# Pneumoperitoneum without significant bowel perforation in patients with blunt trauma: a systematic review and meta-analysis

## TL;DR

This study finds that about half of trauma patients with CT-detected free intraperitoneal air do not have significant bowel perforation, suggesting that not all cases require surgery.

## Contribution

The study quantifies the proportion of blunt trauma patients with CT-detected free intraperitoneal air who do not have significant bowel perforation, challenging traditional surgical guidelines.

## Key findings

- Approximately 51% of patients with CT-detected free intraperitoneal air had no significant bowel perforation.
- The pooled analysis showed that 34% of free intraperitoneal air cases did not require surgery.
- Benign pneumoperitoneum was often associated with high-energy trauma and managed nonoperatively in some cases.

## Abstract

Free intraperitoneal air (FIA) after blunt trauma is traditionally considered a radiological marker of hollow viscus perforation requiring urgent laparotomy. However, emerging reports have described pneumoperitoneum without surgically meaningful bowel injury, raising concerns about unnecessary operations. This systematic review and meta-analysis aimed to quantify the proportion of patients with blunt trauma with computed tomography (CT)-detected FIA who had no significant bowel perforation, defined as either (1) non-therapeutic laparotomy with no identified macroscopic perforation or (2) successful nonoperative management without subsequent clinical deterioration.

This review followed the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and was prospectively registered in the International Prospective Register of Systematic Reviews (CRD42020202174). PubMed, Embase, Scopus, and Web of Science were searched through November 13, 2025, for observational studies reporting the outcomes of patients with blunt trauma with CT-detected FIA. Two reviewers independently performed study selection, data extraction, and quality assessment using the Newcastle–Ottawa Scale. A random-effects meta-analysis was performed to estimate the pooled proportion of FIA cases without significant perforation. Heterogeneity was assessed using I-squared statistic (measure of heterogeneity) and τ2; small-study effects were examined using contour-enhanced funnel plots and Egger’s regression. Case reports meeting eligibility criteria were narratively summarized.

Fourteen studies comprising 8,972 patients with blunt trauma were included. Among them, 239 (2.7%) had CT-detected FIA. In the FIA subgroup, 117 patients (49.0%) had surgically confirmed bowel perforation, whereas 122 (51.0%) had no significant perforation, defined as a non-therapeutic laparotomy or a stable nonoperative clinical course. Among patients without FIA on CT, 56 of 8,733 (0.6%) had bowel perforation identified during surgery. The pooled analysis showed that 34% (95% CI 14–63%) of patients with FIA had no significant perforation. Substantial heterogeneity was observed (I-squared statistic = 80.3%, τ2 = 3.26, p < 0.001), reflecting variations in CT acquisition, diagnostic criteria, and operative thresholds. Funnel plot asymmetry suggested potential small-study effects. Additionally, 19 case reports (20 patients) published between 1999 and 2025 illustrated that benign pneumoperitoneum most often occurred in young men following high-energy trauma, commonly associated with pneumothorax or pneumomediastinum; most underwent nontherapeutic laparotomy, whereas several were successfully managed nonoperatively.

A noteworthy subgroup of patients with blunt trauma with CT-detected FIA did not exhibit clinically significant bowel perforations. Although FIA remains an important radiologic warning sign, it is not an independent diagnostic indicator of significant hollow viscus injury. Clinical decision-making should integrate clinical assessment with adjunctive CT findings rather than rely on FIA alone. Owing to the rarity of FIA and limited sample sizes, larger prospective studies are required to refine the diagnostic performance of FIA and optimize selective nonoperative management strategies.

The online version contains supplementary material available at 10.1186/s13017-026-00673-3.