Preoperative Clinical Factor of Small Intestinal Necrosis Caused by Intestinal Strangulation

Background: Intestinal necrosis induced by strangulation is life-threatening, thus it is important to do operation before leading to intestinal necrosis. The purpose of this study is to find preoperative clinical factor whether intestinal resection is needed in patients of intestinal strangulation. Methods: 184 patients diagnosed with intestinal strangulation during operation were focused from April 2008 to March 2016. Among 184 patients, 98 patients were resected intestine and 86 patients were not resected. Preoperative clinical factors for Resection Group (RG) and Non-Resection Group (NRG) were compared retrospectively. Results: In univariate analysis, patients in RG were older and had higher proportion of female, Systemic Inflammatory Response Syndrome (SIRS), severe American Society of Anesthesiologists (ASA), White Blood Cell (WBC), C-Reactive Protein (CRP), Base Excess (BE), Lactate, and closed loop on computed tomography. In a logistic regression analysis age, sex, BE, Lactate, closed loop was identified significant factors. Each of these variables was assigned a score of 1, except for close loop which was scored 2. The Receiver Operating Characteristic (ROC) curve analysis of the score revealed a great discriminatory ability for the system (AUC: 0.821). Conclusion: We identified an independent risk factor for intestinal resection in patients of intestinal strangulation. The scoring system is very simple and might be planning surgical procedure and postoperative care in patients with intestinal strangulation.


Introduction
Intestinal strangulation is an intestinal obstruction associated with ischemia of the intestinal tract. Previous studies reported that clinical signs, laboratory findings, CT finding, and a combination of these parameters are important factors to distinguish intestinal strangulation or simple small bowel obstruction [1][2][3]. But it is difficult to detect intestinal strangulation at the beginning. Furthermore, the ischemic intestine became necrosis, thus it is an indication for emergency operation because it would become fatal if it develops from perforation to peritonitis [4]. Upregulation of lactate was reported as a good predictive factor for bowel strangulation and associated with intestinal ischemia using mouse model [5,6]. Nonetheless, surgeons often have trouble in distinguishing ischemia of the intestine from an intestinal necrosis before operation. Thus, we conducted this study to find preoperative clinical factors whether the intestinal resection is needed and to find the new prediction model for the intestinal necrosis. patients did not need resection. All patients resected intestine were proved intestinal necrosis at a pathological diagnosis. Data was collected immediately after arriving at the Emergency room or the Surgery outpatient clinic. Recorded variables: age, sex, time between the onset of symptoms and arrival at the first visit, vital signs (systolic and diastolic arterial blood pressures, heart rate, breathing rate, and body temperature), symptoms and physical examination findings, laboratory findings, imaging features, types of management, time between arrival and operation, operative findings, etiology of obstruction, incidence and causes of bowel ischemia, necrosis, and perforation, hospital stay. The associations between the clinical factors and the intestinal necrosis were retrospectively analyzed in the development dataset.

Statistical Analysis
Categorical variables between the groups were analyzed by the Chi-square test. The Mann-Whitney U test was used to test continuous variables. A stepwise regression analysis was performed with factors found to be significantly associated with the intestinal necrosis based on the univariate analysis results. For statistical analyses, we used the JMP (Version 10, SAS Institute) and SAS software programs (Version 10, SAS Institute). A twotailed P value of < 0.05 was considered statistically significant.

Results
During the study period, a total of 184 patients with intestinal strangulation underwent emergency operation. Among these, 98 patients were resected the intestine (resection group; RG), 86 patients were not resected (non-resection group; NRG). The characteristics of study populations are summarized in (Table 1). The average age of the patients was older in RG than in NRG (77.7 years vs 72.2 years p<0.01). Female was more compared with male in RG than in NRG (p<0.01). The surgical history and the time to operation were not associated with the intestinal necrosis. The proportion of patients in SIRS and severe ASA was higher in RG than in NRG (P<0.01, p<0.05, respectively). The examination findings and the surgical findings are shown in (Table 2). WBC, CRP, LDH, and CK in peripheral blood were higher in RG than in NRG (p<0.05, p<0.001, p<0.05, respectively). Base Excess was lower, and Lactate was higher in arterial blood gas (p<0.05, p<0.001, respectively). The proportion of closed loop intestine in CT examination was higher in RG than in NRG. Operation time was longer in RG than in NRG (p<0.001). Univariate analysis and multivariate analysis of preoperative data in RG compared with that in NRG is shown in (Table 3). The significant factors were age, sex, SIRS, ASA, WBC, CRP, BE, Lactate, closed loop in univariate analysis. Furthermore, the significant factors were age, sex, BE, Lactate, closed loop in multivariate analysis. Based on a logistic regression analysis, the following five variables were selected: age (over 71 years), female, BE (under 0.3), Lactate (over 1.8), closed loop. Each of these variables was assigned a score of 1, except for close loop which was scored 2 to obtain a total of 6 according to the odds ratio (Table 4).

Table 4: Risk score for intestinal resection and Scattering in accord with Prediction scoring system. A) The Prediction Scoring (PS) system
for intestinal resection in intestinal strangulation was shown. The score was distributed for each of these variables: age (over 71 years), 1; female, 1; BE (under 0.3), 1; Lactate (over 1.8), 1; closed loop, 2; B) The scattering was shown by total points which each score was added up.
Scattering in accord with the prediction scoring system is shown Table4, and the patients with higher points in this system tend to be resected the intestine. The prediction scoring system for intestinal resection in intestinal strangulation was demonstrated based on a stepwise regression analysis. The Receiver Operating Characteristic (ROC) curve analysis revealed a great discriminatory ability for the system (AUC: 0.821) (Figure1).

Discussion
Intestinal strangulation is defined as ischemia of intestinal tract due to SBO. It was regarded as an indication for an emergency surgical treatment. If the ischemic intestine was left without operation, the intestine became necrosis, leading to perforation and peritonitis [4]. Therefore, it is important to diagnose it as soon as possible before intestinal necrosis. Various clinical data vary if the intestine occurred intestinal strangulation. LDH, CK, and lactate elevates due to hypoperfusion of the intestinal tissue after progression to ischemia [7]. Especially, increase of lactate was reported as a good predictive factor for bowel strangulation and associated with intestinal ischemia using mouse model [5,6]. On the other hand, there were not different WBC and CRP levels between patients with conservative management and those with surgical treatment, thus WBC and CRP may not be useful to detect bowel ischemia [8-10]. Past studies reported the utility of CT to make a diagnosis of small bowel obstruction [11,12]. In contrast it is difficult to find small bowel ischemia by CT with sensitivities ranging from 75% to 100% and specificities from 61% to 93% [13,14]. It is insufficient to make a diagnosis with intestinal necrosis by single examination, and we should use a combination of various examination which is simple and minimal as much as possible. The current study demonstrated that the various clinical factors were related with intestinal necrosis in intestinal strangulation, and using five factors (age, sex, BE, Lactate, and closed loop) the new prediction scoring system for the intestinal necrosis was constructed. The ROC curve analysis revealed a great discriminatory ability for the system (AUC: 0.821). The system is very simple because of using just data of arterial blood gas and CT, and reliable to detect intestinal necrosis in patients of intestinal strangulation. There are some problems associated with this study. This study was done by single institution and a retrospective analysis, therefore it is necessary to validate the results at multiple institutions in prospective manner.

Conclusion
The current study suggests that the system may help to diagnose an intestinal necrosis.