<b><i>Introduction:</i></b> Obesity and associated risk factors have been linked to cognitive decline before. <b><i>Objectives:</i></b> In the present study, we evaluated potential cumulative negative effects of overweight and obesity on cognitive performance in euthymic patients with bipolar disorder (BD) in a longitudinal design. <b><i>Methods:</i></b> Neurocognitive measures (California Verbal Learning Test, Trail Making Test [TMT] A/B, Digit-Symbol-Test, Digit-Span, d2 Test), anthropometrics (e.g., body mass index [BMI]), and clinical ratings (Hamilton Depression Scale, Young Mania Rating Scale) were collected over a 12-month observation period. Follow-up data of 38 patients with BD (mean age 40 years; 15 males, 23 females) were available. <b><i>Results:</i></b> High baseline BMI predicted a decrease in the patient’s performance in the Digit-Span backwards task measuring working memory performance. In contrast, cognitive performance was not predicted by increases in BMI at follow-up. Normal weight bipolar patients (<i>n</i> = 19) improved their performance on the TMT B, measuring cognitive flexibility and executive functioning, within 1 year, while overweight bipolar patients (<i>n</i> = 19) showed no change in this task. <b><i>Conclusions:</i></b> The results suggest that overweight can predict cognitive performance changes over 12 months.