Lobectomy and pneumonectomy for primary lung cancer were identified in 12,201 patients between January 2002 and June 2008. After excluding procedures for missing clinical staging or end points; institutions with more than 10 % missing data for clinical stage, discharge mortality, or length of stay; and patients treated with chemotherapy or radiation for unrelated disease, there remained 5376 resections. Study end points were discharge mortality, length of stay more than 14 days, and major morbidity. Multivariate analysis using propensity scores stratified into quintiles measured the effect of induction therapy.
In 525 of 5376 procedures (9.8 % ), chemotherapy (n = 153), radiotherapy (23), or chemoradiotherapy (349) preceded resection. Compared with resection only, patients receiving induction therapy were younger and had fewer comorbidities, more reoperative surgery, and higher rates of pneumonectomy. Clinical IIIA-N2 disease was treated with induction therapy in only 203 of 397 patients (51.1 % ). Propensity-adjusted rates detected no difference in discharge mortality, prolonged length of stay, or a composite of major morbidity for patients receiving induction therapy. Similar results were obtained in a logistic regression model (discharge mortality P = .9883; prolonged hospital stay P = .9710; major morbidity P = .9678).
Less than 10 % of all major lung resections for primary carcinoma and just more than half of all resections for clinical stage IIIA-N2 disease are preceded by neoadjuvant chemotherapy or radiation. This study does not support concerns over excessive operative risk of induction therapy.