BackgroundPara-aortic lymph node metastasis (PALNM) is a rare occurrence in colorectal cancer (CRC), and the high risk of radical lymphadenectomy leads to persistent debate about the best treatment strategy. This study aims to evaluate the predictor for PALNM and the clinical value of para-aortic lymph node dissection (PALND) in CRC patients with radiologically suspected synchronous PALNM. MethodsPatients who has synchronous radiologically suspected PALNM and underwent primary tumor resection were included. Logistic regression and receiver operating characteristic curve analysis were used to assess the predictive value of lymph node short axis in preoperative CT, identifying the optimal cut-off value. Propensity score matching and Cox regression explored factors affecting overall and disease-free survival, while Kaplan-Meier curves and decision tree models identified patient characteristics suitable for synchronous paraaortic lymph node dissection. ResultsA total of 578 patients were enrolled, and 125 patients received synchronous PALND. We found that simultaneous PALND significantly improved overall survival (HR, 0.56; 95% CI, 0.35-0.91; P = .019) in multivariate analysis, while disease-free survival showed no significant difference (P = .41). The short axis diameter of PALN on preoperative CT is a crucial predictor of PALNM (P < .001, AUC = 0.759) with a threshold of > 7 mm. N-stage and distant metastasis were included as independent predictors in the diagnostic model to enhance accuracy. A larger short axis diameter of PALN correlated with advanced tumor stage and poorer prognosis. Subgroup analysis revealed that PALND offers survival benefits for colorectal cancer patients at all stages with a short axis diameter >10mm on preoperative CT (P = .037) and for stage III patients with a diameter between 7 to10 mm (P < .001, AUC = 0.810). ConclusionSynchronous PALND can improve overall survival in CRC patients with suspected PALNM, with the maximum short axis diameter of PALN serving as a key criterion for selecting patients for surgery.