Seasonal pattern in bipolar disorder and its relationship with clinical features

INTRODUCTION: Since ancient times, it has been known that seasonal changes have effects on clinical features, course and treatment of patients with bipolar disorder. The term seasonal pattern is described differently in various classification systems. The number of studies investigating the relation of seasonal pattern with clinical features using description in the DSM 5, the latest of these classifications, is very limited. Studies in this area have the potential to arise new explanations and approaches. OBJECTIVE: The aim of this study was to investigate the differences in clinical features between bipolar disorder patients with and without seasonal pattern according to the DSM 5 criteria. METHODS: The study was conducted in Affective Disorders Unit in Ege University School of Medicine, Department of Psychiatry between 4 January 2018 and 30 April 2018. Every consecutive patient who admitted to the outpatient clinic as well as recovered hospitalized patients who were planned to be discharged was screened for eligibility. Patients who were diagnosed with bipolar disorder according to the DSM 5 criteria, had experienced more than one same kind of mood episode, gave informed consent and did not have a disorder that would preclude giving valid and reliable information were recruited to the study. The processes of interviewing patients and their relatives, diagnosing the patient, deciding whether the inclusion criteria were met and collecting clinical information were done by two psychiatrists (Y.Z.Ö. and Ö.İ.K.). All possible sources of information were tried to be accessed and used in order to increase the reliability and quality of the information collected retrospectively. Primary sources for information were patient, at least one relative of patient, and all available medical records and documents. As a first step a life chart was formed with the gathered information for each patient. The necessary sociodemographic data, clinical information and treatment related information were collected with the guidance of the life chart and the Data Collection Form (Annex 2) prepared by the study team. RESULTS: One hundred and four patients were included in the study. The percentage of female patients was 57.7 while the percentage of male patients was 42.3. A seasonal pattern was associated with depressive episodes in 4.8%, with manic episodes in 1.9% and with hypomanic episodes in 1.9% of patients according to the seasonal pattern description in the DSM 5. When "any two consecutive years" criterion was used instead of the "last two years" in criterion C of the seasonal pattern description in the DSM 5, depressive episodes associated seasonal pattern rate was 23.1%, manic episodes associated seasonal pattern rate was 15.4% and the hypomanic episodes associated seasonal pattern rate was 4.8% among patients. Since there were only very few patients fulfilling DSM 5 seasonal pattern criteria this did not allow us to have comprehensive statistical analyses with these data. On the other hand, sociodemographic, clinical and treatment characteristics of patients were compared between two groups: patients with and without seasonal pattern after the change in criterion C. Results from these comparisons are as follows: The following characteristics were found to be significantly higher among patients who have depressive episodes associated seasonal pattern in compare to patients who have not depressive episodes associated seasonal pattern: number of lifetime depressive episodes (p<0.01), rates of patients who have atypical features in any of their depressive episodes (p=0.019), total number of mood episodes (p<0.01), rates of patients who have residual symptoms between episodes (p=0.006), rates of patients who have residual depressive symptoms (p<0.01) and mean frequency of mood episodes (p=0,032). Mean cycle lengths were significantly lower (p=0.014) in patients with depressive episodes associated seasonal pattern. The following characteristics were found to be significantly higher among patients with manic or hypomanic episodes associated seasonal pattern compare to patients without any manic or hypomanic episodes associated seasonal pattern: total number of manic episodes (p=0,02), total number of mood episodes (p=0,042), manic predominant polarity (p=0,018) and number of hospitalizations (p=0,018). On the other hand, the age of onset of illness in patients with manic or hypomanic episodes associated seasonal pattern was significantly lower (p = 0.023). Several characteristics were found to be significantly higher among patients with any kind of mood episode (depressive, manic or hypomanic) associated seasonal pattern compare to patients without any seasonal pattern. These are number of lifetime depressive episodes (p=0.013), number of lifetime manic episodes (p=0.001), rates of patients who have atypical features in any of their depressive episodes (p=0.003), rates of patients who have melancholic features in any of their depressive episodes (p=0.025), total number of mood episodes (p<0.01), mean frequency of mood episodes (p=0,012), rates of patients who have residual symptoms between episodes (p=0.034), number of hospitalizations (p=0.01) and rates of patients who have predominantly DMI pattern of course (p=0.025). CONCLUSION: Perhaps, the most important result, though not the primary aim of our study, is showing that criterion C of seasonal pattern in the DSM 5 has a restrictive description. Another important result is that bipolar disorder patients with seasonal pattern show many differences in clinical features, course and treatment characteristics compared to patients without seasonal pattern. These differences may be related to relatively more severe episode features and clinical course. However, the fact that the study conducted at a single centre with tertiary centre characteristics, limits the generalizability of the results. It is clear that there is a need for further studies with samples that better represent the bipolar disorder population. KEYWORDS: Bipolar disorder, seasonality, seasonal pattern, clinical features, course, treatment