Objective To analyze the association between postoperative pulmonary complications (PPCs) and diaphragm function in the elderly patients undergoing hepatobiliary surgery. Methods Clinical data were collected of 16 elderly patients who underwent hepatobiliary surgery at the Department of General Surgery of the Second Medical Center of Chinese PLA General Hospital. According to postoperative pulmonary complications, they were divided into six patients in the PPCs group (n=6) and non-PPCs group (n=10). Bedside ultrasound was used to monitor diaphragmatic mobility (DE) on the right side of the patients on the preoperative day 1 and the postoperative day 1,3, and 7. The two groups were compared in preoperative DE and postoperative minimal DE, ΔDE, operation time, and surgical approach. SPSS 23.0 were used for data analysis and multivariate logistic regression for analysis of the risk factors affecting PPCs. Receiver operating characteristic (ROC) curves were drawn to evaluate their predictive value for PPCs. Results There was no significant difference between the two groups in the preoperative DE and surgical approach (P>0.05). Compared with the non-PPCs group, the PPCs group had longer operation time [(247.500±68.099) vs (162.300±66.111) min], smaller postoperative minimal DE [(1.071±0.202) vs (1.414±0.236) cm] and greater ΔDE [(0.536±0.106) vs (0.343±0.139) cm], the differences being all statistically significant (P<0.05 for all). Multivariate logistic regression analysis revealed that operation time and minimal postoperative DE were independent risk factors for PPCs. The area under the ROC curve for predicting PPCs was 0.825 (95%CI 0.670-0.980) by operation time and 0.867 (95%CI 0.693-0.974) by postoperative minimum DE, and the optimal cut-off points were 210 min and 1.19cm (sensitivity 83.33% vs 90.00%; specificity 80.00% vs 83.33%). Conclusion The decline of diaphragm function after operation and prolonged operation are risk factors affecting PPCs in the elderly patients undergoing major hepatobiliary surgery. PPCs can be predicted by monitoring diaphragm function through ultrasound.