Bipolar disorder bordering neurological conditions: A study of phenotypic and genotypic overlaps
MetadataVis full innførsel
Humans have tried to discern – or diagnose – mental disorders for thousands of years. The validity of diagnoses can be assessed in many ways. Construct validity assumes that patients with the same diagnosis should have specific phenotypic and genotypic features that converge within the group and diverge from other diagnoses. Bipolar disorder is a diagnosis defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM 5). Symptoms of bipolar disorder are in some cases regarded a consequence of neurological conditions like traumatic brain injury (TBI), epilepsy, or Alzheimer’s disease. If so, the DSM 5 diagnosis of bipolar and related disorders due to the neurological condition should be used. However; indices of neurological injury, dysfunction, or disease do not per se qualify for the latter diagnosis in the DSM 5. The overall aim of the thesis is to assess the construct validity of bipolar disorder with and without indices of neurological conditions. The hypothesis of Paper I is that bipolar disorder with premorbid TBI has phenotypic overlaps with bipolar and related disorders due to TBI (i.e. more irritability and aggression and an increased hypomania/mania:depression ratio). The hypothesis of Paper II is that bipolar II disorder with epileptiform discharges or other sharp activity on encephalograms has phenotypic overlaps with bipolar and related disorders due to epilepsy (i.e. an altered course of illness). The hypothesis of Paper III is that bipolar disorder per se has genotypic overlaps with Alzheimer’s disease. In Paper I we compared clinical characteristics of bipolar disorder between 37 patients with and 468 patients without premorbid TBI. We obtained data from the cross-sectional multi-center Bipolar Research and Innovation Network Norway (BRAIN)-study. Premorbid TBI was independently associated with more disruptive symptoms and comorbid migraine, but not with an increased hypomania/mania:depression-ratio. In Paper II we used data from the BRAIN- 9 study to compare course of illness in bipolar II disorder among 12 patients with and 74 patients without epileptiform discharges or other sharp activity on electroencephalograms (EEGs). We found an increased hypomania:depression-ratio among patients with the specified findings on EEGs. In Paper III we used summary statistics from genome wide association studies (GWASs) obtained from the International Genomics of Alzheimer’s Project part 1 (17,008 cases, 37,154 controls) and the Psychiatric Genomics Consortium Bipolar Disorder Working Group (20,352 cases, 31,358 controls). We found a polygenic overlap without genetic correlation between bipolar disorder and Alzheimer’s disease and identified two novel loci associated with both conditions, whereof one of the loci had the same direction of effect on risk. Ours and previous findings suggest that bipolar disorder with premorbid TBI is phenotypically more similar to bipolar disorder in general than to bipolar and related disorders due to TBI. Bipolar II disorder with epileptiform discharges or other sharp activity on EEG, however, might have a distinct course of illness overlapping with bipolar and related disorders due to epilepsy. Bipolar disorder in general can largely be phenotypically and genotypically separated from Alzheimer’s disease, but the conditions probably have some overlapping genetic influences. Conceptualizing bipolar disorder as a pathway disorder implies that neurological conditions could contribute to the pathophysiology of bipolar disorder symptoms through biological pathways distinct from or shared with those typical for bipolar disorder, or be a consequence of biological pathways underlying bipolar disorder. Our results can have some consequences for the diagnostic evaluation, treatment, and prognosis of patients with bipolar disorder and indices of neurological conditions. However, most of our findings warrant replication with more robust methods before they could inform clinical practice.