"Another proposed mechanism of action is that the seizures induced by ECT cause a profound change in sleep architecture (a reversible inhibition of REM sleep); it is this change in the state of the organism that drives the therapeutic effects of ECT and not any simple change in the release of neurotransmitters, neurotrophic factors and/or hormones."

So a small percentage of patients not relapsing after one sleep deprivation would actually make it the best available cure with reduced risks compared to ECT.

Other treatments possibly aren't curative and have not as serious side effects as ECT, more similar to sleep deprivation.

I'm not really interested in giving anyone medical advice, etc, based on my selective reading of pubmed, I just think it is important to list treatment options neutrally and with more consideration of what self treatment patients are probably very effectively applying if managed medical treatment is supposed to have a safer and better outcome.