According to the "kindling" hypothesis, when people who are genetically predisposed toward bipolar disorder experience stressful events, the amount of stress required to bring about a change in mood becomes progressively lower, until the episodes eventually start (and recur) spontaneously.
The term kindling is appropriate because the process bears similarity to the process of setting a log on fire. On its own, a log is not quick to burn. However, by surrounding the log with more easily combustible materials, the log soon catches fire. Eventually, once the log has kindled, it will continuously burn without any help, and will be harder to put out. Source
Dr. Robert M. Post of the National Institute of Mental Health (USA) is credited with first applying the kindling model to bipolar disorder (NARSAD). Demitri and Janice Papolos, in their excellent book The Bipolar Child, describe this model as follows:
- ... initial periods of cycling may begin with an environmental stressor, but if the cycles continue or occur unchecked, the brain becomes kindled or sensitized - pathways inside the central nervous system are reinforced so to speak - and future episodes of depression, hypomania, or mania will occur by themselves (independently of an outside stimulus), with greater and greater frequency.
In addition, it has been shown that substances such as cocaine and alcohol have their own kindling effects which can contribute to bipolar kindling. In fact, it was the knowledge that cocaine causes seizures that led Dr. Post to connect kindling in epilepsy with mood disorders, after he had studied the unexpected effects of cocaine on severely depressed patients (NARSAD).
A study led by Dr. Joseph Goldberg found that patients diagnosed with both bipolar disorder and substance abuse were much more likely to respond to treatment that included an anticonvulsant/mood stabilizer, divalproex (Depakote) or carbamazepine (Tegretol), with or without Lithium, than treatment with Lithium alone. At the same time, patients who had bipolar disorder but no history of substance abuse had similar remission rates with both types of treatment. Dr. Goldberg did note that more controlled studies are needed on the role of anticonvulsants in treating dual diagnosis patients. (Substance Abuse, 2000) Source