Background: General anesthesia can affect intestinal function, but there is no objective and effective indicator to evaluate the inhibition and recovery of intestinal function. The main objective of this study is to assess whether bowel sounds (BSs) change before and after general anesthesia, then it can be explained that the BS can be an effective indicator of intestinal function. Methods: We randomly selected 26 inpatients and collected three sets of 5-minute continuous BSs before operation (Pre-op), after operation (Pro-op) and three hours after operation(3h-Pro-op) separately for each patient. The data were de-noised with adaptive filtering and wavelet threshold denoising, and processed with fractal dimension to identify the effective bowel sounds (EBSs). Then the linear and nonlinear characteristic values (CVs) of each EBS were extracted and paired t -test and rank-sum test were used to evaluate the changes of the BSs after general anesthesia. Results: For the difference between Pre-op and Pro-op, as well as between Pro-op and 3h-Pro-op, there are statistical differences ( p <0.05). Specifically, the linear CVs that can reflect the occurrence frequency, overall energy and overall duration of EBSs and the nonlinear CVs that can reflect the dispersion degree of stability and complexity of EBSs were statistically significant. However, there is no statistical difference in the CVs reflecting the energy and duration, as well as the stability and complexity of locally EBSs ( p >0.05). Also, there is no statistically significant difference between all the characteristic values between Pre-op and 3h-Pro-op ( p >0.05). Conclusion: The BSs change after general anesthesia. Furthermore, the BSs are weakened after general anesthesia and recovered to the state before general anesthesia three hours later. Therefore, the BS can be an indicator of intestinal function under general anesthesia, so as to provide guidance for postoperative feeding , which is of great clinical significance.