Causes

The causes of bipolar disorder likely vary between individuals. Twin studies have been limited by relatively small sample sizes but have indicated a substantial genetic contribution, as well as environmental influence. For bipolar I, the (probandwise) concordance rates in modern studies have been consistently put at around 40% in monozygotic twins (same genes), compared to 0 to 10% in dizygotic twins. A combination of bipolar I, II andcyclothymia produced concordance rates of 42% vs 11%, with a relatively lower ratio for bipolar II that likely reflects heterogeneity. The overall heritability of the bipolar spectrum has been put at 0.71. There is overlap with unipolar depression and if this is also counted in the co-twin the concordance with bipolar disorder rises to 67% in monozigotic twins and 19% in dizigotic. The relatively low concordance between dizygotic twins brought up together suggests that shared family environmental effects are limited, although the ability to detect them has been limited by small sample sizes.

Genetic

Genetic studies have suggested many chromosomal regions and candidate genes appearing to relate to bipolar disorder's development, but the results are not consistent and often not replicated.

Although the first genetic linkage finding for mania was in 1969, the linkage studies have been inconsistent. Meta-analyses of linkage studies detected either no significant genome-wide findings or, using a different methodology, only two genome-wide significant peaks, on chromosome 6q and on 8q21. Neither have genome-wide association studies brought a consistent focus — each has identified new loci. Nonparametric linkage analysis using rank based methods did not detect genome-wide significant linkage findings whereas joint analysis of all linkage data sets identified two genome wide significant peaks on chromosome 6q and on 8q21. 

Findings point strongly to heterogeneity, with different genes being implicated in different families. A review seeking to identify the more consistent findings suggested several genes related to serotonin (SLC6A4 and TPH2), dopamine (DRD4 and SLC6A3), glutamate (DAOA and DTNBP1), and cell growth and/or maintenance pathways (NRG1, DISC1 and BDNF), although noting a high risk of false positives in the published literature. It was also suggested that individual genes are likely to have only a small effect and to be involved in some aspect related to the disorder (and a broad range of "normal" human behavior) rather than the disorder per se.

Advanced paternal age has been linked to a somewhat increased chance of bipolar disorder in offspring, consistent with a hypothesis of increased newgenetic mutations.

Physiological

Abnormalities in the structure and/or function of certain brain circuits could underlie bipolar. Meta-analyses of structural MRI studies in bipolar disorder report an increase in the volume of the lateral ventricles, globus pallidus and increase in the rates of deep white matter hyperintensities. Functional MRI findings suggest that abnormal modulation between ventral prefrontal and limbic regions, especially the amygdala, likely contribute to poor emotional regulation and mood symptoms.

According to the "kindling" hypothesis, when people who are genetically predisposed toward bipolar disorder experience stressful events, the stress threshold at which mood changes occur becomes progressively lower, until the episodes eventually start (and recur) spontaneously. There is evidence of hypothalamic-pituitary-adrenal axis (HPA axis) abnormalities in bipolar disorder due to stress.

Other brain components which have been proposed to play a role are the mitochondria, and a sodium ATPase pump, causing cyclical periods of poor neuron firing (depression) and hypersensitive neuron firing (mania). This may only apply for type one, but type two apparently results from a large confluence of factors.^] Circadian rhythms and melatonin activity also seem to be altered.

Environmental

Evidence suggests that environmental factors play a significant role in the development and course of bipolar disorder, and that individual psychosocial variables may interact with genetic dispositions. There is fairly consistent evidence from prospective studies that recent life events and interpersonal relationships contribute to the likelihood of onsets and recurrences of bipolar mood episodes, as they do for onsets and recurrences of unipolar depression. There have been repeated findings that between a third and a half of adults diagnosed with bipolar disorder report traumatic/abusive experiences in childhood, which is associated on average with earlier onset, a worse course, and more co-occurring disorders such as PTSD. The total number of reported stressful events in childhood is higher in those with an adult diagnosis of bipolar spectrum disorder compared to those without, particularly events stemming from a harsh environment rather than from the child's own behavior.

Neurological

Less commonly bipolar disorder or a bipolar-like disorder may occur as a result of or in association with a neurological condition or injury. Such conditions and injuries may include (but are not limited to) stroke, traumatic brain injury, HIV infection, multiple sclerosis, porphyria and rarely temporal lobe epilepsy.