Background: In chronic obstructive pulmonary disease (COPD), the “obesity paradox” is a phenomenon without a clear cause. The objective is to analyze the complications of COPD patients according to their body mass index (BMI). Methods: An observational study with a six-year prospective follow-up of 273 COPD patients who attended a spirometry test in 2011. Survival and acute events were analyzed according to the BMI quartiles. Results: A total of 273 patients were included. BMI quartiles were ≤24.23; 24.24–27.69; 27.70–31.25; ≥31.26. During the follow-up, 93 patients died. No differences were found in exacerbations, pneumonia, emergency visits, hospital admissions or income in a critical unit. Survival was lower in the quartile 1 of BMI with respect to each of the 2–4 quartiles (p-value 0.019, 0.013, and 0.004, respectively). Advanced age (hazard ratio, HR 1.06; 95% confidence interval, CI 1.03–1.09), low pulmonary function (HR 0.93; 95% CI 0.86–0.99), exacerbator with chronic bronchitis phenotype (HR 1.76; 95% CI 1.01–3.06), high Charlson (HR 1.32, 95% CI 1.18–1.49), and the quartile 1 of BMI (HR 1.99, 95% CI 1.08–3.69) were identified as risk factors independently associated with mortality. Conclusions: In COPD, low BMI conditions a lower survival, although not for having more acute events.