The RANZCP professional practice guidelines for electroconvulsive therapy: Time for evidence-based,best practice electroconvulsive therapy to become the uniform standard across Australia and New Zealand

Weiss et al. (2019) have provided clear, practical guidelines for psychiatrists in New Zealand and Australia to encourage best practice electroconvulsive therapy (ECT) management of patients receiving ECT. The evidence-based recommendations for placement, pulse width and stimulus dosing should be uniformly followed by prescribing psychiatrists across both countries. ECT should be provided at the highest possible standard, given that it is subjected to more public scrutiny than other psychiatric treatments.

It is pleasing to see bitemporal (BT) ECT relegated to a second- or third-line option, to be infrequently used in relatively exceptional circumstances. However, there may be some psychiatrists who still prefer this placement, based on the historical view that it is more effective than other placements, especially unilateral ECT. As the authors point out, Semkovska’s non-inferiority trial in 2016 and Kolshus’ meta-analysis in 2017 clearly demonstrated that for depression at least, right unilateral (RUL) ECT with a 1-millisecond pulse width (PW) and a stimulus dose set at 6x seizure threshold (ST) is equivalent to BT ECT across all efficacy measures, including speed of response, but with a preserved cognitive advantage. With regard to ECT for psychosis, the authors note Phutane’s study in 2013, showing bifrontal (BF) ECT to be slightly superior to BT ECT for schizophrenia in both efficacy and side effects. This finding has now been replicated by Bansod et al. (2018). Therefore, there are very few situations where BT ECT should be given as a first-line prescription, and psychiatrists who still routinely use BT ECT should now update their practice.

The guidelines also make it clear that the effective stimulus dose for RUL ECT with a 1-millisecond PW is 5–6x ST, and not lower. They also find no support for BF or BT ECT given with an ultrabrief (UB; 0.25–0.3 milliseconds) PW, as speed of response is even slower than RUL UB ECT (Sienaert et al., 2009). If there are centres in Australia or New Zealand that still use doses of 3x or 4x ST for RUL ECT, or provide BF or BT ECT with a 0.3-millisecond stimulus, they need to adopt the evidence-based standard of 5–6x ST for RUL ECT and abandon UB bilateral ECT altogether.

I was pleased to see the guidelines caution against using a poor-quality seizure electroencephalogram (EEG) as the sole basis to escalate the stimulus dose during an ECT course. As the authors rightly point out, there are many factors other than rising seizure threshold that can interfere with EEG quality. The practice of rapidly increasing the stimulus dose to ‘chase a good EEG’ can result in excessive and unnecessary cognitive side effects. The authors suggest first considering a sensible range of factors that might have caused a poor EEG. However, a missing option is to re-titrate the seizure threshold during the course if the EEG quality declines, to be more certain of any change in threshold. Some centres now advocate an automatic retitration of threshold at session 6 or 7, particularly for RUL UB ECT (Suetani and Waite, 2014).

The guidelines include an appropriate debate about the relative merits of providing a suprathreshold stimulus 60–90 seconds after obtaining the threshold seizure at the titration session, versus ending the session after ST is found and treating with the suprathreshold stimulus at the next session. The main argument in favour of the first approach is that this ensures a therapeutic seizure is given at the first session. However, the guidelines point out that there is emerging evidence that the threshold seizure at the titration session may still be therapeutic and that providing a suprathreshold stimulus 60–90 seconds later may only add more cognitive side effects for little benefit, especially if a ‘top-up’ bolus of anaesthetic has been required to keep the patient asleep. This is an area that would benefit from further research.

Finally, the guidelines give some prominence to ECT with a 0.5-millisecond PW. The evidence base is probably too weak for it to have been included in Table 1 alongside the more robust options of 1-millisecond PW ECT and RUL UB ECT, but at least the guidelines do acknowledge the lower level of evidence for 0.5-millisecond ECT. However, I would take issue with their assertion that 0.5-millisecond ECT is ‘at the lower end of the brief pulse range’. The evidence that exists about 0.5-millisecond PW ECT suggests it is fact closer to 0.3 milliseconds than 1 millisecond in terms of efficacy and adverse effects. In fact, Sienaert et al. (2018) argue persuasively that 0.5-millisecond PW ECT should be considered a wider form of UB ECT rather than a narrower form of brief pulse ECT. However, there is also a good argument for simply naming the actual pulse width in milliseconds and abandoning the terms ‘ultrabrief’ and ‘brief pulse’ altogether.