Survival of patients with or without symptoms undergoing potentially curative resections for primary lung cancer

Numerous historical screening programs to detect lung cancer have been undertaken. With technologic advances, complimentary diagnostic tests have been developed; however, only the National Lung Cancer Trial has demonstrated increased survival. Following the success of this study, screening programs are being trialled in several countries. Screening should, in theory, reduce lung cancer deaths by identifying asymptomatic patients with earlier tumors. This study asked whether lung cancer patients who are asymptomatic at presentation have a better survival than those who present with symptoms.

This was a retrospective analysis of a validated prospective thoracic surgery database from a tertiary center in the Northwest of England. Included were 1,546 consecutive patients (826 men, 720 women) who received operative intervention for non-small cell lung cancer. The main outcome measures included 5-year survival and univariate and multivariate Cox regression analysis.

Cancer stage, age, and operation type were confirmed as being of prognostic importance, validating previous studies. Survival between asymptomatic or symptomatic patients did not differ significantly (p = 0.489), regardless of stage. The hazard ratios (with 95% confidence intervals) for variables associated with poorer outcome identified by Cox's regression analysis were male sex, 1.34 (1.15 to 1.56); advancing age, 1.03 (1.02 to 1.04); advancing stage, 3.30 (2.69 to 4.04); and pneumonectomy, 1.24 (1.01 to 1.52). Symptoms were not a significant variable affecting survival on multivariate analysis.

This retrospective study from the Northwest of England showed that in our subset of lung cancer patients undergoing resection, asymptomatic patients with non-small cell lung cancer do not have improved survival, implying it is a systemic disease in many at diagnosis. Care should be taken when generalizing the results of the National Lung Screening Trial to all populations until further validation has been performed.