The Effects of Electroconvulsive Therapy on Depression and Suicidality in Transgender and Gender Diverse Individuals

To the Editor,

Transgender and Gender Diverse (TGD) individuals experience incongruity between their sex assigned at birth and their current gender identity. Due to a variety of factors including stigma and gender minority stress, TGD people experience higher rates of mood disorder than the general public, with a recent national registry-based Swedish study finding a 6-fold higher rate of mood disorders in TGD people compared to cisgender people.¹ Electroconvulsive therapy (ECT) is an effective and cost-effective treatment for mood disorders. While in prospective trials ECT has demonstrated equivalent tolerability and efficacy in patients identifying as male or female, its effectiveness in patients with other gender identities has not been significantly explored. This study reports the outcomes of a series of TGD patients who were treated with ECT at a single center.

ECT patients provided self-reported outcomes data using the Quick Inventory of Depressive Symptomatology Self Report 16 item scale (QIDS)² and the Behavior and Symptom Identification Scale-24 (BASIS-24).³ Two items of the BASIS-24 enquire about the frequency of self-harm thoughts and suicidal ideation (SI) in the past week, with a 5 point scale ranging from 0 “never” to 4 “always.” For demographics, the BASIS-24 asks patients “What is your gender?” with response options of “male,” “female,” and “other,” with individuals selecting “other” asked to provide a free-text description of their gender.

The study population consists of patients aged 16 or older undergoing first ECT treatment course from Jan 2015 to Mar 2020, and who selected “other” for their gender. All patients received ECT using a Mecta Spectrum 5000Q (Tualatin, OR), with individualized seizure threshold determination at the time of first treatment,⁴ and subsequent treatments administered at a default frequency of thrice weekly with electrode placement and charge determined by the treating psychiatrist. Methohexital was the default anesthetic agent, with succinylcholine as the muscle relaxant. Patients were followed for the first 10 treatments or until ECT discontinuation, whichever came first, with QIDS and BASIS-24 repeated after every 5^th treatment. This retrospective cohort study was approved by the Partners Healthcare IRB with a waiver of informed consent.

A total of 19 patients met inclusion criteria. Mean age was 24.1±5.5 years, and primary clinical diagnoses were major depressive disorder (13; 68.4%), bipolar disorder (5; 26.3%), and other (1; 5.3) (Table S1). Patients self-reported a range of gender identities, with 9 (47.3%) identifying as gender non-binary (Table S2).

At baseline patients had severe-range depression on the QIDS, with a mean score of 17.0 ± 4.9. Median BASIS-24 SI subscale was 1.25 (IQR 0.42 to 3.00) (Figure). At treatment #5, mean QIDS was 11.2 ± 4.5, with a median SI subscale of 0.58 (IQR 0.00 to 1.79). At treatment #10, 14 patients continued in ECT; mean QIDS was 9.6 ± 4.8, and median BASIS-24 SI subscale was 0.50 (IQR 0.00 to 1.46). Looking at the individual trajectories of patients, 16 of 19 patients had an improvement in QIDS score between baseline and the time of last follow-up, while the remaining 3 had unchanged QIDS. For suicidality, 10 of 19 patients had lower BASIS-24 self-harm subscale at the time of last treatment, while 4 remain unchanged and 5 had worsened (Figure 1).

Figure:

Box plot of QIDS (top) and BASIS-24 suicidal ideation and self-harm index (bottom) scores between baseline and treatment #10. Boxes show the data range, with a horizontal line at the median. Solid lines trace the individual trajectories for the patients in the sample.

Literature in ECT in TGD patients is sparse, with the only prior reports focusing on transgender patients specifically, and cumulatively reporting on only 9 patients.^5–7 In addition to being the largest case series yet reported, this study includes patients with gender identities besides transgender. Among our 19 TGD patients, ECT was associated with a decrease in depressive symptoms as measured by the QIDS. The magnitude of this improvement (from a baseline score of 17.0 ± 4.9 to 9.6 ± 4.8 following treatment #10) is similar to the magnitude of improvement seen among all 424 patients aged 30 and younger treated at our study site, where mean QIDS reduced from 17.0 ± 4.9 to 10.3 ± 5.3 over 10 ECT treatments.⁸

There is clear evidence that TGD individuals suffer from higher rates of suicidal ideation and non-suicidal self-injury relative to cisgender individuals.⁹ At present there are no randomized controlled trials of interventions to reduce SI in the TGD population, and limited evidence in general for pharmacological strategies to reduce suicide generally.¹⁰ The results of this study suggest that TGD patients with significant depression and SI may have improvement in their suicidal thinking when treated with ECT, although notably reduction in SI occurred in only 10 of 19 patients in this sample, and SI actually worsened in 5 patients despite consistent improvement in depressive symptoms.

Psychiatry has a long and complicated history with sexual orientation and gender expression, some of which shamefully involved aversive treatment, including electric shocks, in an attempt to change sexual identity. The legacy of this past may have ongoing effects on the perception of ECT. In contrast to these purposely aversive methods, modern ECT is a safe and effective treatment for underlying depression without any attempt to modify gender identity. These results support the idea that ECT is an effective treatment of depressive disorders regardless of the gender identity of the patient receiving treatment.

Limitations of this study include its retrospective observational nature without control group, which prevents discussion of alternative treatment methodologies. Additionally, the one question used to assess gender identity in this study is suboptimal, and a better strategy would include two questions asking about sex assigned at birth and present gender identity.¹¹ It is likely that the methodology used here significantly undercounts transgender individuals, who are likely to select the sex corresponding to their gender identity and thus would not be identified in this sample.