Pathophysiology 21 (2014) 105–110
Current concepts in the surgical approach to necrotizing enterocolitis Mehul V. Raval, R. Lawrence Moss ∗ Division of Pediatric Surgery, Nationwide Children’s Hospital, The Ohio State University College of Medicine, Columbus, OH, USA
Abstract Necrotizing enterocolitis (NEC) is the most common surgical emergency occurring in neonatal intensive care unit (NICU) patients. Among patients with NEC, those that require surgery experience the poorest outcomes and highest mortality. Surgical intervention, while attempting to address the intestinal injury and ongoing mulitfactorial physiologic insults in NEC is associated with its own stresses that may compound the ongoing physiologic derangement. Surgery is thus reserved for those patients with clear indication for intervention such as pneumoperitoneum, confirmed stool or pus in the peritoneal cavity, or worsening clinical status. The purpose of this review is to briefly describe the physiologic stress induced by surgical intervention in the preterm, low birth weight patient with NEC and to provide a contemporary overview of available surgical management options for NEC. The optimal surgical plan employed is strongly influenced by clinical judgment and theoretical benefits in terms of minimizing physiologic stressors while providing temporary and/or definitive treatment in a timely fashion. While the choice of operation has not been shown to have a significant effect on any clinically important outcomes, ongoing investigations continue to study both short and long-term outcomes in patients with NEC. © 2013 Elsevier Ireland Ltd. All rights reserved. Keywords: Necrotizing enterocolitis; Surgery; Neonatal surgery; Neonatal intensive care unit; Prematurity
1. Introduction Necrotizing enterocolitis (NEC) remains one of the most challenging diseases to diagnose and treat among neonatal intensive care unit (NICU) patients. NEC is the most common surgical emergency and the most common cause of gastrointestinal-related morbidity in the NICU [1,2]. Patients with NEC who undergo surgical interventions have the worst outcomes and highest costs [3]. Though the definitive pathophysiology of NEC has not been fully elucidated, NEC is predominantly a disease of low birth weight, premature infants, which is only occasionally seen in term infants [2]. The changes in intestinal microbiota, loss of the gut barrier, microciculatory changes, and other pathophysiologic etiologies observed in NEC are more pronounced in the premature newborn with naïve physiology, limited reserve, heightened fluid and electrolyte balance sensitivity, increased energy ∗ Corresponding author at: Nationwide Children’s Hospital, E. Thomas Boles Jr., Professor of Surgery, The Ohio State University, College of Medicine, 700 Children’s Drive, Columbus, OH 43205, USA. Tel.: +1 614 722 5163; fax: +1 614 722 5995. E-mail address: [email protected] (R.L. Moss).
0928-4680/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.pathophys.2013.11.017
requirements, and altered immunologic and stress response. As overall NICU care has advanced and the survival of preterm infants has improved, a corresponding increase in prevalence of NEC has been observed. Surgical intervention, while attempting to address the intestinal injury and ongoing mulitfactorial physiologic insult, is associated with a variety of inherent stresses that may compound the physiologic derangement observed in NEC. While the pathophysiologic basis of NEC is extensively covered in other chapters in this compendium, the purpose of this review is to briefly describe the physiologic stress induced by surgical intervention. Specific topics that may be of significance in the low birth weight, premature infant undergoing surgery will be addressed. Subsequently, the various surgical options available in the management of NEC will be explored. The surgical option employed is strongly influenced by clinical judgment and theoretical benefits in terms of minimizing physiologic stressors while providing temporary and/or definitive treatment in a timely fashion. While the choice of operation has not been shown to have a significant effect on any clinically important outcomes, ongoing investigations continue to study both short and long-term outcomes in patients with NEC.
106
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
2. Presentation and diagnosis A variety of factors have been proposed to be associated with the development of NEC including prematurity, enteral feedings, steroid use, hypotension/ischemia, presence of a patent ductus arteriosus, and infection [2,4,5]. Among these, only prematurity has been consistently associated with NEC and need for surgical intervention [6,7]. Early recognition of NEC development is a clinical challenge. Early NEC can be difficult to distinguish from relatively benign conditions such as feeding intolerance commonly associated with prematurity to severe conditions such as neonatal sepsis. Subtle initial signs such as feeding intolerance and abdominal distention are often followed by temperature instability, lethargy, apnea, and bradycardia. Severe or “surgical” disease often includes the development of abdominal tenderness, peritonitis, abdominal wall discoloration, and identification of palpable abdominal mass. Though fixed abdominal mass and abdominal wall erythema are highly specific signs of intestinal necrosis, less than 10% of patients with surgical NEC will have these findings [8]. At the other end of the clinical spectrum, NEC can progress to multi system organ failure or death within a few hours of onset. Though most commonly occurring in the terminal ileum, NEC may involve any and/or non-contiguous portions of intestine. When the entire gut is ischemic, the clinical picture is referred to as “NEC totalis”. Laboratory findings are often non-specific during the initial phases of NEC. Metabolic acidosis, neutropenia, left shift of segmented neurtrophils, and hyponatremia are the most concerning laboratory aberrations. Thrombocytopenia, especially with a rapid drop in the platelet count, is a poor prognostic sign [1]. A combination of these laboratory values has been used in scoring systems to predict the timing and type of surgical intervention in NEC [9]. Attempts to identify biochemical markers specific to necrotizing enterocolitis have not been successful and remain an area of ongoing research. A variety of imaging modalities are now available in the NICU population but abdominal plain films have remained the mainstay for NEC diagnosis for decades. Pneumoperitoneum identified on plain films mandates surgical intervention [10]. Pneumotosis intestinalis, the most specific finding for diagnosing NEC, as well as portal venous gas or fixed dilated air-filled loops can confirm the diagnosis but are not necessarily indications for surgical intervention (Fig. 1). Abdominal ultrasound is increasingly available in most NICU settings and can be used to assess intestinal peristalsis, bowel wall thickness, and to identify to fluid collections. Ultrasound technology has improved to include Doppler flow assessment as well as more detailed imaging ability with high frequency transducers. Ultrasound can be readily performed at the bedside, is noninvasive, and avoids ionizing radiation. Though several institutions with experienced technicians have shown high sensitivity and specificity of ultrasound to detect necrotic bowel in NEC [11,12], the generalizability of this imaging modality to consistently provide information beyond that
Fig. 1. Plain radiograph of extensive pneumotosis intestinalis and free air in a patient with necrotizing enterocolitis.
provided in plain films remains to be determined. Other imaging studies such as contrast studies, computed tomography scans, and magnetic resonance imaging are available but have not proven to be of significant utility in the acute management of NEC [13]. Ultimately, surgical intervention is reserved for those patients with pneumoperitoneum, confirmed stool or pus in the peritoneal cavity, or worsening clinical status.
3. The pysiologic stress of surgery Once the decision to proceed with surgery has been made it is imperative to anticipate the physiologic impact of the ensuing stress response of the surgical intervention itself. Operative intervention is often described as a controlled, induced trauma. Numerous changes occur in response to the surgical intervention from increased metabolic demands, to triggered inflammatory and immune cascades, and to alterations in the endocrine balance [14]. All of these changes are magnified in the preterm newborn [15]. After operative trauma, metabolism increases to provide substrates needed for regenerating and healing tissue. In newborns undergoing major abdominal surgery, catecholamine release, oxygen consumption, and resting energy expenditure peak at 4 h postoperatively and then return rapidly to baseline within the first 24 h of surgery [16,17]. Moderate intraoperative opiate dosing has been shown to blunt the immediate stress response and may be associated with an improved postoperative
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
outcomes [17]. The surge of opiate release around the time of umbilical cord clamping may partially explain the reduced metabolic stress response seen in newborns undergoing surgical procedures in the first few hours of birth [18]. After surgery, the immune system is both stimulated to combat opportunistic postoperative infection while at the same time dampened to facilitate healing. Key cytokines released include interleukins (IL) such as IL-1 and IL-6 as well as tumor necrosis factor ␣. These pro-inflammatory cytokines are balanced by anti-inflammatory IL1ra and IL-10. These cytokines and others trigger specific gene expression in target cells, trigger release of regulatory hormones, and facilitate cell-to-cell communication. These actions have a direct impact on third spacing of fluids, metabolic demands, and culminate in wound healing. Peak levels of specific cytokines, such as IL-6, may directly correlate with the degree of operative trauma in the neonate [19]. Other mediators of cellular response altered by surgical trauma include histamine, serotonin, cortisol, and various components of the kinin and complement systems. The endocrine response to surgery has also been well studied with changes in the insulin/glucose ratio in the immediate postoperative period. In neonates, hyperglycemia is often observed after surgical intervention and the degree of hyperglycemia may be directly correlated to the degree of surgical stress [20]. Ultimately, the balance of pro and anti-inflammatory response, increased metabolic needs, fluid-shifts, and overall homeostasis must be closely observed in the postsurgical neonate.
4. The role of peritoneal drainage The two main options available for the surgical management of NEC are exploratory laparotomy and primary peritoneal drainage (PPD). Though laparotomy with bowel resection has been the classic surgical approach, PPD has been considered an option for over four decades (Fig. 2) [21]. The role of PPD has changed from initial use as a temporizing measure to more broad use as a potential definitive treatment for NEC [21–23]. Ein et al. published their experience with neonates weighing less than 1000 grams demonstrating a 32% success of PPD alone without the need of subsequent laparotomy [23,24]. While several groups reported similar success [25,26], others reports showed conflicting results [24,27,28]. Most of these studies included small numbers of patients, few were prospective, and none were randomized. A 2001 meta-analysis attempted to adjust for the fact that PPD was often employed in the most premature and ill cohorts, and a difference in outcome could not be assessed [27]. In 2006, a prospective, observational study was published comparing laparotomy to PPD for patients of 1 kg or less using data accrued at 16 centers with the National Institute of Child Health and Human Development (NICHD) Neonatal Research Network [29]. Regardless of surgical treatment, outcomes for the entire cohort were poor with 60% either having died or received prolonged parenteral nutrition
107
Fig. 2. Drawing of primary peritoneal drain placement in a neonate with necrotizing enterocolitis.
by discharge and 72% having died or shown to have had neurodevelopmental delay by the time of follow-up. When comparing the patients based on type of surgical intervention the PPD group had lower gestational age, postnatal age, systolic blood pressure at the time of intervention, and were more likely to be treated with high frequency oscillatory ventilation or have higher peak inspiratory pressures on conventional ventilation compared to those undergoing laparotomy. This study suggested a possible difference in long-term neurodevelopmental outcomes favoring laparotomy prior to disease severity adjustment and serves as the basis for the current NICHD Neonatal Research Network NEST trial [30]. Two randomized controlled trials have been performed comparing PPD to laparotomy. The North American trial was published in 2006 (NECSTEPS) and the European trial in 2008 (NET trial) [31,32]. Both trials were well conducted and represented multiple centers (15 centers in North America for NECSTEPS and 18 centers across 11 countries in the NET trial). These trials focused on low birth weight, premature infants with NECSTEPS outlining a 1500 g birth weight limit and a gestational age less than 34 weeks. The NET trial limited inclusion to those infants less than or equal to 1000 g. Key differences between the trials involved the technique used and patient management in the peritoneal drainage groups. In NECSTEPS, operating surgeons were encouraged to irrigate through the peritoneal drain incision prior to drain placement. In the NET trial, surgeons were discouraged from performing any irrigation to wash away stool or pus. Along the same lines, NECSTEPS patients were managed with peritoneal drainage as a definitive therapy. If patients worsened then additional drains were allowed. Only 5 patients in this trial were converted to laparotomy for early clinical deterioration. Conversely, the NET trial considered peritoneal
108
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
drainage a temporizing treatment with 74% of patients in the NET trial assigned to peritoneal drainage ultimately undergoing laparotomy at a median of 2.5 days after randomization. Peritoneal drainage was utilized as effective definitive treatment in 11% of infants in the NET trial and nearly 33% in the NECSTEPS trial. The primary outcome for both trials was survival. This was measured at one and six months in the NET trial and at 90 days in NECSTEPS. Both studies identified survival rates that were not statistically different between peritoneal drainage and laparotomy. Length of stay was evaluated in both trials as a secondary outcome and both trials found no statistical difference in length of stay between laparotomy and PPD. Similarly, there were no statistical differences in various other outcomes of interest in one or both trials studied various time points including ventilator dependence, parenteral nutrition dependence, and time to full enteral nutrition. Neither study addressed neurodevelopmental outcomes or cost-effectiveness. The conclusions garnered from these studies differed based on how the studies approached use of peritoneal drains. In NECSTEPS, where PPD was considered definitive treatments, authors concluded that there was no significant difference between primary peritoneal drainage and laparotomy for initial management. Conversely, the NET trial, where peritoneal drains were considered a temporizing measure to laparotomy, authors concluded that primary peritoneal drainage is ineffective as either a temporizing measure or definitive treatment. A recent Cochrane Review combining the results of the two studies concluded that no significant benefits or harms of peritoneal drainage versus laparotomy could be identified [33].
5. Laparotomy and stoma creation Laparotomy with resection of necrotic bowel and with enterostomy creation has been the traditional approach to NEC [34]. Advantages include an expeditious intervention that addresses the physiologic insult. The disadvantages of this approach that must be considered include a need for parenteral nutrition and the risk of high stoma output and salt and water loss. Stoma specific complications include retraction, stricture, prolapse, obstruction, and skin/wound issues [35]. A second anesthetic is also required for stoma takedown typically when the patient’s weight is over 2 kg and at least 4 weeks after stomas maturation to avoid encountering vascular intra-abdominal adhesions [36].
6. Surgical alternatives Laparotomy with resection and primary anastomosis is another option that eliminates the mandatory need for a second operative intervention to reestablish intestinal continuity [37]. Much of the physiologic stress of stoma creation such as
fluid losses and electrolyte abnormalities can thus be avoided. Some groups have even endorsed performing numerous anastomoses in the setting of multiple areas of disease [38]. Disadvantages to performing even a single anastomoses in a NEC patient undergoing laparotomy include leak and stricture related to working with inflamed, edematous tissue in a hemodynamically unstable patient. Anastomotic complications have the potential to serve as spark that could easily lead the fragile neonate back into sepsis, multisystem organ dysfunction, and death. Primary anastomsis in the acutely ill NEC patient remains an area of ongoing investigation. Other surgical options include a ‘clip and drop’ technique where grossly necrotic bowel is resected and the proximal and distal aspects are closed using titanium clips [39]. These ends are placed back into the peritoneal cavity and a subsequent operation is then performed within 48–72 h to assess for possible bowel anastomosis in a more physiologic stable patient [40]. Another technique described involves laparotomy with transverse closure of areas of frank bowel wall injury (‘patch’), placement of bilateral Penrose drains (‘drain’), and a delayed laparotomy at 14 days after providing total parenteral nutrition (‘wait’) [41]. ‘NEC totalis’ is a term used to describe pan-intestinal involvement with near total bowel compromise. Typically NEC totalis is defined as disease in which less than 25% of bowel is viable. In this desperate clinical scenario, there has been some anecdotal success with proximal diversion without resection followed by a delayed second-look at 6–8 weeks after initial surgery [42,43]. Mortality and morbidity remain high with almost all survivors having short-bowel syndrome. These patients may benefit from long-term, multidisciplinary intestinal rehabilitation with close monitoring of parenteral nutrition delivery. Also, a variety of intestinal lengthening procedures have been described including longitudinal intestinal lengthening and tailoring operation as described by Bianchi as well as serial transverse enteroplasty procedure [44,45]. Intestinal transplantation can be considered in patients with short bowel syndrome from NEC who develops cholestasis and other complications related to dependence on parenteral nutrition [46]. ‘NEC totalis’ carries a grim prognosis. Withdrawal of treatment may be considered after a candid discussion with the patient’s family [47].
7. Long-term complications and outcomes A common delayed complication after recovering from both surgically and medically managed NEC is the formation of intestinal stricture. The incidence of intestinal stricture may be as high as 36% with most strictures occurring in the left colon [10,48,49]. These patients will develop feeding intolerance and distension that resolves when feeds are withheld. Radiologically, strictures usually can be identified using contrast studies. Another frequent complication is short-bowel syndrome and the associated cholestatic liver
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
disease that develops from long-term parenteral nutrition dependence. These patients are also at risk for infectious complications related to long-term central venous access. Recurrent NEC can also manifest after patients undergo surgical intervention. The overall incidence of recurrent NEC is estimated at 6% [10,50]. Fortunately, the majority of patients with recurrent NEC are successfully treated nonoperatively [50]. Among low birth weight infants, patients with surgically managed NEC have worse long-term neurodevelopmental and growth outcomes when compared to patients without NEC [51,52]. Meanwhile, patients with medically managed NEC have similar neurodevelopmental and growth outcomes to patients without NEC [51]. The currently ongoing NICHD Neonatal Research Network sponsored NEST trial is designed to assess long-term outcomes among patients undergoing surgical treatment of NEC [30].
8. Conclusions When evaluating patients with NEC, surgical intervention is reserved for those patients with pneumoperitoneum, confirmed stool or pus in the peritoneal cavity, or worsening clinical status. Various surgical options range from peritoneal drain placement to laparotomy with enterostomy creation, anastomoses, or temporizing measures with planned second look operations. The choice of operation has not been shown to have a significant effect on clinically significant outcomes and must be made after careful consideration of the physiologic state of the patient. Ultimately, patients with NEC requiring surgical intervention have more morbidity and mortality than patients with NEC treated medically and the mortality rate for surgically managed NEC remains dependent upon disease severity and is directly correlated to gestational age.
References [1] D.A. Kafetzis, C. Skevaki, C. Costalos, Neonatal necrotizing enterocolitis: an overview, Curr. Opin. Infect. Dis. 16 (4) (2003) 349–355. [2] S.O. Guthrie, P.V. Gordon, V. Thomas, et al., Necrotizing enterocolitis among neonates in the United States, J. Perinatol. 23 (4) (2003) 278–285. [3] J.A. Bisquera, T.R. Cooper, C.L. Berseth, Impact of necrotizing enterocolitis on length of stay and hospital charges in very low birth weight infants, Pediatrics 109 (3) (2002) 423–428. [4] K. Sankaran, B. Puckett, D.S. Lee, et al., Variations in incidence of necrotizing enterocolitis in Canadian neonatal intensive care units, J. Pediatr. Gastroenterol. Nutr. 39 (4) (2004) 366–372. [5] B. Lemyre, W. Xiu, N.R. Bouali, et al., A decrease in the number of cases of necrotizing enterocolitis associated with the enhancement of infection prevention and control measures during a Staphylococcus aureus outbreak in a neonatal intensive care unit, Infect. Control Hosp. Epidemiol. 33 (1) (2012) 29–33.
109
[6] R.L. Moss, L.A. Kalish, C. Duggan, et al., Clinical parameters do not adequately predict outcome in necrotizing enterocolitis: a multiinstitutional study, J. Perinatol. 28 (10) (2008) 665–674. [7] W.H. Yee, A.S. Soraisham, V.S. Shah, et al., Incidence and timing of presentation of necrotizing enterocolitis in preterm infants, Pediatrics 129 (2) (2012) e298–e304. [8] A.M. Kosloske, Indications for operation in necrotizing enterocolitis revisited, J. Pediatr. Surg. 29 (5) (1994) 663–666. [9] J.J. Tepas 3rd, R. Sharma, M.L. Hudak, et al., Coming full circle: an evidence-based definition of the timing and type of surgical management of very low-birth-weight (
Current concepts in the surgical approach to necrotizing enterocolitis Mehul V. Raval, R. Lawrence Moss ∗ Division of Pediatric Surgery, Nationwide Children’s Hospital, The Ohio State University College of Medicine, Columbus, OH, USA
Abstract Necrotizing enterocolitis (NEC) is the most common surgical emergency occurring in neonatal intensive care unit (NICU) patients. Among patients with NEC, those that require surgery experience the poorest outcomes and highest mortality. Surgical intervention, while attempting to address the intestinal injury and ongoing mulitfactorial physiologic insults in NEC is associated with its own stresses that may compound the ongoing physiologic derangement. Surgery is thus reserved for those patients with clear indication for intervention such as pneumoperitoneum, confirmed stool or pus in the peritoneal cavity, or worsening clinical status. The purpose of this review is to briefly describe the physiologic stress induced by surgical intervention in the preterm, low birth weight patient with NEC and to provide a contemporary overview of available surgical management options for NEC. The optimal surgical plan employed is strongly influenced by clinical judgment and theoretical benefits in terms of minimizing physiologic stressors while providing temporary and/or definitive treatment in a timely fashion. While the choice of operation has not been shown to have a significant effect on any clinically important outcomes, ongoing investigations continue to study both short and long-term outcomes in patients with NEC. © 2013 Elsevier Ireland Ltd. All rights reserved. Keywords: Necrotizing enterocolitis; Surgery; Neonatal surgery; Neonatal intensive care unit; Prematurity
1. Introduction Necrotizing enterocolitis (NEC) remains one of the most challenging diseases to diagnose and treat among neonatal intensive care unit (NICU) patients. NEC is the most common surgical emergency and the most common cause of gastrointestinal-related morbidity in the NICU [1,2]. Patients with NEC who undergo surgical interventions have the worst outcomes and highest costs [3]. Though the definitive pathophysiology of NEC has not been fully elucidated, NEC is predominantly a disease of low birth weight, premature infants, which is only occasionally seen in term infants [2]. The changes in intestinal microbiota, loss of the gut barrier, microciculatory changes, and other pathophysiologic etiologies observed in NEC are more pronounced in the premature newborn with naïve physiology, limited reserve, heightened fluid and electrolyte balance sensitivity, increased energy ∗ Corresponding author at: Nationwide Children’s Hospital, E. Thomas Boles Jr., Professor of Surgery, The Ohio State University, College of Medicine, 700 Children’s Drive, Columbus, OH 43205, USA. Tel.: +1 614 722 5163; fax: +1 614 722 5995. E-mail address: [email protected] (R.L. Moss).
0928-4680/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.pathophys.2013.11.017
requirements, and altered immunologic and stress response. As overall NICU care has advanced and the survival of preterm infants has improved, a corresponding increase in prevalence of NEC has been observed. Surgical intervention, while attempting to address the intestinal injury and ongoing mulitfactorial physiologic insult, is associated with a variety of inherent stresses that may compound the physiologic derangement observed in NEC. While the pathophysiologic basis of NEC is extensively covered in other chapters in this compendium, the purpose of this review is to briefly describe the physiologic stress induced by surgical intervention. Specific topics that may be of significance in the low birth weight, premature infant undergoing surgery will be addressed. Subsequently, the various surgical options available in the management of NEC will be explored. The surgical option employed is strongly influenced by clinical judgment and theoretical benefits in terms of minimizing physiologic stressors while providing temporary and/or definitive treatment in a timely fashion. While the choice of operation has not been shown to have a significant effect on any clinically important outcomes, ongoing investigations continue to study both short and long-term outcomes in patients with NEC.
106
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
2. Presentation and diagnosis A variety of factors have been proposed to be associated with the development of NEC including prematurity, enteral feedings, steroid use, hypotension/ischemia, presence of a patent ductus arteriosus, and infection [2,4,5]. Among these, only prematurity has been consistently associated with NEC and need for surgical intervention [6,7]. Early recognition of NEC development is a clinical challenge. Early NEC can be difficult to distinguish from relatively benign conditions such as feeding intolerance commonly associated with prematurity to severe conditions such as neonatal sepsis. Subtle initial signs such as feeding intolerance and abdominal distention are often followed by temperature instability, lethargy, apnea, and bradycardia. Severe or “surgical” disease often includes the development of abdominal tenderness, peritonitis, abdominal wall discoloration, and identification of palpable abdominal mass. Though fixed abdominal mass and abdominal wall erythema are highly specific signs of intestinal necrosis, less than 10% of patients with surgical NEC will have these findings [8]. At the other end of the clinical spectrum, NEC can progress to multi system organ failure or death within a few hours of onset. Though most commonly occurring in the terminal ileum, NEC may involve any and/or non-contiguous portions of intestine. When the entire gut is ischemic, the clinical picture is referred to as “NEC totalis”. Laboratory findings are often non-specific during the initial phases of NEC. Metabolic acidosis, neutropenia, left shift of segmented neurtrophils, and hyponatremia are the most concerning laboratory aberrations. Thrombocytopenia, especially with a rapid drop in the platelet count, is a poor prognostic sign [1]. A combination of these laboratory values has been used in scoring systems to predict the timing and type of surgical intervention in NEC [9]. Attempts to identify biochemical markers specific to necrotizing enterocolitis have not been successful and remain an area of ongoing research. A variety of imaging modalities are now available in the NICU population but abdominal plain films have remained the mainstay for NEC diagnosis for decades. Pneumoperitoneum identified on plain films mandates surgical intervention [10]. Pneumotosis intestinalis, the most specific finding for diagnosing NEC, as well as portal venous gas or fixed dilated air-filled loops can confirm the diagnosis but are not necessarily indications for surgical intervention (Fig. 1). Abdominal ultrasound is increasingly available in most NICU settings and can be used to assess intestinal peristalsis, bowel wall thickness, and to identify to fluid collections. Ultrasound technology has improved to include Doppler flow assessment as well as more detailed imaging ability with high frequency transducers. Ultrasound can be readily performed at the bedside, is noninvasive, and avoids ionizing radiation. Though several institutions with experienced technicians have shown high sensitivity and specificity of ultrasound to detect necrotic bowel in NEC [11,12], the generalizability of this imaging modality to consistently provide information beyond that
Fig. 1. Plain radiograph of extensive pneumotosis intestinalis and free air in a patient with necrotizing enterocolitis.
provided in plain films remains to be determined. Other imaging studies such as contrast studies, computed tomography scans, and magnetic resonance imaging are available but have not proven to be of significant utility in the acute management of NEC [13]. Ultimately, surgical intervention is reserved for those patients with pneumoperitoneum, confirmed stool or pus in the peritoneal cavity, or worsening clinical status.
3. The pysiologic stress of surgery Once the decision to proceed with surgery has been made it is imperative to anticipate the physiologic impact of the ensuing stress response of the surgical intervention itself. Operative intervention is often described as a controlled, induced trauma. Numerous changes occur in response to the surgical intervention from increased metabolic demands, to triggered inflammatory and immune cascades, and to alterations in the endocrine balance [14]. All of these changes are magnified in the preterm newborn [15]. After operative trauma, metabolism increases to provide substrates needed for regenerating and healing tissue. In newborns undergoing major abdominal surgery, catecholamine release, oxygen consumption, and resting energy expenditure peak at 4 h postoperatively and then return rapidly to baseline within the first 24 h of surgery [16,17]. Moderate intraoperative opiate dosing has been shown to blunt the immediate stress response and may be associated with an improved postoperative
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
outcomes [17]. The surge of opiate release around the time of umbilical cord clamping may partially explain the reduced metabolic stress response seen in newborns undergoing surgical procedures in the first few hours of birth [18]. After surgery, the immune system is both stimulated to combat opportunistic postoperative infection while at the same time dampened to facilitate healing. Key cytokines released include interleukins (IL) such as IL-1 and IL-6 as well as tumor necrosis factor ␣. These pro-inflammatory cytokines are balanced by anti-inflammatory IL1ra and IL-10. These cytokines and others trigger specific gene expression in target cells, trigger release of regulatory hormones, and facilitate cell-to-cell communication. These actions have a direct impact on third spacing of fluids, metabolic demands, and culminate in wound healing. Peak levels of specific cytokines, such as IL-6, may directly correlate with the degree of operative trauma in the neonate [19]. Other mediators of cellular response altered by surgical trauma include histamine, serotonin, cortisol, and various components of the kinin and complement systems. The endocrine response to surgery has also been well studied with changes in the insulin/glucose ratio in the immediate postoperative period. In neonates, hyperglycemia is often observed after surgical intervention and the degree of hyperglycemia may be directly correlated to the degree of surgical stress [20]. Ultimately, the balance of pro and anti-inflammatory response, increased metabolic needs, fluid-shifts, and overall homeostasis must be closely observed in the postsurgical neonate.
4. The role of peritoneal drainage The two main options available for the surgical management of NEC are exploratory laparotomy and primary peritoneal drainage (PPD). Though laparotomy with bowel resection has been the classic surgical approach, PPD has been considered an option for over four decades (Fig. 2) [21]. The role of PPD has changed from initial use as a temporizing measure to more broad use as a potential definitive treatment for NEC [21–23]. Ein et al. published their experience with neonates weighing less than 1000 grams demonstrating a 32% success of PPD alone without the need of subsequent laparotomy [23,24]. While several groups reported similar success [25,26], others reports showed conflicting results [24,27,28]. Most of these studies included small numbers of patients, few were prospective, and none were randomized. A 2001 meta-analysis attempted to adjust for the fact that PPD was often employed in the most premature and ill cohorts, and a difference in outcome could not be assessed [27]. In 2006, a prospective, observational study was published comparing laparotomy to PPD for patients of 1 kg or less using data accrued at 16 centers with the National Institute of Child Health and Human Development (NICHD) Neonatal Research Network [29]. Regardless of surgical treatment, outcomes for the entire cohort were poor with 60% either having died or received prolonged parenteral nutrition
107
Fig. 2. Drawing of primary peritoneal drain placement in a neonate with necrotizing enterocolitis.
by discharge and 72% having died or shown to have had neurodevelopmental delay by the time of follow-up. When comparing the patients based on type of surgical intervention the PPD group had lower gestational age, postnatal age, systolic blood pressure at the time of intervention, and were more likely to be treated with high frequency oscillatory ventilation or have higher peak inspiratory pressures on conventional ventilation compared to those undergoing laparotomy. This study suggested a possible difference in long-term neurodevelopmental outcomes favoring laparotomy prior to disease severity adjustment and serves as the basis for the current NICHD Neonatal Research Network NEST trial [30]. Two randomized controlled trials have been performed comparing PPD to laparotomy. The North American trial was published in 2006 (NECSTEPS) and the European trial in 2008 (NET trial) [31,32]. Both trials were well conducted and represented multiple centers (15 centers in North America for NECSTEPS and 18 centers across 11 countries in the NET trial). These trials focused on low birth weight, premature infants with NECSTEPS outlining a 1500 g birth weight limit and a gestational age less than 34 weeks. The NET trial limited inclusion to those infants less than or equal to 1000 g. Key differences between the trials involved the technique used and patient management in the peritoneal drainage groups. In NECSTEPS, operating surgeons were encouraged to irrigate through the peritoneal drain incision prior to drain placement. In the NET trial, surgeons were discouraged from performing any irrigation to wash away stool or pus. Along the same lines, NECSTEPS patients were managed with peritoneal drainage as a definitive therapy. If patients worsened then additional drains were allowed. Only 5 patients in this trial were converted to laparotomy for early clinical deterioration. Conversely, the NET trial considered peritoneal
108
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
drainage a temporizing treatment with 74% of patients in the NET trial assigned to peritoneal drainage ultimately undergoing laparotomy at a median of 2.5 days after randomization. Peritoneal drainage was utilized as effective definitive treatment in 11% of infants in the NET trial and nearly 33% in the NECSTEPS trial. The primary outcome for both trials was survival. This was measured at one and six months in the NET trial and at 90 days in NECSTEPS. Both studies identified survival rates that were not statistically different between peritoneal drainage and laparotomy. Length of stay was evaluated in both trials as a secondary outcome and both trials found no statistical difference in length of stay between laparotomy and PPD. Similarly, there were no statistical differences in various other outcomes of interest in one or both trials studied various time points including ventilator dependence, parenteral nutrition dependence, and time to full enteral nutrition. Neither study addressed neurodevelopmental outcomes or cost-effectiveness. The conclusions garnered from these studies differed based on how the studies approached use of peritoneal drains. In NECSTEPS, where PPD was considered definitive treatments, authors concluded that there was no significant difference between primary peritoneal drainage and laparotomy for initial management. Conversely, the NET trial, where peritoneal drains were considered a temporizing measure to laparotomy, authors concluded that primary peritoneal drainage is ineffective as either a temporizing measure or definitive treatment. A recent Cochrane Review combining the results of the two studies concluded that no significant benefits or harms of peritoneal drainage versus laparotomy could be identified [33].
5. Laparotomy and stoma creation Laparotomy with resection of necrotic bowel and with enterostomy creation has been the traditional approach to NEC [34]. Advantages include an expeditious intervention that addresses the physiologic insult. The disadvantages of this approach that must be considered include a need for parenteral nutrition and the risk of high stoma output and salt and water loss. Stoma specific complications include retraction, stricture, prolapse, obstruction, and skin/wound issues [35]. A second anesthetic is also required for stoma takedown typically when the patient’s weight is over 2 kg and at least 4 weeks after stomas maturation to avoid encountering vascular intra-abdominal adhesions [36].
6. Surgical alternatives Laparotomy with resection and primary anastomosis is another option that eliminates the mandatory need for a second operative intervention to reestablish intestinal continuity [37]. Much of the physiologic stress of stoma creation such as
fluid losses and electrolyte abnormalities can thus be avoided. Some groups have even endorsed performing numerous anastomoses in the setting of multiple areas of disease [38]. Disadvantages to performing even a single anastomoses in a NEC patient undergoing laparotomy include leak and stricture related to working with inflamed, edematous tissue in a hemodynamically unstable patient. Anastomotic complications have the potential to serve as spark that could easily lead the fragile neonate back into sepsis, multisystem organ dysfunction, and death. Primary anastomsis in the acutely ill NEC patient remains an area of ongoing investigation. Other surgical options include a ‘clip and drop’ technique where grossly necrotic bowel is resected and the proximal and distal aspects are closed using titanium clips [39]. These ends are placed back into the peritoneal cavity and a subsequent operation is then performed within 48–72 h to assess for possible bowel anastomosis in a more physiologic stable patient [40]. Another technique described involves laparotomy with transverse closure of areas of frank bowel wall injury (‘patch’), placement of bilateral Penrose drains (‘drain’), and a delayed laparotomy at 14 days after providing total parenteral nutrition (‘wait’) [41]. ‘NEC totalis’ is a term used to describe pan-intestinal involvement with near total bowel compromise. Typically NEC totalis is defined as disease in which less than 25% of bowel is viable. In this desperate clinical scenario, there has been some anecdotal success with proximal diversion without resection followed by a delayed second-look at 6–8 weeks after initial surgery [42,43]. Mortality and morbidity remain high with almost all survivors having short-bowel syndrome. These patients may benefit from long-term, multidisciplinary intestinal rehabilitation with close monitoring of parenteral nutrition delivery. Also, a variety of intestinal lengthening procedures have been described including longitudinal intestinal lengthening and tailoring operation as described by Bianchi as well as serial transverse enteroplasty procedure [44,45]. Intestinal transplantation can be considered in patients with short bowel syndrome from NEC who develops cholestasis and other complications related to dependence on parenteral nutrition [46]. ‘NEC totalis’ carries a grim prognosis. Withdrawal of treatment may be considered after a candid discussion with the patient’s family [47].
7. Long-term complications and outcomes A common delayed complication after recovering from both surgically and medically managed NEC is the formation of intestinal stricture. The incidence of intestinal stricture may be as high as 36% with most strictures occurring in the left colon [10,48,49]. These patients will develop feeding intolerance and distension that resolves when feeds are withheld. Radiologically, strictures usually can be identified using contrast studies. Another frequent complication is short-bowel syndrome and the associated cholestatic liver
M.V. Raval, R.L. Moss / Pathophysiology 21 (2014) 105–110
disease that develops from long-term parenteral nutrition dependence. These patients are also at risk for infectious complications related to long-term central venous access. Recurrent NEC can also manifest after patients undergo surgical intervention. The overall incidence of recurrent NEC is estimated at 6% [10,50]. Fortunately, the majority of patients with recurrent NEC are successfully treated nonoperatively [50]. Among low birth weight infants, patients with surgically managed NEC have worse long-term neurodevelopmental and growth outcomes when compared to patients without NEC [51,52]. Meanwhile, patients with medically managed NEC have similar neurodevelopmental and growth outcomes to patients without NEC [51]. The currently ongoing NICHD Neonatal Research Network sponsored NEST trial is designed to assess long-term outcomes among patients undergoing surgical treatment of NEC [30].
8. Conclusions When evaluating patients with NEC, surgical intervention is reserved for those patients with pneumoperitoneum, confirmed stool or pus in the peritoneal cavity, or worsening clinical status. Various surgical options range from peritoneal drain placement to laparotomy with enterostomy creation, anastomoses, or temporizing measures with planned second look operations. The choice of operation has not been shown to have a significant effect on clinically significant outcomes and must be made after careful consideration of the physiologic state of the patient. Ultimately, patients with NEC requiring surgical intervention have more morbidity and mortality than patients with NEC treated medically and the mortality rate for surgically managed NEC remains dependent upon disease severity and is directly correlated to gestational age.
References [1] D.A. Kafetzis, C. Skevaki, C. Costalos, Neonatal necrotizing enterocolitis: an overview, Curr. Opin. Infect. Dis. 16 (4) (2003) 349–355. [2] S.O. Guthrie, P.V. Gordon, V. Thomas, et al., Necrotizing enterocolitis among neonates in the United States, J. Perinatol. 23 (4) (2003) 278–285. [3] J.A. Bisquera, T.R. Cooper, C.L. Berseth, Impact of necrotizing enterocolitis on length of stay and hospital charges in very low birth weight infants, Pediatrics 109 (3) (2002) 423–428. [4] K. Sankaran, B. Puckett, D.S. Lee, et al., Variations in incidence of necrotizing enterocolitis in Canadian neonatal intensive care units, J. Pediatr. Gastroenterol. Nutr. 39 (4) (2004) 366–372. [5] B. Lemyre, W. Xiu, N.R. Bouali, et al., A decrease in the number of cases of necrotizing enterocolitis associated with the enhancement of infection prevention and control measures during a Staphylococcus aureus outbreak in a neonatal intensive care unit, Infect. Control Hosp. Epidemiol. 33 (1) (2012) 29–33.
109
[6] R.L. Moss, L.A. Kalish, C. Duggan, et al., Clinical parameters do not adequately predict outcome in necrotizing enterocolitis: a multiinstitutional study, J. Perinatol. 28 (10) (2008) 665–674. [7] W.H. Yee, A.S. Soraisham, V.S. Shah, et al., Incidence and timing of presentation of necrotizing enterocolitis in preterm infants, Pediatrics 129 (2) (2012) e298–e304. [8] A.M. Kosloske, Indications for operation in necrotizing enterocolitis revisited, J. Pediatr. Surg. 29 (5) (1994) 663–666. [9] J.J. Tepas 3rd, R. Sharma, M.L. Hudak, et al., Coming full circle: an evidence-based definition of the timing and type of surgical management of very low-birth-weight (
