Gasless abdomen has been defined as a state of scanty or invisible intestinal gas on simple abdominal radiography [1]. Until recently, the incidence of gasless abdomen in premature infants has not been well-documented. A recent study reported that 57 (11.5%) of 496 extremely-low-birth-weight infants (ELBWIs) with birth weight less than 1,000 g showed gasless abdomen on radiography, and 12 (21.1%) of them had been confirmed to have spontaneous intestinal perforation (SIP) [1]. However, the incidence of SIP in ELBWIs with gasless abdomen remains unclear. Several other studies have reported that gasless abdomen may commonly present in ELBWIs with SIP (52.0% to 67.0%) [1-6].

SIP is a common complication observed in ELBWIs with birth weight less than 1,000 g, and it has been described as a distinct clinical entity that differs from necrotizing enterocolitis (NEC), in terms of both clinical and histologic presentation [1,4,5]. Unlike NEC that accompanies radiologic hallmarks on simple abdominal radiography, such as pneumatosis intestinalis, or portal vein gas, and fixed bowel loop, SIP does not show distended, fixed bowel loops, pneumatosis intestinalis, or portal venous gas on abdominal radiography, and the most common radiological finding in infants with SIP has been gasless abdomen [1,4,6]. In addition, SIP associated with gasless abdomen in ELBWIs may occur without any signs suggestive of intestinal perforation, such as pneumoperitoneum, which can be found in only 33.3% to 50.0% of cases [4,6]. This low rate of intraperitoneal free air in SIP could be explained by the fact that perforation in SIP occurs as a more gradual process with slow spillage of meconium or bile-stained intraluminal material into the peritoneum rather than as a sudden event caused by rupture of subserosal cavity in NEC [1,4,6]. Moreover, intestinal structures other than the perforation site usually tend to be relatively well preserved [4,6]. Thus, it is often difficult to diagnose SIP in infants with gasless abdomen by simple abdominal radiography in the absence of pneumoperitoneum [2,7].

Given that the delayed treatment of SIP in ELBWIs owing to difficulty in early diagnosis may increase the likelihood of subsequent dismal outcomes, a more careful clinical and radiologic assessment is required if the gasless abdomen is persistently observed in these infants [3]. In this regard, previous studies have reported that abdominal ultrasonography had an important role in the early diagnosis of SIP in ELBWIs with gasless abdomen [1,7].

Thus, in this study, we intended to evaluate the clinical and radiologic findings for early diagnosis of SIP in ELBWIs with persistent gasless abdomen, and to determine the risk factors associated with SIP. We also aimed to evaluate the usefulness of abdominal ultrasonography for the diagnosis of SIP in ELBWIs with gasless abdomen.

This study showed the clinical and radiological findings associated with SIP occurring in ELBWIs with persistent gasless abdomen on simple abdominal radiography. PDA treated with intravenous ibuprofen or indomethacin might be regarded as a significant risk factor associated with SIP in ELBWIs with gasless abdomen. Pneumoperitoneum was found in only one of the eight infants (12.5%) with SIP on simple abdominal radiography; abdominal ultrasonography in the non-SIP group showed ascites or focal fluid collection in three infants; the other findings were mostly unremarkable. Meanwhile, abdominal ultrasonography of infants in the SIP group exclusively showed findings suggestive of intestinal perforation such as intramural echogenicity (75.0%), echogenic extramural material (50.0%) and ascites (50%), with or without bowel wall thickening (50.0%). Thus, this study suggested that abdominal ultrasound might help detect intestinal perforation when SIP is suspected in ELBWIs with gasless abdomen.

Several causes of SIP in extremely premature infants have been suggested. Multifocal partial or complete defects of musculature in the intestinal muscular layer have been frequently observed in the histologic findings of SIP [8-12]. Thus, defects in the muscle layer owing to immaturity in premature infants might be considered as one of the causes of SIP. Additionally, ischemia caused by stress, hypoxia, and/or hypotension before or after birth may lead to disruptions in both musculature and mucosa followed by the regeneration of mucosa, which results in total or partial absence of musculature [13,14]. Thus, SIP occurs typically in the terminal ileum, also known as the “watershed zone,” which is vulnerable to ischemia [15,16]. In this study, similar to the results of previous studies, the most common perforation site was the terminal ileum. However, it is not fully understood whether perforation occurs in other parts of the intestine, such as the colon or proximal ileum, which have more blood supply, as well as in multiple sites.

SIP of the newborn occurs primarily in ELBWIs with an incidence of approximately 3% in this specific population [17-19]. The median age at presentation of SIP ranges from 7 to 15 days [4,15,19-22], which was 11.6±4.4 days in this study. It is well-known that prostaglandin inhibitors such as indomethacin used for the treatment of PDA in premature infants could reduce blood flow to the intestinal tract and induce bowel wall ischemia, subsequently leading to SIP [4,23-25]. Combined with indomethacin for treating PDA, administration of glucocorticoids is commonly reported to be a risk factors for SIP, and exposure to both indomethacin and hydrocortisone in the first week of life might significantly increase the risk of SIP in very-low-birth-weight infants [26]. Moreover, a study reported that the use of ibuprofen could be related to the occurrence of SIP in premature infants [27]. Similarly, this study suggested that PDA treated with intravenous ibuprofen or indomethacin might be a risk factor associated with the occurrence of SIP. However, we were unable to investigate whether hydrocortisone might be a risk factor for SIP, because it was used in only one infant with SIP in this study. Other risk factors reported in previous studies associated with SIP, such as exposure to inotropic agents and surfactants [28], were not significant. Immature gastrointestinal motility and thickening of the meconium would exert mechanical compression on the bowel wall, making the patient susceptible to mucosa damage, ischemia, and perforation [6,13]. In addition, various infections, especially those caused by Candida species, have been reported to be related to SIP [6]. In this study, one case of Enterococcus and one case of GBS were detected in the blood cultures of infants in the SIP group. However, the relationship between sepsis and the occurrence of SIP in ELBWIs with gasless abdomen could not be evaluated in this study because of the small number of cases. Further research in this aspect is warranted.

In general, infants with SIP have no typical symptoms or signs of NEC, such as abdominal distension, abdominal rigidity, tenderness, abdominal wall edema, or erythema [29]. These infants do not meet the traditionally accepted Bell’s criteria for NEC [6]. They are often strikingly stable in the early stage and have no signs of severe illness or peritonitis [5,15]. Their abdominal radiography commonly shows gasless abdomen [6]. Hence, perforation may be difficult to diagnose, particularly in ELBWIs with gasless abdomen, which may delay prompt and adequate surgical intervention [1,2,4,5,29]. Gray or bluish discoloration of the abdominal wall, resulting from peritoneal staining by spillage of meconium into the peritoneal cavity, that it should be regarded as a typical sign of perforation, and this discoloration usually starts in the groin or scrotum [5,16]. In this study, these findings were observed in six of the infants (75%) with SIP, and non-specific findings such as greenish or red gastric residues, abdominal distension, and decreased bowel sounds, were common in infants with SIP.

Previous studies have reported that abdominal ultrasonography may be a very useful tool for diagnosis of SIP [1,2]. In particular, the presence of extraluminal free-floating hyperechogenic materials could suggest the occurrence of intestinal perforation in ELBWIs with gasless or scanty gas abdomen [1]. Extraluminal echogenic materials and intraluminal echogenicity were observed in four (50%) and six infants (75%) with SIP, respectively, in this study. However, these findings should not be interpreted as definite signs of intestinal perforation, because their specificity was reported to be only 89.0% [1]. All infants with the above-mentioned abdominal ultrasonographic findings were operated on, and the presence of SIP was confirmed.

Treatment options for SIP in ELBWIs are primary laparotomy with resection or primary peritoneal drainage [19,30-33]. Although previous studies did not reveal a clear difference in prognosis between infants who underwent peritoneal drainage and those who underwent primary surgical repair [32], there is compelling evidence that infants who underwent peritoneal drainage show inferior neurologic outcomes at 18 to 22 months of age [33]. In this study, three patients underwent peritoneal drainage followed by salvage laparotomy. These three infants had unstable vital signs at the first intervention, and peritoneal drainage was preferred over primary laparotomy with resection of perforated segments. Primary laparotomy was performed in the other five infants. However, despite the surgical treatment and subsequent intensive care, four infants (50.0%) died after operation. It was presumed that three of them died owing to postoperative bleeding resulting from disseminated intravascular coagulation occurring during the surgery. Therefore, early diagnosis and surgery are required.

This study has several limitations. First, this study was conducted retrospectively in a single center, and included only a small number of ELBWIs with persistent gasless abdomen on abdominal radiography. Thus, the exact incidence and prognosis of SIP in ELBWIs with gasless abdomen could not be predicted. Second, because the clinical and radiologic findings suggestive of SIP in ELBWIs in this study would be non-specific and might present in other gastrointestinal disorders, such as NEC, the results indicating that these clinical findings may be associated with intestinal perforation in ELBWIs with gasless abdomen should be interpreted with caution. However, despite the above limitations, the findings of abdominal ultrasonography in this study such as extraluminal echogenic materials and intraluminal echogenicity may be helpful for early detection of a slowly progressive intestinal perforation in ELBWIs with gasless abdomen.

In conclusion, this study showed that SIP could commonly occur in ELBWIs who show persistent gasless abdomen on simple abdominal radiography, and gray or bluish discoloration of the abdomen suggestive of meconium peritonitis was frequently observed in infants with SIP. Moreover, this study showed that abdominal ultrasound could be a highly useful for the diagnosis of SIP.