Text version
Guilty of double standards? Who? Me? Okay, okay. If a new drug treatment for postpartum depression funded by a pharmaceutical company was tested in a randomized trial and produced the results seen in this paper, I would be pretty skeptical. But since this is light therapy, can we overlook this small sample size of 30, the 23% dropout rate, and some doubt about the blinding? Because, after all, wouldn’t it be great to have a rapid-acting treatment for postpartum depression that’s not taking on the unknown potential complications of messing with the allopregnanolone system as brexanolone and zuranolone do?
Hi! Jim Phelps here for the Psychopharmacology Institute. Let’s start with the risk side as that might influence how you look at efficacy. It does for me. Out of 15 participants in the light therapy group, 6 reported headaches as a side effect. Five of them completed the study, and 1 dropped out. So, 6 out of 15 patients—that’s 40% with headache vs 1 patient amongst the 15 on placebo. And what was the placebo? Well, it was the same light box but emitting only 500 lx instead of 10,000 lx. No data are reported regarding the success of the blinding—no pun intended, which leaves some room for doubt there. So, that’s it for risk of headache, which was tolerable for all but 1 patient.
To my knowledge, no long-term risks have been described for short-term use of light therapy except the induction of hypomanic or manic switch, of which there were none in this study. That might be due to the exclusion of patients with a prior psychiatric history. Note that there are no concerns here about breastfeeding with light therapy, whereas brexanolone and zuranolone have only been studied in women who were not breastfeeding.
Now, let’s look at efficacy. A prestudy power analysis suggested that 16 participants in each group would be sufficient, even with a 20% dropout rate. So, the investigators got pretty close to their recruitment goal. Outcome measures were partly investigator rated—the Hamilton Depression Rating (HAM-D) and Montgomery-Asberg Depression Rating Scales (MADRS)—and partly participant completed—the Edinburgh Postnatal Depression Scale (EDPS). Instructions called for sitting 45 cm from the light box for 45 minutes in the first hour after awakening at a fixed wake-up time.
And the results? Depression scores fell sharply from baseline to week 1 in both the treatment and the placebo groups. By week 3, decreases on the MADRS and the EDPS were significantly greater for the light therapy compared with the placebo group, although not on the HAM-D. Clinically significant? Well, probably. Look at the response rates, meaning patients who experienced a 50% reduction in depression scores: 75% of the light therapy group responded vs 18% of the placebo group on the MADRS. That wasn’t significant on the HAM-D or the EDPS, but clinical significance is somewhat independent, especially in a small study like this.
What about remission, which was defined by a HAM-D of less than 8? Rates were also impressive—67% in the light therapy group vs 18% in the placebo group. Light therapy also had a strong effect on sleep quality as measured by the Pittsburgh Sleep Quality Index, with 25% improvement on the light therapy group vs 4% in the placebo group. The results aren’t perfectly affirmative of efficacy but strongly suggest that you’d see clinically meaningful changes with relatively low risk.
Perhaps the bigger issue is how your patients are going to try this. In our practice, we obtained a small grant and bought 4 light boxes, which are signed out for 2–3 weeks like a library book. That’s long enough to see whether light therapy works for postpartum depression or other conditions. It might even be a sufficient treatment in postpartum depression. This study was only 3 weeks long, but we are not clear what happens after that. If longer treatment is required, patients then have to purchase their own unit, but at least by that time they know it works.
For more on this, the paper’s discussion of circadian malsynchronization as a plausible mechanism of action of light in postpartum depression is of interest. The article is linked here at the Psychopharmacology Institute.
Abstract
Background: Uncertainties and difficulties associated with the current treatment modalities for perinatal depression (PND) may cause some mothers to avoid treatment. Raising awareness about the effectiveness and safety of bright light therapy (BLT) may help to alleviate the challenges of PND. The main goal of this study was to evaluate the efficacy and safety of BLT versus placebo in PND.
Method: A total of 30 women who were either pregnant or in first year postpartum and diagnosed with major depressive disorder were enrolled; 23 completed the study. Patients were randomly assigned to either the BLT (10,000 lux) or placebo (<500 lux) group. BLT and placebo light were applied for 45 min in the morning every day for a 3-week period. The Montgomery-Åsberg Depression Rating Scale (MADRS), Hamilton Depression Rating Scale (HAM-D), and Edinburgh Postnatal Depression Scale (EPDS) were administered weekly to evaluate response and remission rates and depression scores.
Results: There was no significant difference between the two groups in terms of baseline depression scores. At the end of the study, the response rates assessed according to MADRS were 75% for BLT and 18.2% for placebo (p = .006), and remission rates were 41.7% vs. 0% (p = .016), respectively. There was no significant difference between the groups (p > .05) in terms of treatment-related side effects. The main limitation of this study is its small sample size, which limits the generalizability of the study’s findings.
Conclusion: The results indicate that BLT is more effective than placebo and is reliable in terms of side effects in PND patients. In order to expand the use of BLT in PND, new studies with larger sample sizes are needed.
Download PDF and other files
Reference
Donmez, M., Yorguner, N., Kora, K., & Topcuoglu, V. (2022). Efficacy of bright light therapy in perinatal depression: A randomized, double-blind, placebo-controlled study. Journal of Psychiatric Research, 149, 315-322.
