Efficacy of anti‐inflammatory treatment on major depressive disorder or depressive symptoms: meta‐analysis of clinical trials

Abstract

Efficacy of anti-inflammatory treatment on major depressive disorder or depressive symptoms: meta-analysis of clinical trials K€ ohler-Forsberg O, Nicolaisen Lydholm C, Hjorthøj C, Nordentoft C, Mors O, Benros ME. Efficacy of anti-inflammatory treatment on major depressive disorder or depressive symptoms: meta-analysis of clinical trials Background: No study has gathered evidence from all randomized clinical trials (RCTs) with anti-inflammatory drugs measuring antidepressant effects including a detailed assessment of side-effects and bias. Methods: We performed a systematic review identifying RCTs published prior to January 1, 2018, studying antidepressant treatment effects and sideeffects of pharmacological anti-inflammatory intervention in adults with major depressive disorder (MDD) or depressive symptoms. Outcomes were depression scores after treatment, remission, response, and side-effects. Pooled standard mean differences (SMD) and risk ratios (RR) including 95% confidence intervals (95%-CI) were calculated. Results: We identified 36 RCTs, whereof 13 investigated NSAIDs (N = 4214), 9 cytokine inhibitors (N = 3345), seven statins (N = 1576), 3 minocycline (N = 151), 2 pioglitazone (N = 77), and 2 glucocorticoids (N = 59). Anti-inflammatory agents improved depressive symptoms compared to placebo as add-on in patients with MDD (SMD = À0.64; 95%-CI = À0.88, À0.40; I 2 = 51%; N = 597) and as monotherapy (SMD = À0.41; 95%-CI = À0.60, À0.22; I 2 = 93%, N = 8825). Antiinflammatory add-on improved response (RR = 1.76; 95%-CI = 1.44-2.16; I 2 = 16%; N = 341) and remission (RR = 2.14; 95%-CI = 1.03-4.48; I 2 = 57%; N = 270). We found a trend toward an increased risk for infections, and all studies showed high risk of bias. Conclusion: Anti-inflammatory agents improved antidepressant treatment effects. Future RCTs need to include longer follow-up, identify optimal doses and subgroups of patients that can benefit from anti-inflammatory intervention. Summations• Among 36 RCTs including almost 10 000 patients, five out of six anti-inflammatory drugs improved depression scores compared to placebo.• Short-term (i.e., weeks) anti-inflammatory add-on to antidepressants showed improved antidepressant effects in MDD without increasing the risk for side-effects.• The effect size for anti-inflammatory add-on to antidepressants was similar to the effect size of antidepressants alone, indicating the potential clinical benefit. Considerations• Few studies reported on side-effects and all studies were associated with risk of bias.• Few studies investigated biomarkers and none explored specific depressive symptoms to predict better response to anti-inflammatory treatment.• Hence, future studies need to include longer follow-up and measures of inflammation (blood, CSF, and brain scans) to identify subgroups of patients with specific depressive symptoms who could be offered anti-inflammatory treatment.

Treatment of depression with anti‐inflammatory medications—the role of pain relief

“…In the discussion of limitations of their meta-analysis, they briefly mention that 'although the antidepressant effects of NSAIDs have been shown to be independent of their painrelieving effects, the antidepressant effect of anti-inflammatory agents may be mediated via their effect on somatic diseases'. Kohler-Forsberg et al (1) use the study by Iyengar et al (2) as supportive evidence of their claim that the NSAIDs antidepressant effects have been shown to be independent of their pain relieving effects (1, p 412). However, the Iyengar et al study (2) (which itself is a pooled analysis of 5 postapproval trials) does not support this claim.…”

“…In their meta-analysis, Kohler-Forsberg et al (1) analyzed the efficacy or rather effect of anti-inflammatory treatment on major depressive disorder or depressive symptoms. In the discussion of limitations of their meta-analysis, they briefly mention that 'although the antidepressant effects of NSAIDs have been shown to be independent of their painrelieving effects, the antidepressant effect of anti-inflammatory agents may be mediated via their effect on somatic diseases'.…”

“…Nevertheless, the impact that pain reduction had on mood elevation cannot be easily discarded. One can simply ask patients in his/her practice or patients in several studies in the meta-analysis by Kohler-Forsberg et al (1) (several studies included patients with significant comorbidity) whether their improvement in mood and fatigue correlates with pain relief.…”

Steroids and antidepressant response

“…The recent meta-analytic review of trials involving anti-inflammatory agents to treat or to supplement treatment of patients with major depressive disorder (MDD) or depressive symptoms by K€ ohler-Forsberg et al (1) includes data derived from 36 controlled trials. Treatments included glucocorticoids as well as cytokine inhibitors, minocycline, non-steroidal antiinflammatory drugs, pioglitazone, and statins.…”

“…Potentially relevant to this work are efforts in the 1980s to evaluate the ability of the dexamethasone suppression test (DST) to predict responsiveness to treatment of MDD subjects with various antidepressants. Since non-suppression of cortisol output by dexamethasone indicates the presence of elevated circulating corticosteroid concentrations, one might expect to find superior treatment responses among cortisol non-suppressors (DST-positive), based on the proposal that various antiinflammatory agents, including glucocorticosteroids, might enhance responses to antidepressants (1). The extensive body of research on the DST, though heterogeneous and including some findings of an association of cortisol non-suppression and favorable response to some antidepressants, led to the overall conclusion that initial DST status had little predictive association with antidepressant response (2)(3)(4)(5).…”