Spontaneous intestinal perforation of the newborn

INTRODUCTION — Spontaneous intestinal perforation (SIP) of the newborn, also referred to as isolated perforation or focal intestinal perforation (FIP) of the newborn, is a single intestinal perforation that is typically found at the terminal ileum [1-5]. SIP occurs primarily in preterm infants with very low birth weight (VLBW, birth weight <1500 g) and extremely low birth weight (ELBW, birth weight <1000 g). SIP is a separate clinical entity from necrotizing enterocolitis, the most severe gastrointestinal complication of preterm infants. This differentiation is important because of management considerations. (See "Neonatal necrotizing enterocolitis: Clinical features and diagnosis".)

The epidemiology, clinical features, diagnosis, and management of spontaneous intestinal perforation of the newborn will be reviewed here.

EPIDEMIOLOGY — Spontaneous intestinal perforation (SIP) is most commonly found in very low birth weight (VLBW, BW <1500 g) and extremely low birth weight (ELBW, BW <1000 g) preterm infants. The reported incidence for VLBW infants with a gestational age (GA) ≤32 weeks was 1.6 percent based on the National Inpatient Sample dataset from 2002 to 2017 of 658,001 infants [6]. In this cohort, the incidence of SIP increased with decreasing gestation, and 90 percent of cases with SIP were ≤28 weeks gestation, and 82 percent were ELBW infants. SIP occurs more frequently in male infants than female infants [6].

Risk factors — Prematurity is the only well-established risk factor for SIP. Other reported antenatal and postnatal risk factors are based upon limited data that include findings from case series [1,7-9]. It remains uncertain whether or not there is an association between SIP and the following factors.

●Antenatal

•Chorioamnionitis – Severe placental chorioamnionitis appears to be an antenatal risk factor for SIP. In one case-control study of 16 preterm infants with SIP, infants with SIP versus matched controls were more likely to have severe placental chorioamnionitis with evidence of fetal vascular response based upon a blinded retrospective histologic examination of the placenta and umbilical vessels (40 versus 12 percent) [7]. In this study, mothers of infants with SIP were more likely to have received antibiotics before or at delivery (93 versus 57 percent).

•Antenatal drugs – Although antenatal administration of glucocorticoids, nonsteroidal anti-inflammatory drugs (NSAIDs) and magnesium sulfate had been initially reported to increase the risk of SIP, but the quality of evidence is poor on whether antenatal exposure to any of these drugs is associated with SIP [9-14].

•Other – In a small retrospective study, additional reported risk factors for SIP in VLBW infants in 8 cases versus 32 controls included oligohydramnios, velamentous cord insertion, lower one minute Apgar scores, and primigravity [15]. However, these findings need to be confirmed in a larger cohort of cases.

●Postnatal

•Exposure to glucocorticoids – Early administration of postnatal glucocorticoids increases the risk of SIP [16-18]. This was best illustrated in a meta-analysis of four trials of prophylactic dexamethasone for the prevention of bronchopulmonary dysplasia in which VLBW infants treated with dexamethasone before 48 hours of life had an increased risk of SIP compared with controls (odds ratio [OR] 1.91, 95% CI 1.21-3.07) [16]. (See "Postnatal use of glucocorticoids for prevention of bronchopulmonary dysplasia (BPD) in preterm infants", section on 'Adverse effects'.)

•Exposure to indomethacin – Although the postnatal use of indomethacin had been previously reported to increase the risk of SIP [11], subsequent publications have reported no association between indomethacin exposure and SIP [8,19,20]. (See "Patent ductus arteriosus (PDA) in preterm infants: Management and outcome", section on 'Indomethacin'.)

However, conflicting data from a Canadian retrospective cohort study of 4268 ELBW infants who were born at <30 weeks gestation found that the prophylactic administration of indomethacin, combined with early feeding, was associated with an increased odds of SIP, but early feeding alone was not associated with SIP. The study also noted marked variation in the use of prophylactic indomethacin across the 13 treatment centers [21].

•Combination of steroids and indomethacin versus steroids – A study from the Canadian Neonatal Network of 4729 preterm infant (GA <26 weeks or birth weight [BW] <750 g) reported that the combination of antenatal steroids with postnatal indomethacin increased the risk of SIP compared with antenatal steroids alone (adjusted OR, 1.61, 95% CI 1.14-2.28) [13].

•Other reported postnatal risk factors include the use of inotropic agents, delayed onset of feeding and intraventricular hemorrhage of Grade III or higher [22,23]. However, these risk factors might be surrogate markers for ill premature infants, rather than causative.

PATHOLOGY AND PATHOGENESIS — In very low birth weight (VLBW) neonates with spontaneous intestinal perforation (SIP), there is a single isolated perforation that is typically found on the antimesenteric border of the terminal ileum, but perforation has also been reported in the jejunum and colon [1-5]. At the area of perforation, focal hemorrhagic necrosis with well-defined margins is observed in contrast to the ischemic and coagulative necrosis seen in necrotizing enterocolitis (NEC) [4,24]. In SIP, the bowel proximal and distal to the perforation appears normal (picture 1).

The etiology of SIP remains unknown. Several studies have explored a variety of different pathogenetic pathways.

●Thinning or absence of the muscularis propria at the perforation site has been reported in several cases [2,24-27], but it is unclear whether these changes are involved in the pathogenesis of SIP.

●A role for abnormal or delayed nitric oxide synthase (NOS) has been hypothesized based upon a single study of a NOS knock-out mouse model that demonstrated ileal perforation with exposure to indomethacin and/or dexamethasone [28].

●Changes in immunoregulatory pathways including angiogenesis, arginine metabolism, cell adhesion and chemotaxis, extracellular matrix remodeling, hypoxia and oxidative stress, inflammation, and muscle contraction. Although, there is upregulation of these processes in SIP, it is milder in intestinal and plasma samples taken from patients with SIP versus those with NEC [29,30].  

CLINICAL MANIFESTATIONS

Age at presentation — SIP occurs in very low birth weight (VLBW) and extremely low birth weight (ELBW) preterm infants with a gestational age between 25 and 27 weeks. In several case series of SIP, the median age at perforation is seven days with a range of 0 to 15 days [1,2,4,8,31-33]. An analysis from the Pediatrix Medical Group data set demonstrated that neonates with SIP presented earlier in life than those with NEC (median age 7 versus 15 days of life) [8].

Risk factors for SIP may vary according to the time of presentation. In a Pediatrix analysis of 633 neonates with SIP, patients with SIP were divided into two groups based upon their age of presentation [33]. Infants in the first group (n = 116), who presented between zero and three days of life, were larger (median birth weight 1.4 kg) and less likely to have received antenatal glucocorticoids, indomethacin, surfactant, or mechanical ventilation compared with infants in the second group who presented between 7 and 10 days of age with a median birth weight of 775 g.

Physical findings — Infants with SIP present with an acute onset of abdominal distension. Abdominal distention usually occurs without the abdominal wall erythema, crepitus, and induration commonly seen in patients with NEC.

A black-bluish discoloration of the abdominal wall may be seen in SIP, and is not a finding that is typical of NEC (picture 2) [1,4,25,32,34,35]. The discoloration may extend into the groin and, in males, the scrotum.

Other associated findings include hypotension, tachycardia, and signs of peritonitis.

Associated infections — In several case series, concomitant sepsis due to coagulase-negative Staphylococcus or fungemia due to Candida albicans have been reported in neonates with SIP and may be a major cause of morbidity and mortality [1,2,7,32]. It is unknown whether the infections precede or are a result of bowel perforation.

Laboratory tests — Laboratory findings are nonspecific and are not useful in the diagnosis of SIP. They include the following:

●Leukocytosis (white cell counts between 14,000 and 25,000/microL) occurs in most infants with SIP [32,35].

●Serum alkaline phosphatase and bilirubin are often increased [35].

●A decrease in platelet count and hematocrit may be observed [1].

DIAGNOSIS — A diagnosis for suspected spontaneous intestinal perforation (SIP) is based upon the presence of the characteristic clinical features in a preterm infant within the first 10 days of life. These findings include abdominal distension, often with the classical bluish discoloration of the abdominal wall in the absence of abdominal wall erythema (picture 2), crepitus, and induration. Imaging studies are useful in supporting the diagnosis of SIP, if there are findings that differentiate SIP from necrotizing enterocolitis (NEC). (See 'Imaging' below.)

The diagnosis of SIP is strongly suspected if there are clinical and radiologic findings of pneumoperitoneum in the absence of pneumatosis intestinalis and portal venous air. The diagnosis of SIP is made by operative findings that demonstrate an isolated bowel perforation in the setting of otherwise normal bowel (picture 1), which is confirmed by histopathologic examination. (See 'Surgical treatment' below.)

Imaging — Abdominal radiographs that demonstrate pneumoperitoneum support the diagnosis of SIP (image 1 and image 2). Images are performed in the supine position and, to detect pneumoperitoneum, in either the supine cross-table lateral view or in the lateral decubitus position with the left side down (image 2). However, a gasless abdomen may be seen in patients with SIP when the perforation is walled off and free air is not present, but this is not a typical finding [4,32]. In these patients, ultrasound may be useful in making the diagnosis of perforation if echogenic free fluid is detected [36].

DIFFERENTIAL DIAGNOSIS: NECROTIZING ENTEROCOLITIS — The primary differential diagnosis of SIP is necrotizing enterocolitis (NEC). SIP and NEC are gastrointestinal conditions with  overlap in presentation and treatment that typically occur in very low birth weight (VLBW) and extremely low birth weight (ELBW) preterm infants with a gestational age between 25 and 27 weeks [37]. However, since the optimal management of these two disorders may differ, it remains helpful to distinguish SIP from NEC.

Despite the overlap, the following common findings of SIP typically distinguish it from NEC [1,4,25,32,34,38]:

●SIP generally presents within the first week of life. In contrast, NEC typically presents after the first week of life after the infant has begun to feed.

●SIP has the characteristic physical finding of significant abdominal distention, often accompanied by a bluish discoloration (picture 2).

●Specific radiographic findings on abdominal radiographs diagnostic for NEC are not observed in patients with SIP include pneumatosis intestinalis, portal venous air, transient thickening of the intestinal wall, and fixed dilated small bowel loops. Although pneumoperitoneum is a common supportive radiologic finding for SIP occurring in approximately 30 to 50 percent of patients with SIP, it is not a distinguishing feature as it also frequently occurs in patients with NEC.

MANAGEMENT

Initial management — Initial management of a patient with suspected spontaneous intestinal perforation (SIP) is directed toward stabilization of the patient prior to surgery and includes the following:

●Cessation of all feeds and enteral medications

●Nasogastric suction to decompress the abdomen

●Supportive care including fluid resuscitation and inotropic medications to correct hypotension

●Intravenous antibiotics

Surgical treatment — Surgery has been considered the definitive treatment for SIP [4,25,39], although many pediatric surgeons now consider primary peritoneal drainage (PPD) as the initial treatment of choice. If successful, PPD, which is performed at the bedside, avoids laparotomy, the need for general anesthesia, and transport to the operating room for these very preterm and often severely ill patients. Many pediatric surgeons support the use of PPD in infants with SIP, as many of these patients will recover without any further surgical intervention [40].

In a single multicenter randomized trial of infants with birth weights (BW) <1000 g (mean gestational age 25 weeks), the risk of the primary outcome of death or neurodevelopmental impairment (NDI) at 18 to 22 months corrected age was similar between infants with SIP randomized to initial laparotomy (n = 99) and PPD (n = 102), (69 versus 63 percent, adjusted relative risk 1.11, 95% CI 0.95 to 1.31) [41]. However, this trial was limited by the sample size, which in part was due to a low enrollment of eligible infants (31 percent).

The results of this trial are consistent with previous observational data, including a multicenter, retrospective study of 171 infants with SIP [42-47]. In this report, similar outcomes were noted for infants treated by PPD (n = 110) and primary laparotomy (n = 61) in terms of complications, time to resume feeds, length of stay, and mortality [42].

In addition, there is an observational report of the successful use of peritoneal needle aspiration in 18 of 31 patients who presented with SIP in a single French center [46]. However, we do not recommend this procedure until there are further studies that confirm efficacy and safety.

Our approach — At our centers, PPD performed at the bedside is the initial preferred procedure for infants with SIP. Our practice is to use only a single drainage site, although other surgeons perform a counter incision in the left lower quadrant and thread the Penrose between the two incisions. After the administration of intravenous pain medication, the abdomen is prepped with iodine solution, and local anesthesia is administered. A small transverse incision is made at McBurney's point (two-thirds of the distance from the umbilicus to the anterior superior iliac crest in the right lower quadrant). The layers of the abdominal wall are bluntly dissected to enter the peritoneal cavity. In many cases, there will be a rush of air followed by drainage of meconium. Cultures are taken and a Penrose drain is gently threaded into the abdomen and secured. In some centers, the peritoneal cavity is irrigated with warm saline solution prior to the insertion of the Penrose drain, although there are no data to support this practice.

The drain site is observed over the subsequent days. When there is no evidence of persistent intestinal or meconium drainage from the drain site, the drain is backed out daily until it is removed. It is not necessary to keep the drain in place until the infant is fed. After the return of bowel function, a trial of feeding can be started or the patency of the gastrointestinal tract can be determined with a contrast study.

Laparotomy is indicated following PPD in the following settings [45]:

●Reaccumulation of free air after the drain is removed, indicating the perforated bowel did not seal.

●Ongoing sepsis, indicating a failure of the drain to control contamination and ongoing peritonitis.

●Fistula with persistent intestinal drainage that fails to close after several weeks of observation.

●Bowel obstruction, either due to adhesions or, more commonly, a stricture at the site of the perforation.

OUTCOME — SIP is associated with increased mortality [6]. However, long-term survival of infants with spontaneous intestinal perforation (SIP) has improved over the past 30 years with reported survival rates of 64 to 90 percent, regardless of treatment with primary peritoneal drainage (PPD) or laparotomy [4-6,32,34,43,48,49]. The increase in survival is due to advances in neonatal care including improved parenteral nutrition and antibiotics.

After adjusting for confounding factors, patients with SIP appear to have a lower mortality rate compared with patients with NEC [6,50-52]. Patients with SIP and NEC appear to have similar risks for either death or neurodevelopmental impairment [50,53,54].

In a study from the National Institute of Child Health and Human Development (NICHD) Neonatal Research Network, death or neurological impairment occurred in 82, 79, and 53 percent of patients with surgical NEC, SIP, and those without NEC or SIP, respectively. Multivariate analysis demonstrated an increased risk of death and neurologic impairment for patients with SIP compared with controls (adjusted odds ratio [OR] 2.21, 95% CI 1.5-3.2) and patients with surgical NEC (adjusted OR 2.11, 95% 1.5-9.2). In a separate analysis of survivors, the risk of neurodevelopmental impairment was increased in those with SIP (adjusted OR 2.17, 95% CI 1.4-3.2) and surgical NEC (adjusted OR 1.7, 95% CI 1.2-2.4) compared with controls.

Infants with SIP also appear to be at increased risk to develop retinopathy of prematurity and periventricular leukomalacia compared with those without SIP [38]. (See "Overview of the long-term complications of preterm birth" and "Retinopathy of prematurity (ROP): Risk factors, classification, and screening", section on 'Risk factors'.)

A retrospective case-matched cohort study from a single center reported poorer neurodevelopmental outcome for patients with SIP was due to an increase in the risk of severe intraventricular hemorrhage in patients with SIP compared with matched controls (21 versus 7 percent) [49]. In this study, based on parental phone surveys, there was no difference in gastrointestinal outcome and quality of life between the two groups.

SUMMARY AND RECOMMENDATIONS

●Introduction and pathology – Spontaneous intestinal perforation (SIP) of the newborn is a single intestinal perforation typically found on the antimesenteric border of the terminal ileum. (See 'Pathology and pathogenesis' above.)

●Epidemiology – SIP occurs primarily in preterm, very low birth weight (VLBW, birth weight <1500 g) infants, with a reported prevalence of approximately 2 percent. Although prematurity is the only well-established risk factor, other reported risk factors include severe chorioamnionitis, postnatal glucocorticoid therapy, and possibly postnatal indomethacin therapy. (See 'Epidemiology' above.)

●Clinical manifestations – SIP generally presents in VLBW infants within the first 10 days of life as an acute onset of abdominal distension and hypotension. The classical physical finding is a black-bluish discoloration of the abdominal wall. In contrast, abdominal wall erythema, crepitus, and induration commonly seen with necrotizing enterocolitis (NEC) are not usually present in infants with SIP. (See 'Clinical manifestations' above.)

●Diagnosis – The diagnosis for SIP is based upon clinical presentation and, physical findings consistent with SIP in the presence of pneumoperitoneum without other radiographic findings of NEC. (See 'Diagnosis' above.)

●Differential diagnosis: NEC – NEC is the main disease in the differential diagnosis of SIP. Differentiation between the two disorders is usually based upon the clinical manifestations and abdominal radiographic findings, although there is considerable overlap between the two conditions. (See 'Differential diagnosis: Necrotizing enterocolitis' above.)

●Management

•Stabilization – The initial management for a neonate with SIP is directed towards stabilization of the patient. This includes discontinuing feeds and enteral medications, nasogastric suctioning, and supportive care, such as fluid resuscitation, antibiotics, and inotropic medications to correct hypotension. (See 'Initial management' above.)

•Surgical treatment – The treatment for SIP is surgical. We suggest primary peritoneal drainage as the initial treatment for SIP versus  laparotomy with bowel resection with either primary anastomosis or creation of an ostomy (Grade 2B). (See 'Surgical treatment' above.)

●Outcome – Long-term survival of infants with SIP has improved over the past 30 years with reported survival rates of 64 to 90 percent. However, survivors are at risk for neurodevelopmental impairment. (See 'Outcome' above.)