Karnofsky and ECOG performance status scoring in lung cancer: a prospective, longitudinal study of 536 patients from a single institution

Abstract

The Karnofsky's index of performance status (KPS) and the Eastern Cooperative Oncology Group Performance Status Scale (ECOG PS) are widely used methods of assessing the functional status of cancer patients. In this study, we compare their predictive validity, and suggest a table of transformation between scales. 536 consecutive lung cancer patients were assigned both KPS and ECOG PS scores before, during and after treatment (in all, 1656 assignments). Patients were accurately staged at diagnosis, and carefully re-evaluated at each follow-up visit. Multiple clinical, laboratory and instrumental data were recorded along with performance status assessments. Survival times were measured from the pathological diagnosis. KPS and ECOG PS assignments were strongly related to each other (Spearman R = -0.869). Correlation between scales persisted unchanged in pretreatment and post-treatment assessments, advanced and limited diseases, response or non-response to treatment, and different assessors (R indices ranging from -0.825 to -0.901). A three-point conversion table showed the highest rate of success with an overall percentage of agreement exceeding 84% (grade 1: KPS = 100, 90, 80 and ECOG PS = 0, 1; grade 2: KPS = 70, 60 and ECOG PS = 2; grade 3: KPS < 60 and ECOG PS = 3, 4). Both univariate and multivariate analyses of survival documented the predictive validity of the two scales. However, KPS showed less ability than ECOG PS to discriminate patients with different prognosis. Because of the better predictive ability shown in this study, ECOG PS should be preferred to KPS. A general consensus on the scale to use could avoid problems of conversion, which is not always easy and free of errors.