The 6-minute walk test (6MWT) is not intended to document oxygen (O) desaturation during exertion but is often used for this purpose. Because of this, it only has modest reproducibility in determining the need for ambulatory O therapy in patients with cardiopulmonary disease. The diagnostic and prognostic value of detecting exertional O desaturation is still unknown. The aims of this study were to estimate the prevalence of O desaturation during a 6MWT based on pulse oximetry measurements at the beginning and end of a 6MWT in a clinical population of patients with suspected cardiopulmonary disease and to determine whether the pulmonary function test (PFT) can predict exercise-induced desaturation during a 6MWT. This retrospective cohort study reviewed the results of the 6MWT and the PFT (i.e., spirometry, lung volumes, and diffusion capacity) of all patients who were evaluated for suspected cardiopulmonary disease at an academic medical center during a 5-year study period. The patients were categorized into three groups based on the change in O saturation by pulse oximetry (SpO) from start to end of the 6MWT: (1) SpO decreased by ≥3%; (2) SpO unchanged (-2 ≤ Δ ≤ 0%); and (3) SpO increased by ≥1%. Demographic, anthropometric, and lung function measurements were analyzed to determine which factors predicted O desaturation during the 6MWT. Of the 319 patients who underwent the 6MWT and the PFT from November 2005 until December 2010 (mean age = 54 ± 0.78 years, 63% women, 58% Whites, body mass index = 29.63 ± 8.10 kg/m), 113 (35%) had a decreased SpO, 146 (46%) had no change, and 60 (19%) had an increased SpO from the start to end of test. Our bivariate analysis found age, spirometric measures, and diffusion capacity for carbon monoxide (DLCO) had statistically significant inverse associations with the SpO change category ( < 0.001). Both a 3% and 4% drop in SpO during the 6MWT were statistically significantly associated with an older age, a higher prevalence of obstruction, and reduced forced vital capacity (FVC), forced expiratory volume in one second (FEV), FEV/FVC, DLCO and 6-minute walk distance (6MWD). Multivariable logistic regression analyses revealed that only DLCO was a significant independent predictor of the change in SpO and a ≥ 4% O desaturation during a 6MWT. Receiver operating curve analysis indicates DLCO cut-off of 45% is 82% sensitive and 40% specific in identifying ≥4% O desaturators, with an area under the curve of 0.788 ± 0.039 ( < 0.001). The prevalence of a ≥ 3% oxygen desaturation via pulse oximetry during a 6MWT in our clinical population of patients with suspected cardiopulmonary disease was 35%. Although age, spirometric lung volumes, and DLCO had statistically significant unadjusted inverse associations with the change in SpO during a 6MWT, the DLCO is the only significant independent predictor of both the magnitude of the change in SpO and the occurrence of O desaturation of at least 4%, respectively, during the test. Clinical Implications: A DLCO cut-off of 45% may be useful in identifying patients at risk for exertional hypoxemia during a 6MWT.
