Cases of NOMI are increasingly common with the progressive aging of societies during the last decades [17]. Despite a greater awareness of the life-threatening disorder, the diagnosis and treatment of NOMI remain challenging. Explorative laparotomies for the assessment of bowel ischemia, without the performance of any therapeutic procedures, often have been indicated. If the prognostic criteria for the conservative treatment of NOMI patients were to be clarified, the knowledge would be helpful in determining a treatment plan and avoiding unnecessary laparotomy. This study is the first attempt to evaluate the predictive factors which affect the outcomes of non-operative therapy for NOMI. It is noteworthy that 31% of the patients in our study group survived to discharge without surgical intervention, this group included a substantial number of patients ineligible for surgical treatment because of severe comorbidities or poor general condition. Some biological measures of tissue ischemia, cell lysis and inflammation (LDH or CRP), with possibly effective cutoff values calculated, were shown to be promising predictors of survival in conservative therapy for NOMI. Importantly, the index of general condition at the onset of NOMI, represented by the SOFA score (≤ 3 points), was found to be the most reliable factor. The serum levels of albumin and T-Bil were also associated with prognosis and seem mainly to represent the general condition of patients.

The key underlying mechanism for the development of NOMI can be explained by an excessive physiological response to maintain perfusion of vital organs at the expense of mesenteric perfusion, resulting in persistent splanchnic vasoconstriction [3]. According to this hypothesis, the most important treatment for NOMI is to limit the duration and severity of systemic circulatory failure. Fluid resuscitation, with care to avoid overload, is a critical component of the initial care [4]. Administration of vasodilators may be effective for relieving persistent vasoconstriction and improving the prognosis, when the hemodynamics of the systemic circulation are restored, as shown in a previous study [12]. Nonetheless, most of the patients in our series who survived, except for one, did not undergo treatment with a vasodilator.

We should note the heterogeneity in diagnostic methods and criteria for NOMI among previous studies. The diagnostic value of CT has been questioned, and mesenteric angiography or surgical exploration has been considered to be necessary for reliable diagnosis and classification of NOMI [2]. However, recently, the diagnostic precision of CT has been improved with the development of MDCT, in which the resolution of the multi-planar reconstructed images is comparable to that of angiography [18]. In the current study, the diagnosis of NOMI was based on the findings of contrast-enhanced MDCT, such as attenuated bowel wall enhancement, pneumatosis intestinalis and portal venous gas. These findings have been shown to be highly specific to bowel ischemia, despite the sensitivity of detection having been unsatisfactory [17, 19]. Furthermore, the patients with the diagnosis of “pneumatosis cystoides intestinalis” associated with benign prognosis [13, 14], which may include some cases of mild bowel ischemia, were excluded from this study. Therefore, the diagnostic accuracy of NOMI is considered to be reliable in our patient group, although there is the possibility that patients with subclinical or undetected ischemia are missing.

The presence of pneumatosis intestinalis and portal venous gas has been deemed to be definitive findings for severe bowel ischemia, and patients with these signs need urgent explorative laparotomy [9, 20]. Nonetheless, pneumatosis intestinalis is leakage of gas within the bowel wall due to mucosal injury, and portal venous gas is the progression of the pneumatosis intestinalis to the portal venous system [19]. These findings do not necessarily indicate irreversible and transmural necrosis of the bowel. Interestingly, a substantial proportion of patients with these radiological findings survived without surgical intervention in this study.

Our results suggest that NOMI patients with a low SOFA score could be indicated for conservative management. The patients, when supported with non-operative treatment, should be monitored closely by means of serial abdominal examinations, as well as evaluations of their general condition with blood tests. A timely follow-up CT scan may be needed for precise evaluation. Surgical treatment should be considered without delay if the ischemic damage progresses.

The present study had some limitations, it being a retrospective study in a single institution. Although the patients with definitive radiological findings of NOMI according to the uniform diagnostic criteria were selected in our study group, the indication or reason for conservative treatment varied in each individual. In addition, the sample size was too small to perform multivariate analysis. Thus, the possibility cannot be excluded that some confounders may have affected our results. Further studies with larger cohorts are necessary to confirm the clinical relevance of the prognostic factors for the conservative treatment of NOMI.