| Summary: | BACKGROUND: Postoperative fever may serve as an indirect sign to reflect the alterations of the host milieu caused by surgery. It still remains open to investigation whether postoperative fever has a bearing on prognosis in patients with lymph node negative breast cancers. METHODS: We performed a retrospective study of 883 female unilateral patients with lymph node negative breast cancer. Fever was defined as an oral temperature ≥38 in one week postoperation. Survival curves were performed with Kaplan-Meier method, and annual relapse hazard was estimated by hazard function. FINDINGS: The fever patients were older than those without fever (P<0.0001). Hypertensive patients had a propensity for fever after surgery (P = 0.011). A statistically significant difference was yielded in the incidence of fever among HR+/ERBB2-, ERBB2+, HR-/ERBB2- subgroups (P = 0.012). In the univariate survival analysis, we observed postoperative fever patients were more likely to recur than those without fever (P = 0.0027). The Cox proportional hazards regression analysis showed that postoperative fever (P = 0.044, RR = 1.89, 95%CI 1.02-3.52) as well as the HR/ERBB2 subgroups (P = 0.013, HR = 1.60, 95%CI 1.09-2.31) was an independent prognostic factor for relapse-free survival. CONCLUSION: Postoperative fever may contribute to relapse in node negative breast cancer patients, which suggests that changes in host milieu related to fever might accelerate the growth of micro-metastatic foci. It may be more precise to integrate both tumor- and host-related factors for the evaluation of relapse risk.
|
|---|
