The belief that specific stages of the lunar cycle are related to human biology or behaviour has been present in popular culture since antiquity. Based on folklore or on clinical experience, several studies investigating possible links between the lunar cycle and mental illness were performed, with controversial results. To date, according to our knowledge, no study has yet considered a possible variability of the relationship with the seasons.

Our study aims to explore the relationship between the lunar cycle and acute manic states in bipolar patients, including a differentiation by seasons.

We designed an exploratory study, retrospectively analysing admissions of manic episodes over the period 2007/2013. Admission dates and dates of last symptomatic worsening were separated into seasons and then distributed into one of four quarters of the moon's synodic cycle. The last symptomatic worsening dates were obtained using information from admission notes, being considered whenever mentioned accurately and within no more than one week of admission.

Concerning the Autumn/Winter (n=221), we found a statistically significant disparity in the distribution of cases by lunar quarters in the admission dates of manic episodes (p=0.038), especially in episodes with psychotic symptoms (n=196; p=0.018), and in the last symptomatic worsening of manic episodes (n=60; p<0.001). No statistical significance was found during the Spring/Summer.

The lunar cycle appears to be strongly associated with acute manic states in bipolar patients, but the association seems determined by the season. Further studies are needed to corroborate or refute these results.