Inflammation is reportedly associated with the development and progression of various malignancies. However, the clinical significance of preoperative and postoperative inflammation in lung cancer patients undergoing surgery is unknown. The relationship between preoperative and postoperative C-reactive protein (CRP), an indicator of inflammation, and survival was retrospectively analyzed in 356 patients who underwent complete resection of pathologic Stage I and II non-small cell lung cancers. Cutoffs for preoperative CRP (CRPpre), postoperative maximum levels of CRP (CRPmax), and postoperative CRP levels 30 days after surgery (CRP30) were determined as 0.2 mg/dL, 6.4 mg/dL, and 0.2 mg/dL, respectively. Patients with CRPprehigh, CRPmaxhigh, or CRP30high status had significantly poorer overall survival (OS) and relapse-free survival (RFS) than those with CRPprelow, CRPmaxlow, or CRP30low. Patients were stratified into 4 groups according to perioperative CRP grades, combining CRPprehigh, CRPmaxhigh, and CRP30high statuses, yielding groups with grades 0, 1, 2, and 3. OS and RFS significantly worsened with increasing grade. After controlling for potential confounders, the multivariate Cox proportional hazard model revealed perioperative CRP grade as an independent poor prognostic factor for OS (grade 3 vs grade 0): adjusted hazard ratio, 5.05; 95% confidence interval, 1.59-19.6; P = 0.005), and RFS (adjusted hazard ratio, 3.62; 95% confidence interval, 1.50-9.33; P = 0.004). Perioperative inflammation was associated with a long-term negative prognostic impact after lobectomy for lung cancer. Further prospective analysis is required to identify whether control of perioperative inflammation may improve prognosis after lung cancer surgery.