data, an expedient laparotomy with resection of the diseased extent of bowel resection to avoid resultant short gut syndrome. intestinal segment and creation of a diverting ostomy may yield The standard of care is to remove all necrotic areas, taking great the best outcome in this subset. care to preserve any bowel that appears viable.8 After resection of the necrotic segments, an area of viable intestine is used to create an ostomy with or without a mucus fistula. There are occasional Surgical options in patients <1500 g: to drain or not to reports of primary anastomosis following resection of an isolated drain? necrotic intestinal segment; however, this is not a widely accepted In patients weighing <1500 g, the optimal choice of surgical approach.14 In patients with multiple areas of questionable intervention has been more controversial. Some surgeons argue intestinal viability, a ‘second-look’ procedure may be planned after that laparotomy with intestinal resection in VLBW infants is 24 to 48 h to reevaluate the bowel. Weber and Lewis15 reported on associated with high rates of morbidity and mortality. Therefore, 32 patients with pan-intestinal involvement and investigated the they advocate PD as the preferred initial, and sometimes definitive, impact of a second-look operation on outcome. The survival rate procedure for patients with complicated NEC or FIP.21–23 However, was 71% (10/14) for patients who underwent a second-look it is still unclear whether PD is appropriate for both NEC and FIP, procedure, and 67% (12/18) for those who did not. While this since many infants with NEC eventually require exploratory difference was not statistically significant, the principal benefit of laparotomy. In 1977, Ein et al.24 reported on a series of the second-look procedure was probably to spare marginally viable hemodynamically unstable extremely low birth weight patients who intestine. Second-look procedures have been recommended in underwent PD for NEC. Three infants survived; two of them did not concert with the ‘clip and drop back technique’.16 Proponents of require a subsequent operation. They postulated that PD allowed this approach advocate resection of the entire diseased bowel at the relief of intra-abdominal pressure, drainage of intestinal initial exploration without anastomosis, followed by reexploration perforation and time for medical optimization and stabilization. at 48 to 72 h. However, the patient sample size (five patients) is too Over the ensuing two decades, this approach has steadily gained small to comment on efficacy. Proximal diversion alone has been popularity, and many surgeons now employ it both as a shown to limit bowel resection without increasing morbidity or temporizing measure and as definitive therapy for extremely low mortality.17 In addition, second-look procedures may be of benefit birth weight patients with Bell stage III disease. In fact, Rovin in infants who are not improving 24 to 48 h after initial diversion. et al.25 advocate primary PD for all infants with stage III NEC; However, there are no randomized prospective studies comparing however, several reports note that the majority of NEC survivors this approach to others. who undergo PD eventually require laparotomy, regardless of birth In the past, when multiple segments of necrotic intestine were weight.26 Consistent with these observations, a prospective found at laparotomy for NEC, multiple stomas were created. This European multi-institutional study of PD in 44 infants with NEC27

Journal of Perinatology Optimal surgical management of NEC and FIP CJ Hunter et al S16 reported that 86% of the neonates improved after PD; however, 54% Neurodevelopmental impairment affects 50% of NEC survivors. of those who underwent PD required delayed laparotomy. The In a recent nonrandomized study, Blakely et al.32 reported that overall survival in this cohort was 82%, with 57% survival in patients with NEC who underwent laparotomy instead of PD had a infants with a birth weight <1000 g, and 95% for those weighing lower risk of future developmental delay.32 Because prior >1000 g. Of note, the authors did not distinguish between patients investigations suggested that neurodevelopmental delay was related with NEC and FIP. mainly to prematurity and prolonged hospital stay, Blakely et al. In 2001, Moss et al.28 published a meta-analysis of 10 studies controlled for the confounding factors of prematurity and severity involving a total of 475 patients with NEC. They concluded that it of disease in their study. Further investigation is warranted to was not possible to determine whether PD was superior to address this issue. laparotomy; therefore, they recommended a randomized control trial to resolve this issue. More recently, Moss and colleagues Conclusions published a multi-institutional, randomized, controlled trial Neonatal intestinal perforation is characterized by a spectrum of comparing primary PD to laparotomy and bowel resection in clinical presentations and anatomical findings. The inciting causes infants with NEC weighing <1500 g.29 They found no statistically of both FIP and NEC are typically multifactorial, affecting the significant difference in mortality, dependence on parenteral smallest and most fragile patients. Fundamental to the treatment of nutrition or length of hospital stay between groups; 38% of the PD both NEC and FIP is prompt diagnosis, adequate resuscitation and required delayed laparotomy. Interestingly, among patients who a combined team approach, by both pediatric surgeons and were evaluated but not enrolled in the study, the mortality rate was neonatologists. Current evidence suggests that the precise form of 15% for those who underwent laparotomy compared to 41% for operative intervention does not affect outcome; however, there are those who underwent PD. Thus, while PD may have an important many confounding variables including patient’s weight, role in the surgical management of extremely low birth weight comorbidities, surgeon preference, experience and timing of the infants, careful individual assessment of each patient by an procedure. The optimal surgical approach to severe NEC is indeed experienced pediatric surgeon is required to determine optimal controversial. While PD may offer some initial advantage because it treatment strategy. is relatively easy to perform and less invasive, it is better reserved for unstable infants as part of the resuscitation phase prior to Complications definitive laparotomy. Specifically, pan-involvement is best managed with laparotomy and proximal diversion alone. The management of NEC is not without complications.30 The most Additionally, second-look procedures may be useful in infants who serious complications of acute NEC include intestinal necrosis and are not improving. Although FIP may be managed with PD, we perforation, which may occur in up to one-third of patients. again advocate the use of drainage as a temporizing measure until However, some patients who initially appear to respond well to definitive laparotomy can be performed. medical management develop signs of intestinal obstruction upon The management of neonatal intestinal perforation (NEC and resuming enteral feedings. This scenario may be due to the FIP) is an area of active clinical and scientific research. As the development of ischemic strictures in the small or large intestine. understanding of the pathophysiologic underpinnings of these Intestinal strictures develop in up to one-third of patients with a diseases progress, it is believed that successful treatment modalities history of NEC.31 The most commonly affected areas include the will continue to emerge. Long-term clinical outcome studies are terminal ileum, splenic flexure and the junction of the descending needed to elucidate the impact of specific therapeutic strategies and sigmoid colon. Radiographic imaging may confirm bowel in this potentially devastating disease. obstruction, with a transition zone and air–fluid levels. If a stricture is suspected a contrast enema (or an upper gastrointestinal study) should be performed to assess intestinal Disclosure patency. If a stricture is demonstrated, surgical resection is This paper was supported by a grant from the NIH AI-49473 (HRF indicated at this time. and VC) and by the Merck Research Fellowship from the Surgical In addition to intestinal strictures, almost 10% of patients with a Infection Society (CH). history of NEC and surgical intervention develop short gut 30 syndrome. These patients depend on long-term intravenous References nutrition and are exposed to its adverse sequelae, including line sepsis, growth retardation, parenteral nutrition associated 1 Stoll BJ. Epidemiology of necrotizing enterocolitis. Clin Perinatol 1994; 21: 205–218. cholestasis, cirrhosis and liver failure. Many of these infants 2 Maayan-Metzger A, Itzchak A, Mazkereth R, Kuint J. Necrotizing enterocolitis in full- eventually require small bowel transplantation and, occasionally, term infants: case-control study and review of the literature. J Perinatol 2004; 24: combined liver and small bowel transplantation. 494–499.

Journal of Perinatology Optimal surgical management of NEC and FIP CJ Hunter et al S17

3 Henry MC, Moss RL. Current issues in the management of necrotizing enterocolitis. 18 Moore TC. Successful use of the ‘patch, drain, and wait’ laparotomy approach to Semin Perinatol 2004; 28: 221–233. perforated necrotizing enterocolitis: is hypoxia-triggered ‘good angiogenesis’ involved? 4 Hsueh W, Caplan MS, Qu XW, Tan XD, De Plaen IG, Gonzalez-Crussi F. Neonatal Pediatr Surg Int 2000; 16: 356–363. necrotizing enterocolitis: clinical considerations and pathogenetic concepts. Pediatr 19 Parigi GB, Bragheri R, Minniti S, Verga G. Surgical treatment of necrotizing Dev Pathol 2003; 6: 6–23. enterocolitis: when? how? Acta Paediatr 1994; 396(Suppl): 58–61. 5 Okuyama H, Kubota A, Oue T, Kuroda S, Ikegami R, Kamiyama M. A comparison of the 20 Cooper A, Ross AJ, O’Neill JA, Schnaufer L. Resection with primary anastomosis for clinical presentation and outcome of focal intestinal perforation and necrotizing necrotizing enterocolitis: a contrasting view. J Pediatr Surg 1988; 23: 64–68. enterocolitis in very-low-birth-weight neonates. Pediatr Surg Int 2002; 18: 704–706. 21 Cass DL, Brandt ML, Patel DL, Nuchtern JG, Minifee PK, Wesson DE. Peritoneal 6 Mintz AC, Applebaum H. Focal gastrointestinal perforations not associated with drainage as definitive treatment for neonates with isolated intestinal perforation. necrotizing enterocolitis in very low birth weight neonates. J Pediatr Surg 1993; 28: J Pediatr Surg 2000; 35: 1531–1536. 857–860. 22 Morgan LJ, Shochat SJ, Hartman GE. Peritoneal drainage as primary management 7 Morecroft JA, Spitz L, Hamilton PA, Holmes SJ. Necrotizing enterocolitisFmultisystem of perforated NEC in the very low birth weight infant. J Pediatr Surg 1994; 29: organ failure of the newborn? Acta Paediatr 1994; 396(Suppl): 21–23. 310–314; discussion 314–315. 8 Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, Barton L et al. Neonatal 23 Lessin MS, Luks FI, Wesselhoeft CW, Gilchrist BF, Iannitti D, DeLuca FG. Peritoneal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg drainage as definitive treatment for intestinal perforation in infants with extremely low 1978; 187:1–7. birth weight (<750 g). J Pediatr Surg 1998; 33: 370–372. 9 Kurscheid T, Holschneider AM. Necrotizing enterocolitis (NEC)Fmortality and 24 Ein SH, Marshall DG, Girvan D. Peritoneal drainage under local anesthesia for long-term results. Eur J Pediatr Surg 1993; 3: 139–143. perforations from necrotizing enterocolitis. J Pediatr Surg 1977; 12: 963–967. 10 Tatekawa Y, Muraji T, Imai Y, Nishijima E, Tsugawa C. The mechanism of focal 25 Rovin JD, Rodgers BM, Burns RC, McGahren ED. The role of peritoneal drainage intestinal perforations in neonates with low birth weight. Pediatr Surg Int 1999; 15: for intestinal perforation in infants with and without necrotizing enterocolitis. 549–552. J Pediatr Surg 1999; 34: 143–147. 11 Stark AR, Carlo WA, Tyson JE, Papile LA, Wright LL, Shankaran S et al. Adverse effects 26 Ahmed T, Ein S, Moore A. The role of peritoneal drains in treatment of perforated of early dexamethasone in extremely-low-birth-weight infants. National Institute of necrotizing enterocolitis: recommendations from recent experience. J Pediatr Surg Child Health and Human Development Neonatal Research Network. N Engl J Med 1998; 33: 1468–1470. 2001; 344: 95–101. 27 Demestre X, Ginovart G, Figueras-Aloy J, Porta R, Krauel X, Garcia-Alix A et al. 12 Tam AL, Camberos A, Applebaum H. Surgical decision making in necrotizing Peritoneal drainage as primary management in necrotizing enterocolitis: a prospective enterocolitis and focal intestinal perforation: predictive value of radiologic findings. study. J Pediatr Surg 2002; 37: 1534–1539. J Pediatr Surg 2002; 37: 1688–1691. 28 Moss RL, Dimmitt RA, Henry MC, Geraghty N, Efron B. A meta-analysis of peritoneal 13 Blakely ML, Lally KP, McDonald S, Brown RL, Barnhart DC, Ricketts RR et al. drainage versus laparotomy for perforated necrotizing enterocolitis. J Pediatr Surg Postoperative outcomes of extremely low birth-weight infants with necrotizing enterocolitis 2001; 36: 1210–1213. or isolated intestinal perforation: a prospective cohort study by the NICHD Neonatal 29 Moss RL, Dimmitt RA, Barnhart DC, Sylvester KG, Brown RL, Powell DM et al. Research Network. Ann Surg 2005; 241: 984–989; discussion 989–994. Laparotomy versus peritoneal drainage for necrotizing enterocolitis and perforation. 14 Ade-Ajayi N, Kiely E, Drake D, Wheeler R, Spitz L. Resection and primary anastomosis N Engl J Med 2006; 354: 2225–2234. in necrotizing enterocolitis. J R Soc Med 1996; 89: 385–388. 30 Horwitz JR, Lally KP, Cheu HW, Vazquez WD, Grosfeld JL, Ziegler MM. Complications 15 Weber TR, Lewis JE. The role of second-look laparotomy in necrotizing enterocolitis. after surgical intervention for necrotizing enterocolitis: a multicenter review. J Pediatr J Pediatr Surg 1986; 21: 323–325. Surg 1995; 30: 994–998; discussion 998–999. 16 Vaughan WG, Grosfeld JL, West K, Scherer LR, Villamizar E, Rescorla FJ. Avoidance of 31 Janik JS, Ein SH, Mancer K. Intestinal stricture after necrotizing enterocolitis. J Pediatr stomas and delayed anastomosis for bowel necrosis: the ‘clip and drop-back’ technique. Surg 1981; 16: 438–443. J Pediatr Surg 1996; 31: 542–545. 32 Blakely ML, Tyson JE, Lally KP, McDonald S, Stoll BJ, Stevenson DK et al. Laparotomy 17 Luzzatto C, Previtera C, Boscolo R, Katende M, Orzali A, Guglielmi M. Necrotizing versus peritoneal drainage for necrotizing enterocolitis or isolated intestinal perforation enterocolitis: late surgical results after enterostomy without resection. Eur J Pediatr in extremely low birth weight infants: outcomes through 18 months adjusted age. Surg 1996; 6: 92–94. Pediatrics 2006; 117: e680–e687.

Journal of Perinatology