Posttraumatic Stress Disorder Symptom Severity Does Not Predict Depression Improvement, but May Impact Clinical Response and Remission

^aVA Rehabilitation Research and Development Service (RR&D) Center for Neurorestoration and Neurotechnology, Providence, Rhode Island
^bDepartment of Psychiatry and Human Behavior, Alpert Medical School of Brown University, Providence, Rhode Island
*Corresponding author: Noah S. Philip, MD (noah_philip@brown.edu).

Published online: March 30, 2021.
Potential conflicts of interest: The authors report no relevant biomedical conflicts of interest.
Funding/support: The authors are supported by the VA RR&D Center for Neurorestoration and Neurotechnology; effort was also supported by US Department of Veterans Affairs grant I01 HX002572 (Drs Bozzay and Philip) and National Institutes of Health (NIH) grant P20 GM130452 (Dr Philip).
Role of the sponsor: The funders had no role in data collection and analysis, decision to publish, or preparation of the manuscript.  
Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or NIH.

J Clin Psychiatry 2021;82(3):20l13751

To cite: Brigido S, Bozzay M, Philip NS. Posttraumatic stress disorder symptom severity does not predict depression improvement, but may impact clinical response and remission. J Clin Psychiatry. 2021;82(3):20l13751.
To share: https://doi.org/10.4088/JCP.20l13751

© Copyright 2021 Physicians Postgraduate Press, Inc.

See reply by Hernandez et al.

To the Editor: With great interest, we read the recent article by Hernandez and colleagues¹ examining whether a diagnosis of comorbid posttraumatic stress disorder (PTSD) was associated with poorer treatment outcomes of veterans receiving transcranial magnetic stimulation (TMS) for major depressive disorder (MDD). This question is of high clinical relevance, as PTSD and depression are highly comorbid in this patient population, and the interaction between these disorders worsens symptom severity,² which can complicate treatment. As such, it is possible that the severity of PTSD symptoms also has ramifications for the effectiveness of treatment. Therefore, we sought to examine the effects tested by Hernandez et al¹ in our Veterans Affairs (VA) neuromodulation clinic while also considering the effects that PTSD symptom severity can have on TMS treatment for depression.

To this end, we conducted a chart review of 57 patients with MDD and comorbid PTSD; this cohort was treated naturalistically between 2012 and 2020 (ie, before the coronavirus disease 2019 [COVID-19] pandemic). As a part of clinical care, patients completed self-reports to evaluate PTSD symptom severity using the PTSD Checklist for DSM-5 (PCL-5)³ and MDD symptom severity using the Patient Health Questionnaire-9 (PHQ-9).⁴ Only patients with both baseline and endpoint scores on both rating scales were included in this analysis.

We examined effects of PTSD diagnosis (defined as meeting threshold-level symptoms of PTSD; ie, PCL-5 score > 33) or symptom severity (as determined by PCL-5 score) on the primary outcomes investigated by Hernandez and colleagues.¹ First, we conducted repeated-measures analyses of variance to examine whether PTSD symptom severity or diagnosis at baseline predicted improvement in depressive symptoms from pre– to post–TMS treatment. In general, patients with worse PTSD symptoms at baseline (PCL total score: F_1,54 = 35.59, P = .000, η_p² = 0.40; threshold: (F_1,54 = 10.93, P = .002, η_p² = 0.17) exhibited more severe depressive symptoms. However, similar to the findings of Hernandez et al,¹ neither PTSD diagnosis (F_1,54 = 0.19, P = .67, η_p² = 0.00) nor symptom severity (F_1,54 = 0.18, P = .67, η_p² = 0.00) differentiated the degree of improvement in depressive symptoms with TMS.

Next, we extended the findings of Hernandez et al¹ by conducting logistic regression to examine the extent to which PTSD diagnosis or symptom severity predicted MDD categorical response (ie, at least 50% reduction in PHQ-9 score) and remission (ie, PHQ-9 score < 5), while controlling for baseline depressive symptom severity. Regarding response, both diagnosis (B = −1.45; SE = 0.66; OR = 0.24; 95% CI, 0.06–0.86; P = .029) and symptom severity (B = 0.06; SE = 0.02; OR = 0.95; 95% CI, 0.91–0.99; P = .026) scores predicted poorer outcomes. For remission, only PTSD diagnosis (B = −1.61; SE = 0.76; OR = 0.20; 95% CI, 0.05–0.90; P = .035), but not PTSD severity (B = –0.05; SE = 0.27; OR = 0.96; 95% CI, 0.91–1.01; P = .088), predicted poorer outcomes.

In summary, we partially replicated the findings of Hernandez et al, as our results suggest that a categorical diagnosis of PTSD was not predictive of depressive symptom improvement. This finding indicates TMS can be an effective treatment for patients with comorbid PTSD and depression, regardless of the severity of symptom presentation. Yet, when we examined categorical outcomes in depression, a more complicated picture emerged. Having sufficient symptom severity to reach threshold for diagnosis (our operational definition of PTSD diagnosis) predicted reduced chance of clinical response and remission, which is a finding reminiscent of that reported in prior randomized controlled trials of TMS for depression.⁵ These results underscore the importance of understanding baseline symptom severity to predict how patients may respond to TMS and provide realistic expectations to those seeking treatment.

Taken together, both our data and those from Hernandez et al¹ suggest that veterans with comorbid depression and PTSD can be effectively treated with TMS, and clinicians should consider PTSD symptom severity at the start of TMS treatment. While these findings require replication in larger and prospective studies, this work and the report from Hernandez et al contribute to the ever-growing effectiveness literature supporting the use of TMS for MDD and commonly comorbid conditions including PTSD^6–8 and suicidality.⁹ The primary limitations of this work are those inherent to chart reviews of naturalistic patient care; we were not powered to evaluate whether medication or other treatment influenced these findings, although, as a requirement at our clinic, prior treatments must be stable for at least 6 weeks. We also did not perform structured clinical interviews, but relied on patient self-reported outcomes. These caveats aside, the results presented by Hernandez et al and the research mentioned in this letter highlight the importance of improving our understanding of TMS outcomes to improve clinical care.