Acute intestinal obstruction (AIO) is a common general surgical emergency and accounts for a considerable share of emergency abdominal admissions and operations worldwide.[1,2,3] It is defined as an arrest of the forward passage of intestinal contents, which may be mechanical (adhesions, hernias, volvulus, tumours, strictures, intussusception) or functional (paralytic ileus, pseudo-obstruction).[1,4] The spectrum of causes differs from region to region. In Western series postoperative adhesions and malignancy predominate, whereas in India external hernias, adhesions and bands, sigmoid and small bowel volvulus, and tuberculous strictures continue to contribute a large proportion of cases.[5,6,7,8] The central pathological event that turns a simple obstruction into a surgical catastrophe is strangulation. Luminal distension raises intramural pressure, impairs venous return and then arterial inflow, and leads in turn to mucosal ischaemia, transmural infarction, bacterial translocation, perforation and peritonitis.[1,9] Once gangrene has developed, resection of non-viable bowel becomes unavoidable, the operation is longer and more hazardous, a primary anastomosis is often unsafe, and the risk of septic shock, multi-organ failure and death rises steeply.[10,11] Delay in the diagnosis of strangulation has repeatedly been shown to be one of the most important modifiable determinants of outcome.[12,13] The surgeon's dilemma is that bowel viability cannot be assessed directly before laparotomy. Non-operative management is appropriate for selected patients with partial adhesive obstruction,[14,15] but a policy of waiting carries the risk of missing strangulation, and early operation has been associated with better survival in adhesive obstruction.[16] Classical clinical signs such as continuous pain, fever, tachycardia and peritonism are neither sensitive nor specific, and even experienced clinicians recognise strangulation correctly in only a modest proportion of patients.[17,18] Several multivariable models and clinicoradiological scores have been proposed to improve this,[19,20,21] CT signs such as reduced bowel-wall enhancement have good specificity,[22,23] and serological markers have been studied extensively,[24,25] but none has been universally adopted. Most of the available evidence comes from high-resource settings and from adhesive small bowel obstruction. The Indian scenario is different. Patients frequently reach tertiary centres after prolonged symptoms, often after referral from primary or rural facilities; neglected hernias and sigmoid volvulus are common; and round-the-clock CT, lactate estimation and blood-gas analysis may not be uniformly available.[5,6,26] There is therefore a need for locally generated, prospective data on which simple, widely available parameters predict bowel gangrene in AIO. The present study was designed to address this gap. Aim To evaluate the predictive factors for bowel gangrene in patients with acute intestinal obstruction at a tertiary care teaching hospital. Objectives To determine the proportion of patients with acute intestinal obstruction who have bowel gangrene at surgery. To compare demographic, clinical, laboratory and radiological features between patients with and without bowel gangrene. To identify independent predictors of bowel gangrene and to assess their diagnostic accuracy. To compare postoperative morbidity, mortality and hospital stay between the two groups.

Study design and setting This was a hospital-based prospective observational study conducted in the Department of General Surgery of a tertiary care teaching hospital in India. The study ran for 24 months, from July 2023 to June 2025. Approval was obtained from the Institutional Ethics Committee before enrolment, and written informed consent (or consent from the legal guardian when the patient was too unwell) was taken from all participants. The study was conducted in accordance with the Declaration of Helsinki and the ICMR National Ethical Guidelines for Biomedical and Health Research Involving Human Participants. Sample size Assuming an anticipated prevalence of bowel gangrene of 25% among patients with AIO, an absolute precision of 8% and a 95% confidence level, the minimum sample size was calculated with the formula n = Z²P(1-P)/d², which gave 113 patients. The sample was rounded up to 120 consecutive patients. Selection criteria Inclusion criteria: • Patients aged 18 years or above presenting to the emergency department with clinical and radiological evidence of acute mechanical intestinal obstruction. • Patients who underwent emergency exploratory laparotomy during the same admission. Exclusion criteria: • Patients whose obstruction resolved with conservative management and who did not require surgery, because bowel viability could not be confirmed. • Paralytic ileus, pseudo-obstruction and obstruction secondary to perforation peritonitis or acute pancreatitis. • Intestinal obstruction in pregnancy, and patients with mesenteric vascular occlusion as the primary diagnosis. • Patients unwilling to participate, and those with incomplete records. Data collection After initial resuscitation, a pre-designed and pre-tested proforma was completed for each patient. It recorded age, sex, comorbidities, previous abdominal surgery, duration of symptoms (hours from onset of pain to presentation), the nature of pain (colicky or continuous), vomiting, constipation, distension, and vital signs. Abdominal examination noted localised tenderness, guarding or rebound tenderness (grouped as peritonism), bowel sounds, and hernial orifices. All patients underwent blood investigations on admission: haemoglobin, total leukocyte count (TLC), renal function, serum electrolytes, C-reactive protein (CRP), and arterial blood gas analysis with lactate. Plain abdominal radiographs and abdominal ultrasonography were performed in every patient. Contrast-enhanced CT (CECT) of the abdomen was performed when the patient's condition and renal function permitted and the diagnosis was in doubt (68 patients). Definitions Bowel gangrene was diagnosed at laparotomy when the bowel segment was black, green or dusky, lustreless, and non-peristaltic, with absent mesenteric arterial pulsation and no recovery after 10-15 minutes of warm saline packs (with or without inhalational oxygen). The resected segment was sent for histopathology, which confirmed transmural necrosis. Tachycardia was a pulse rate above 100/min, hypotension a systolic pressure below 100 mmHg, and fever a temperature above 37.8 °C. Leukocytosis was TLC above 12,000/mm³, hyperlactataemia a serum lactate above 2 mmol/L, and metabolic acidosis an arterial pH below 7.35. The ability to complete the primary outcome did not depend on any single investigation. Management protocol All patients received nasogastric decompression, intravenous fluids, electrolyte correction, broad-spectrum antibiotics and analgesia. Surgery was performed under general anaesthesia by or under the supervision of a consultant surgeon. The operative findings (cause, site and length of gangrenous bowel, peritoneal contamination) and the procedure performed (adhesiolysis, hernia repair, detorsion, resection with anastomosis, stoma formation) were recorded. Patients were followed until discharge or death, and postoperative complications were noted up to 30 days. Statistical analysis Data were entered in Microsoft Excel and analysed with SPSS version 26.0 (IBM Corp., Armonk, NY). Continuous variables were expressed as mean ± standard deviation and compared with the independent-samples t-test; categorical variables were expressed as numbers and percentages and compared with the chi-square or Fisher's exact test. Variables with p<0.05 on univariate analysis were entered into a binary logistic regression model (backward likelihood-ratio method), and results were expressed as adjusted odds ratios (aOR) with 95% confidence intervals (CI).[27] Model fit was tested by the Hosmer-Lemeshow test. ROC curves were constructed to find the area under the curve (AUC), optimal cut-offs, sensitivity, specificity, and positive and negative predictive values (PPV, NPV). A two-sided p-value <0.05 was considered statistically significant.

During the 24-month period, 120 patients fulfilled the criteria and were analysed. Bowel gangrene was found at laparotomy in 30 patients (25.0%; 95% CI 17.5-33.8%), and the remaining 90 patients (75.0%) had viable bowel. Demographic profile The age of the patients ranged from 18 to 86 years (mean 52.4 ± 16.8 years). Patients with gangrene were significantly older than those without (58.6 ± 14.2 vs 50.3 ± 17.1 years; p=0.011), and the proportion with gangrene rose steadily with age, from 11.1% in those aged up to 30 years to 44.4% in those above 70 years (Table 1). There were 78 males (65.0%) and 42 females (35.0%), a male-to-female ratio of 1.9:1, and sex did not differ between the groups. Comorbidity (diabetes, hypertension, ischaemic heart disease, chronic kidney disease) was present in 38.3% and previous abdominal surgery in 46.7%, with no significant difference between the groups. The mean duration of symptoms was significantly longer in the gangrene group (62.4 ± 22.8 vs 38.6 ± 20.4 hours; p<0.001), and 73.3% of patients with gangrene presented after 48 hours compared with 27.8% of those without (Table 2). Table 1: Age distribution and rate of bowel gangrene Age group (years) Total patients, n (%) Gangrene, n Non-gangrene, n Gangrene rate (%) ≤30 18 (15.0) 2 16 11.1 31-40 16 (13.3) 3 13 18.8 41-50 20 (16.7) 3 17 15.0 51-60 22 (18.3) 5 17 22.7 61-70 26 (21.7) 9 17 34.6 >70 18 (15.0) 8 10 44.4 Total 120 (100) 30 90 25.0 Chi-square test for trend, p=0.006. Table 2: Baseline characteristics of patients with and without bowel gangrene Variable Total (n=120) Gangrene (n=30) Non-gangrene (n=90) p-value Age, years (mean ± SD) 52.4 ± 16.8 58.6 ± 14.2 50.3 ± 17.1 0.011 Age >60 years, n (%) 44 (36.7) 17 (56.7) 27 (30.0) 0.009 Male sex, n (%) 78 (65.0) 19 (63.3) 59 (65.6) 0.82 Any comorbidity, n (%) 46 (38.3) 15 (50.0) 31 (34.4) 0.13 Previous abdominal surgery, n (%) 56 (46.7) 12 (40.0) 44 (48.9) 0.40 Duration of symptoms, hours (mean ± SD) 44.6 ± 23.4 62.4 ± 22.8 38.6 ± 20.4 <0.001 Duration >48 hours, n (%) 47 (39.2) 22 (73.3) 25 (27.8) <0.001 SD: standard deviation. Chi-square/Fisher's exact test for categorical and independent t-test for continuous variables. Aetiology of obstruction Adhesions and bands were the leading cause of obstruction (36.7%), followed by obstructed or strangulated external hernia (26.7%), malignancy (12.5%) and volvulus (11.7%), as shown in Table 3. Gangrene was most frequent with volvulus (50.0%) and hernia (34.4%), and least frequent with malignancy (13.3%); none of the five patients with ileocaecal tuberculous stricture had gangrene (Figure 1). Hernias comprised 14 inguinal, 8 femoral, 6 umbilical or paraumbilical and 4 incisional hernias. Table 3: Aetiology of acute intestinal obstruction and associated bowel gangrene Cause of obstruction Patients, n (%) Gangrene, n Gangrene rate (%) Adhesions / bands 44 (36.7) 8 18.2 Obstructed / strangulated hernia 32 (26.7) 11 34.4 Malignancy (colorectal, small bowel) 15 (12.5) 2 13.3 Volvulus (sigmoid 9, small bowel 5) 14 (11.7) 7 50.0 Intussusception 6 (5.0) 1 16.7 Ileocaecal tuberculosis / stricture 5 (4.2) 0 0.0 Others (internal hernia, Meckel's band) 4 (3.3) 1 25.0 Total 120 (100) 30 25.0 Percentages in the last column are row percentages. Clinical features Abdominal distension and vomiting were almost universal and did not discriminate between groups. In contrast, tachycardia, fever, hypotension, continuous (rather than colicky) pain, localised tenderness with guarding or rebound, and absent bowel sounds were all significantly more common in patients with gangrene (Table 4). Localised tenderness with guarding or rebound had the strongest association (unadjusted OR 14.1; 95% CI 5.2-38.2). Even so, 7 of the 30 patients with gangrene (23.3%) had no peritonism and 6 (20.0%) were afebrile and had a pulse rate below 100/min, which illustrates the limitation of clinical examination alone. Table 4: Clinical features in relation to bowel gangrene Feature Total (n=120) Gangrene (n=30) Non-gangrene (n=90) p-value Abdominal distension 101 (84.2) 27 (90.0) 74 (82.2) 0.31 Vomiting 108 (90.0) 28 (93.3) 80 (88.9) 0.73 Continuous pain 41 (34.2) 21 (70.0) 20 (22.2) <0.001 Pulse >100/min 62 (51.7) 24 (80.0) 38 (42.2) <0.001 Temperature >37.8 °C 25 (20.8) 13 (43.3) 12 (13.3) <0.001 Systolic BP <100 mmHg 16 (13.3) 10 (33.3) 6 (6.7) <0.001 Localised tenderness / guarding / rebound 40 (33.3) 23 (76.7) 17 (18.9) <0.001 Absent bowel sounds 28 (23.3) 14 (46.7) 14 (15.6) <0.001 Values are n (%). BP: blood pressure. Laboratory parameters Patients with gangrene had a significantly higher TLC, CRP, serum lactate and serum creatinine, and a greater frequency of metabolic acidosis, than those with viable bowel (Table 5). The mean serum lactate was more than twice as high in the gangrene group (3.9 ± 1.6 vs 1.8 ± 0.9 mmol/L; p<0.001). Haemoglobin did not differ between the groups Table 5: Laboratory findings in relation to bowel gangrene Parameter Gangrene (n=30) Non-gangrene (n=90) p-value Haemoglobin, g/dL (mean ± SD) 12.1 ± 1.9 12.4 ± 2.0 0.46 TLC, ×10³/mm³ (mean ± SD) 15.8 ± 4.6 10.9 ± 3.9 <0.001 TLC >12,000/mm³, n (%) 21 (70.0) 22 (24.4) <0.001 CRP, mg/L (mean ± SD) 118.4 ± 46.2 52.6 ± 38.4 <0.001 Serum lactate, mmol/L (mean ± SD) 3.9 ± 1.6 1.8 ± 0.9 <0.001 Serum lactate >2 mmol/L, n (%) 22 (73.3) 17 (18.9) <0.001 Arterial pH <7.35, n (%) 12 (40.0) 9 (10.0) <0.001 Serum creatinine, mg/dL (mean ± SD) 1.5 ± 0.7 1.1 ± 0.5 0.005 TLC: total leukocyte count; CRP: C-reactive protein. Radiological findings Plain radiographs showed dilated bowel loops with multiple air-fluid levels in all patients, and no patient had free gas under the diaphragm. On ultrasonography and CECT, free peritoneal fluid, bowel-wall thickening and a closed-loop or whirl configuration were significantly more frequent in the gangrene group (Table 6). Reduced or absent bowel-wall enhancement or pneumatosis on CECT was seen in 8 of 22 patients with gangrene who underwent CT (36.4%) but in only 1 of 46 without gangrene (2.2%); this sign was highly specific (97.8%) though insensitive. Table 6: Imaging findings (ultrasonography / CECT) in relation to bowel gangrene Finding Gangrene (n=30) Non-gangrene (n=90) p-value Free peritoneal fluid 21 (70.0) 24 (26.7) <0.001 Bowel-wall thickening >3 mm 18 (60.0) 19 (21.1) <0.001 Closed-loop / whirl sign 9 (30.0) 6 (6.7) 0.001 Reduced wall enhancement / pneumatosis (CECT subset)* 8/22 (36.4) 1/46 (2.2) <0.001 *Denominators are patients who underwent CECT (n=68). Values are n (%). Univariate and multivariate analysis Fourteen variables were significant on univariate analysis (Table 7). When entered into the logistic regression model, six remained independent predictors of bowel gangrene: duration of symptoms >48 hours, tachycardia, localised tenderness with guarding or rebound, leukocytosis, serum lactate >2 mmol/L and free peritoneal fluid on imaging (Table 8). Serum lactate had the largest adjusted odds ratio (6.8). The model fitted the data well (Hosmer-Lemeshow p=0.62; Nagelkerke R² = 0.61) and classified 88.3% of patients correctly. Age, fever, hypotension and acidosis lost significance after adjustment, probably because of their association with the other variables. Table 7: Univariate analysis of predictors of bowel gangrene Variable Unadjusted OR 95% CI p-value Age >60 years 3.05 1.30-7.14 0.009 Duration >48 hours 7.15 2.82-18.1 <0.001 Pulse >100/min 5.47 2.04-14.7 <0.001 Temperature >37.8 °C 4.97 1.93-12.8 <0.001 Systolic BP <100 mmHg 7.00 2.28-21.5 <0.001 Continuous pain 8.17 3.24-20.6 <0.001 Localised tenderness / guarding / rebound 14.1 5.2-38.2 <0.001 Absent bowel sounds 4.75 1.90-11.9 <0.001 TLC >12,000/mm³ 7.21 2.88-18.0 <0.001 Serum lactate >2 mmol/L 11.8 4.49-31.0 <0.001 Arterial pH <7.35 6.00 2.20-16.4 <0.001 Free peritoneal fluid 6.42 2.58-15.9 <0.001 Bowel-wall thickening 5.61 2.31-13.6 <0.001 Closed-loop / whirl sign 6.00 1.92-18.7 0.001 OR: odds ratio; CI: confidence interval. Table 8: Multivariate logistic regression: independent predictors of bowel gangrene Predictor Adjusted OR 95% CI p-value Score points Duration of symptoms >48 hours 4.6 1.5-14.1 0.008 1 Pulse >100/min 3.9 1.2-12.7 0.024 1 Localised tenderness / guarding / rebound 5.2 1.7-16.0 0.004 1 TLC >12,000/mm³ 3.4 1.1-10.6 0.034 1 Serum lactate >2 mmol/L 6.8 2.2-21.0 0.001 1 Free peritoneal fluid on imaging 3.2 1.0-9.9 0.048 1 Hosmer-Lemeshow p=0.62; Nagelkerke R² = 0.61; overall correct classification 88.3%. Diagnostic accuracy and a simple predictive score On ROC analysis, serum lactate (AUC 0.87), CRP (AUC 0.84) and TLC (AUC 0.80) each had good discriminatory power (Table 9, Figure 2). A simple score was then created by assigning one point to each of the six independent predictors (range 0-6). The AUC of the combined score was 0.93 (95% CI 0.88-0.98). A score of 3 or more had a sensitivity of 90.0%, specificity of 82.2%, PPV of 62.8% and NPV of 96.1%. The rate of gangrene increased progressively with the score, from 0% at a score of 0 to 100% at a score of 6 (Table 10). Only 3 of the 77 patients (3.9%) with a score of 2 or less had gangrene. Table 9: Diagnostic accuracy of individual parameters and the combined score Parameter AUC (95% CI) Cut-off Sens. (%) Spec. (%) PPV (%) NPV (%) Serum lactate 0.87 (0.79-0.94) >2.0 mmol/L 73.3 81.1 56.4 90.1 CRP 0.84 (0.76-0.92) >80 mg/L 76.7 78.9 54.8 91.0 TLC 0.80 (0.70-0.89) >12,000/mm³ 70.0 75.6 48.8 88.3 Combined score 0.93 (0.88-0.98) ≥3 points 90.0 82.2 62.8 96.1 AUC: area under the ROC curve; Sens.: sensitivity; Spec.: specificity; PPV/NPV: positive/negative predictive value. Table 10: Distribution of the predictive score and observed bowel gangrene Score Patients, n Gangrene, n Non-gangrene, n Gangrene rate (%) 0 24 0 24 0.0 1 26 1 25 3.8 2 27 2 25 7.4 3 17 7 10 41.2 4 13 9 4 69.2 5 9 7 2 77.8 6 4 4 0 100.0 Total 120 30 90 25.0 Cut-off of ≥3 gives sensitivity 90.0% and NPV 96.1%. Operative findings and procedures Gangrene involved the small bowel in 25 patients (83.3%; ileum 20, jejunum 5) and the large bowel in 5 patients (16.7%; sigmoid colon 4, caecum 1). The mean length of gangrenous bowel resected was 46.8 ± 21.4 cm. Resection with primary anastomosis was done in 18 patients with gangrene, while 12 required a stoma (ileostomy 7; Hartmann's procedure or end colostomy 5). In the non-gangrene group, adhesiolysis or band division was done in 39 patients, hernia repair without resection in 21, detorsion in 7, resection with anastomosis (for tumour, stricture or intussusception) in 16, and a diverting stoma or bypass in 7. Postoperative outcome Postoperative outcomes were significantly worse in patients with gangrene (Table 11). Overall mortality was 9.2% (11 patients): 8 of 30 (26.7%) in the gangrene group versus 3 of 90 (3.3%) in the non-gangrene group (OR 10.5; p<0.001). Septic shock with multi-organ dysfunction was the leading cause of death (7 patients), followed by cardiopulmonary failure (4 patients). Patients with gangrene also needed intensive care more often and stayed in hospital for almost twice as long. Table 11: Postoperative outcomes in relation to bowel gangrene Outcome Gangrene (n=30) Non-gangrene (n=90) p-value Mortality, n (%) 8 (26.7) 3 (3.3) <0.001 ICU admission, n (%) 18 (60.0) 12 (13.3) <0.001 Surgical site infection, n (%) 14 (46.7) 17 (18.9) 0.003 Pulmonary complications, n (%) 9 (30.0) 8 (8.9) 0.003 Septic shock / MODS, n (%) 10 (33.3) 4 (4.4) <0.001 Burst abdomen, n (%) 4 (13.3) 3 (3.3) 0.06 Anastomotic leak, n/N (%) 3/18 (16.7) 1/16 (6.3) 0.60 Hospital stay, days (mean ± SD) 15.8 ± 6.2 8.4 ± 3.1 <0.001 ICU: intensive care unit; MODS: multiple organ dysfunction syndrome. Leak rates are among patients who had a primary anastomosis.

Bowel gangrene is the most feared complication of acute intestinal obstruction, and the ability to anticipate it before laparotomy is of direct practical value. In this prospective study of 120 patients operated for AIO, one in four had non-viable bowel. This is consistent with the wide range of 10-30% reported for strangulation in surgical series of obstruction,[4,10,28] and with Indian data, in which delayed presentation and the high burden of external hernia and volvulus keep the rate of strangulation high.[5,6] The principal finding is that six easily obtainable parameters (duration of symptoms, pulse rate, peritonism, TLC, serum lactate and free peritoneal fluid) independently predicted gangrene, and that a simple additive score derived from them could identify high-risk patients with good accuracy. The mean age of our patients was 52.4 years with a male preponderance, similar to Indian experience.[5] Patients with gangrene were older, and the gangrene rate rose from 11% below 30 years to 44% above 70 years. Older patients tolerate vascular compromise poorly, often have atherosclerotic mesenteric disease and delay seeking care, and advanced age is also a recognised risk factor for adverse outcome after surgery for obstruction.[11,28] Age lost its independent significance in the multivariate model, which suggests that its effect operates through duration, haemodynamic and inflammatory changes. The duration of symptoms was the strongest non-laboratory predictor. Gangrene was present in 46.8% (22 of 47) of patients who presented after 48 hours but in only 11.0% (8 of 73) of those who presented earlier. The time-dependent rise in the risk of resection has been documented in complete small bowel obstruction,[13] and early diagnosis was emphasised decades ago as the key to avoiding strangulation.[12] In India, this delay is often created by travel distances, treatment from unqualified practitioners, and fear of surgery, and the finding reinforces the case for early referral and for a low threshold for operation. Adhesions and bands were the commonest cause in our series, followed closely by hernia, which agrees with a large Eastern Indian series,[5] though in much of rural India hernias still lead. The risk of gangrene, however, depended on the cause rather than its frequency. Volvulus (50.0%) and hernia (34.4%) were the highest-risk causes. A closed loop with a tight neck, as in hernias and volvulus, produces early arterial and venous compromise, and a hernia has been named the most common cause of strangulation in patients with small bowel obstruction.[29] In sigmoid volvulus, gangrene is a major determinant of mortality and of the choice between resection with anastomosis and a Hartmann-type procedure.[6,26] Malignant and tuberculous obstructions, which tend to cause gradual luminal narrowing, were rarely associated with gangrene, as expected for chronic, partial obstructions.[7,8] Surgeons should therefore regard any obstructed external hernia and any case of volvulus as strangulated until proved otherwise. Tachycardia, fever, hypotension, continuous pain, peritonism and absent bowel sounds were significantly associated with gangrene on univariate analysis, but only tachycardia and peritonism were independent. This is a familiar observation. Classic signs of strangulation are late features; in a prospective evaluation, preoperative clinical recognition of strangulation was unreliable, with many cases being missed,[17] and the discriminating value of history and examination for acute small bowel obstruction is limited.[18] In our study, nearly a quarter of patients with gangrene had no peritonism and one in five was afebrile with a normal pulse. The independent role of tachycardia reflects the systemic response to fluid sequestration, third-space loss and endotoxaemia, while localised tenderness with guarding reflects transmural ischaemia and irritation of the parietal peritoneum. Because these signs are masked by analgesics, ageing and obesity, they should be interpreted together with laboratory and imaging data. Leukocytosis was an independent predictor in our study, in keeping with the inflammatory response to transmural ischaemia and translocation of bacteria.[9,19] CRP was also higher in the gangrene group and had a good AUC of 0.84, though it was not retained in the multivariate model because of collinearity with TLC. Serum lactate showed the highest adjusted odds ratio (6.8) and the highest AUC among single tests (0.87). Anaerobic metabolism of ischaemic bowel and reduced hepatic clearance in hypovolaemia raise plasma lactate, and raised lactate has been linked with intestinal ischaemia and with the need for operation in the acute abdomen.[24,30] However, systematic reviews have concluded that no serological marker, including lactate, is accurate enough to be used alone to exclude or confirm ischaemia, mainly because of poor sensitivity when the injury is early or localised.[25,31] Our own sensitivity for lactate of 73.3% supports this view. Lactate should therefore be treated as one component of a composite assessment and not as a stand-alone test; its value is highest when it is combined with clinical and imaging findings, as we did in the score. Ultrasonography is cheap and available in nearly every Indian hospital and was useful for demonstrating free fluid and wall thickening, both of which were more common with gangrene, and free fluid was an independent predictor. CT has been more extensively studied: free mesenteric or peritoneal fluid, mesenteric congestion, wall thickening, closed-loop configuration and, above all, reduced wall enhancement are signs of strangulation, with the last having high specificity but only moderate sensitivity.[22,23,32,33] Combined CT signs improve the prediction of strangulation in adhesive obstruction.[34] Our CT findings were similar: reduced enhancement or pneumatosis had a specificity of 97.8% but missed almost two-thirds of cases. Because CECT was performed in only 57% of our patients, it was not included in the regression model, and a normal CT should never override a worrying clinical picture. Several multivariable models have previously been proposed to predict strangulation or the need for surgery in small bowel obstruction, using clinical and CT variables.[19,20,21] Most were developed in high-income settings and rely on CT. The six-variable score proposed here uses a history item, two bedside signs, two blood tests and an ultrasound finding, and can be applied in district hospitals and in teaching hospitals without mandatory CT. A score of 3 or more identified 90% of patients with gangrene, while a score of 2 or less made gangrene unlikely (NPV 96.1%). Practically, patients scoring 3 or more should be prepared for early laparotomy, with intensive resuscitation, a senior surgeon, ICU bed and a stoma plan, whereas those scoring 0-2 and having partial obstruction may be considered for a short, closely monitored trial of non-operative management, as recommended by current guidelines.[14,15,35] Because the score was derived and tested in the same cohort, it will be over-optimistic and needs external validation. Gangrene was associated with a more than eight-fold higher mortality (26.7% vs 3.3%), more frequent infective and pulmonary complications, a higher need for intensive care and a doubling of hospital stay. These figures are in line with earlier reports in which strangulation and bowel resection were the strongest determinants of death after surgery for obstruction.[10,11,28] The case-fatality of 26.7% in our gangrene group is high, and probably reflects late presentation, a high proportion of elderly patients and limited physiological reserve, as well as the burden of faecal contamination. Early operation has been shown to be associated with a survival advantage in adhesive obstruction,[16] and our results strengthen the argument that prompt recognition of high-risk patients is the most effective way to lower mortality in this group. Strengths and limitations The strengths of this study are its prospective design, consecutive enrolment, predefined variables, histopathological confirmation of gangrene, and the use of multivariate and ROC analysis. It also reflects the real presentation pattern in an Indian tertiary centre. Its limitations include a single-centre design with a modest sample size, which gives wide confidence intervals for some odds ratios (for example, free fluid and TLC); the exclusion of patients who recovered without surgery, which may overestimate the prevalence of gangrene; the subjective element in assessing bowel viability at surgery; the selective use of CECT; and the absence of external validation of the score. Multicentre studies with larger samples are needed.

One-quarter of patients operated for acute intestinal obstruction in this study had bowel gangrene, and the condition carried a nearly eight- to ten-fold higher risk of death. Symptom duration of more than 48 hours, tachycardia, localised tenderness with guarding, leukocytosis, raised serum lactate and free peritoneal fluid were independent predictors, and a simple six-point score based on them showed excellent discrimination. Obstructed hernias and volvulus deserve particular vigilance. The use of such a score, combined with early resuscitation and timely surgery, can help the surgeon in everyday practice to avoid delay in high-risk patients and to reduce the burden of strangulation-related morbidity and mortality. Large multicentre studies are needed to validate it.

1. Williams NS, O'Connell PR, McCaskie AW, editors. Bailey & Love's Short Practice of Surgery. 27th ed. Boca Raton: CRC Press; 2018. Chapter 69, Intestinal obstruction; p. 1195-1215. 2. Brunicardi FC, Andersen DK, Billiar TR, et al., editors. Schwartz's Principles of Surgery. 11th ed. New York: McGraw-Hill Education; 2019. Chapter 28, Small intestine. 3. Foster NM, McGory ML, Zingmond DS, Ko CY. Small bowel obstruction: a population-based appraisal. J Am Coll Surg. 2006;203(2):170-176. 4. Markogiannakis H, Messaris E, Dardamanis D, et al. Acute mechanical bowel obstruction: clinical presentation, etiology, management and outcome. World J Gastroenterol. 2007;13(3):432-437. 5. Adhikari S, Hossein MZ, Das A, Mitra N, Ray U. Etiology and outcome of acute intestinal obstruction: a review of 367 patients in Eastern India. Saudi J Gastroenterol. 2010;16(4):285-287. 6. Raveenthiran V, Madiba TE, Atamanalp SS, De U. Volvulus of the sigmoid colon. Colorectal Dis. 2010;12(7):e1-e17. 7. Sharma MP, Bhatia V. Abdominal tuberculosis. Indian J Med Res. 2004;120(4):305-315. 8. Tandon HD, Prakash A. Pathology of intestinal tuberculosis and its distinction from Crohn's disease. Gut. 1972;13(4):260-269. 9. Mucha P Jr. Small intestinal obstruction. Surg Clin North Am. 1987;67(3):597-620. 10. Shatila AH, Chamberlain BE, Webb WR. Current status of diagnosis and management of strangulation obstruction of the small bowel. Am J Surg. 1976;132(3):299-303. 11. Fevang BT, Fevang J, Stangeland L, Soreide O, Svanes K, Viste A. Complications and death after surgical treatment of small bowel obstruction: a 35-year institutional experience. Ann Surg. 2000;231(4):529-537. 12. Cheadle WG, Garr EE, Richardson JD. The importance of early diagnosis of small bowel obstruction. Am Surg. 1988;54(9):565-569. 13. Bickell NA, Federman AD, Aufses AH Jr. Influence of time on risk of bowel resection in complete small bowel obstruction. J Am Coll Surg. 2005;201(6):847-854. 14. Ten Broek RPG, Krielen P, Di Saverio S, et al. Bologna guidelines for diagnosis and management of adhesive small bowel obstruction (ASBO): 2017 update of the evidence-based guidelines from the World Society of Emergency Surgery ASBO working group. World J Emerg Surg. 2018;13:24. 15. Maung AA, Johnson DC, Piper GL, et al. Evaluation and management of small-bowel obstruction: an Eastern Association for the Surgery of Trauma practice management guideline. J Trauma Acute Care Surg. 2012;73(5 Suppl 4):S362-S369. 16. Teixeira PG, Karamanos E, Talving P, Inaba K, Lam L, Demetriades D. Early operation is associated with a survival benefit for patients with adhesive bowel obstruction. Ann Surg. 2013;258(3):459-465. 17. Sarr MG, Bulkley GB, Zuidema GD. Preoperative recognition of intestinal strangulation obstruction: prospective evaluation of diagnostic capability. Am J Surg. 1983;145(1):176-182. 18. Eskelinen M, Ikonen J, Lipponen P. Contributions of history-taking, physical examination, and computer assistance to diagnosis of acute small-bowel obstruction: a prospective study of 1333 patients with acute abdominal pain. Scand J Gastroenterol. 1994;29(8):715-721. 19. Schwenter F, Poletti PA, Platon A, et al. Clinicoradiological score for predicting the risk of strangulated small bowel obstruction. Br J Surg. 2010;97(7):1119-1125. 20. Zielinski MD, Eiken PW, Bannon MP, et al. Small bowel obstruction: who needs an operation? A multivariate prediction model. World J Surg. 2010;34(5):910-919. 21. Zielinski MD, Eiken PW, Heller SF, et al. Prospective, observational validation of a multivariate small-bowel obstruction model to predict the need for operative intervention. J Am Coll Surg. 2011;212(6):1068-1076. 22. Millet I, Taourel P, Ruyer A, Molinari N. Value of CT findings to predict surgical ischemia in small bowel obstruction: a systematic review and meta-analysis. Medicine (Baltimore). 2015;94(23):e983. 23. Mallo RD, Salem L, Lalani T, Flum DR. Computed tomography diagnosis of ischemia and complete obstruction in small bowel obstruction: a systematic review. J Gastrointest Surg. 2005;9(5):690-694. 24. Murray MJ, Gonze MD, Nowak LR, Cobb CF. Serum D(-)-lactate levels as an aid to diagnosing acute intestinal ischemia. Am J Surg. 1994;167(6):575-578. 25. Evennett NJ, Petrov MS, Mittal A, Windsor JA. Systematic review and pooled estimates for the diagnostic accuracy of serological markers for intestinal ischemia. World J Surg. 2009;33(7):1374-1383. 26. Raveenthiran V. Restorative resection of unprepared left-colon in gangrenous vs viable sigmoid volvulus. Int J Colorectal Dis. 2004;19(3):258-263. 27. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New York: John Wiley & Sons; 2000. 28. Margenthaler JA, Longo WE, Virgo KS, et al. Risk factors for adverse outcomes after the surgical treatment of small bowel obstruction. Ann Surg. 2006;243(4):456-464. 29. Ihedioha U, Alani A, Modak P, Chong P, O'Dwyer PJ. Hernias are the most common cause of strangulation in patients presenting with small bowel obstruction. Hernia. 2006;10(4):338-340. 30. Lange H, Jackel R. Usefulness of plasma lactate concentration in the diagnosis of acute abdominal disease. Eur J Surg. 1994;160(6-7):381-384. 31. Block T, Nilsson TK, Bjorck M, Acosta S. Diagnostic accuracy of plasma biomarkers for intestinal ischaemia. Scand J Clin Lab Invest. 2008;68(3):242-248. 32. Frager D, Baer JW, Medwid SW, Rothpearl A, Bossart PA. Detection of intestinal ischemia in patients with acute small-bowel obstruction due to adhesions or hernia: efficacy of CT and correlation with outcome. AJR Am J Roentgenol. 1996;166(1):67-71. 33. Sheedy SP, Earnest F, Fletcher JG, Fidler JL, Hoskin TL. CT of small-bowel ischemia associated with obstruction in emergency department patients: diagnostic performance evaluation. Radiology. 2006;241(3):729-736. 34. Millet I, Boutot D, Faget C, et al. Assessment of strangulation in adhesive small bowel obstruction on the basis of combined CT findings: implications for clinical care. Radiology. 2017;285(3):798-808. 35. Catena F, De Simone B, Coccolini F, Di Saverio S, Sartelli M, Ansaloni L. Bowel obstruction: a narrative review for all physicians. World J Emerg Surg. 2019;14:20.