Data Sheet 2_Incidence and management outcomes of chylothorax after pulmonary resection for primary lung cancer: a systematic review and meta-analysis.pdf

Exploratory subgroup analyses and meta-regression examined incidence; sensitivity analyses and assessments of small-study effects were conducted for the pooled outcomes.</p>Results<p>Twenty-eight studies involving 212,637 patients were included, with 1,798 cases of postoperative chylothorax. The pooled incidence was 1.4% (95% CI: 1.0%–1.7%; I²=94.87%). Subgroup analyses by surgical approach and diagnostic criteria showed no significant between-subgroup differences, with high within-subgroup heterogeneity.

Across 23 studies, the pooled proportion successfully treated non-surgically was 86.9% (95% CI: 82.0%–91.2%; I²=49.73%). Across 22 studies, the pooled proportion receiving subsequent surgical or interventional treatment was 12.8% (95% CI: 8.4%–17.9%; I²=52.32%). Leave-one-out and GLMM-based sensitivity analyses supported the robustness of the pooled estimates.

Egger’s regression suggested a possible small-study effect for incidence, but trim-and-fill adjustment did not materially change the pooled estimate. Narrative synthesis showed variation in diagnostic criteria, initial non-surgical regimens, escalation strategies, and definitions and reporting of management outcomes.</p>Conclusion<p>Chylothorax after pulmonary resection for primary lung cancer has a low incidence, but its management burden should not be overlooked.

Most cases were controlled with initial non-surgical treatment, whereas some required subsequent surgical or interventional treatment. Diagnostic and treatment-escalation criteria remain inconsistent. Future studies should prespecify and validate diagnostic and treatment-escalation criteria and standardize the definition of successful non-surgical treatment and outcome reporting to improve comparability and provide more reliable evidence for postoperative surveillance and clinical management decisions.</p>Systematic Review Registration<p>crd.york.ac.uk/PROSPERO/view/CRD420261392561, identifier CRD420261392561.</p>