Anthropometric measurements in non-small-cell lung cancer.

There is evidence that malnutrition is an important cause of morbidity and mortality in lung cancer patients and may have an impact on the clinical course of disease. The simplest way to assess nutritional status at the patient's bedside remains recourse to anthropometric measurements. This study was carried out in order to assess the clinical and prognostic significance of triceps skinfold thickness (TST), arm circumference (AC), and wrist circumference (WC) in lung cancer. The patient population was a consecutive series of 388 patients seen for a newly diagnosed primary non-small-cell lung cancer during the last 4 years. A set of 22 anthropometric, clinical, physical, laboratory, radiological, and pathological variables was prospectively recorded for all patients. Patients were carefully followed up, and their subsequent clinical course was recorded. The median values of TST, WC and AC were 8 mm (range 2-25 mm), 18 cm (range 10-27 cm), and 25 cm (range 15-35 cm), respectively. In 107 patients (27.6% of the total) TST values were below the reference value, and 37 of these patients also had a pathologically low small circumference. In all, AC was below the normality range in 60 of the 388 subjects (15.5%). Among the three variables, the strongest relationships were those between AC and WC (r(s)=0.541), and between TST and AC (r(s)=0.521). Univariate analyses of survival showed that TST was strongly predictive of a better prognosis (P<0.001), while WC was unrelated to outcome (P=0.101). Patients with higher values of AC had significantly longer survival than patients with lower values (P<0.018). The multivariate model, in contrast, did not confirm the prognostic capability of any of the anthropometric measures. These data indicate that the anthropometric measures may be significant predictors of survival, although not independently of the other prognostic factors.