Review question
Do pharmacological treatments, non-invasive brain stimulation (electrodes are placed on the scalp and a finely controlled electric current is applied to change brain activity), psychological treatments, or combination treatments reduce the proportion of people with depression or the extent of depressive symptoms after stroke?
Background
Depression is common after stroke yet often is not detected or is inadequately treated.
Search date
We identified studies by searches conducted on 8 February 2022. This is a living systematic review. We search for new evidence every two months and update the review when we identify relevant new evidence. Please refer to the Cochrane Database of Systematic Reviews for the current status of this review.
Study characteristics
We included trials that reported on the use of pharmacological, non-invasive brain stimulation, psychological, and combined interventions to treat depression after stroke. We described the main outcomes as the number of people meeting the criteria for depression (scoring above a predefined scoring threshold) and inadequate response (scoring below 50% of the predefined scoring threshold). Average age of participants ranged from 54 to 78 years. Studies were from Asia (39), Europe (12), America (6), South America (1) and Australia (3).
Key results
We included 65 trials (72 comparisons) involving 5831 participants. Pharmacological treatments resulted in fewer people meeting the study criteria for depression at end of treatment and with inadequate response to treatment. Non-invasive brain stimulation did not reduce the number of people meeting the study criteria for depression at end of treatment and with inadequate response to treatment. Psychological therapy reduced the number of people meeting the study criteria for depression at end of treatment. The combination of pharmacological treatment and non-invasive brain stimulation resulted in fewer people meeting the study criteria for depression but did not affect those with inadequate response to treatment. More people in the pharmacological treatment group reported central nervous system (e.g. confusion, sedation, tremor; in five trials) and gastrointestinal side effects (e.g. constipation, diarrhoea; in four trials) than in the placebo groups. Information on side effects of other treatments was not provided.
Certainty of the evidence
Estimates of treatment effects were imprecise due to small numbers in most studies and recruitment of people with very different baseline characteristics. We rated the certainty of evidence as low to very low due to these and other limitations in study design.
Conclusion
Antidepressant drugs may benefit people with persistent depressive symptoms after stroke, but care is required in their use, as little is known about their effects on overall stroke recovery. Non-invasive brain stimulation may not be of benefit while psychological and combination therapies may offer a treatment option. Future research should include a broader group of people with stroke.
Very low-certainty evidence suggests that pharmacological, psychological and combination therapies can reduce the prevalence of depression while non-invasive brain stimulation had little to no effect on the prevalence of depression. Pharmacological intervention was associated with adverse events related to the CNS and the gastrointestinal tract. More research is required before recommendations can be made about the routine use of such treatments.
Depression is an important morbidity associated with stroke that impacts on recovery, yet is often undetected or inadequately treated.
To evaluate the benefits and harms of pharmacological intervention, non-invasive brain stimulation, psychological therapy, or combinations of these to treat depression after stroke.
This is a living systematic review. We search for new evidence every two months and update the review when we identify relevant new evidence. Please refer to the Cochrane Database of Systematic Reviews for the current status of this review.
We searched the Specialised Registers of Cochrane Stroke, and Cochrane Depression Anxiety and Neurosis, CENTRAL, MEDLINE, Embase, five other databases, two clinical trials registers, reference lists and conference proceedings (February 2022). We contacted study authors.
Randomised controlled trials (RCTs) comparing: 1) pharmacological interventions with placebo; 2) non-invasive brain stimulation with sham stimulation or usual care; 3) psychological therapy with usual care or attention control; 4) pharmacological intervention and psychological therapy with pharmacological intervention and usual care or attention control; 5) pharmacological intervention and non-invasive brain stimulation with pharmacological intervention and sham stimulation or usual care; 6) non-invasive brain stimulation and psychological therapy versus sham brain stimulation or usual care and psychological therapy; 7) pharmacological intervention and psychological therapy with placebo and psychological therapy; 8) pharmacological intervention and non-invasive brain stimulation with placebo and non-invasive brain stimulation; and 9) non-invasive brain stimulation and psychological therapy versus non-invasive brain stimulation and usual care or attention control, with the intention of treating depression after stroke.
Two review authors independently selected studies, assessed risk of bias, and extracted data from included studies. We calculated mean difference (MD) or standardised mean difference (SMD) for continuous data, and risk ratio (RR) for dichotomous data, with 95% confidence intervals (CIs). We assessed heterogeneity using the I² statistic and certainty of the evidence according to GRADE.
We included 65 trials (72 comparisons) with 5831 participants. Data were available for: 1) 20 comparisons; 2) nine comparisons; 3) 25 comparisons; 4) three comparisons; 5) 14 comparisons; and 6) one comparison. We found no trials for comparisons 7 to 9.
Comparison 1: Pharmacological interventions
Very low-certainty evidence from eight trials suggests pharmacological interventions decreased the number of people meeting the study criteria for depression (RR 0.70, 95% CI 0.55 to 0.88; P = 0.002; 8 RCTs; 1025 participants) at end of treatment and very low-certainty evidence from six trials suggests that pharmacological interventions decreased the number of people with inadequate response to treatment (RR 0.47, 95% CI 0.32 to 0.70; P = 0.0002; 6 RCTs; 511 participants) compared to placebo. More adverse events related to the central nervous system (CNS) (RR 1.55, 95% CI 1.12 to 2.15; P = 0.008; 5 RCTs; 488 participants; very low-certainty evidence) and gastrointestinal system (RR 1.62, 95% CI 1.19 to 2.19; P = 0.002; 4 RCTs; 473 participants; very low-certainty evidence) were noted in the pharmacological intervention than in the placebo group.
Comparison 2: Non-invasive brain stimulation
Very low-certainty evidence from two trials show that non-invasive brain stimulation had little to no effect on the number of people meeting the study criteria for depression (RR 0.67, 95% CI 0.39 to 1.14; P = 0.14; 2 RCTs; 130 participants) and the number of people with inadequate response to treatment (RR 0.84, 95% CI 0.52, 1.37; P = 0.49; 2 RCTs; 130 participants) compared to sham stimulation. Non-invasive brain stimulation resulted in no deaths.
Comparison 3: Psychological therapy
Very low-certainty evidence from six trials suggests that psychological therapy decreased the number of people meeting the study criteria for depression at end of treatment (RR 0.77, 95% CI 0.62 to 0.95; P = 0.01; 521 participants) compared to usual care/attention control. No trials of psychological therapy reported on the outcome inadequate response to treatment. No differences in the number of deaths or adverse events were found in the psychological therapy group compared to the usual care/attention control group.
Comparison 4: Pharmacological interventions with psychological therapy
No trials of this combination reported on the primary outcomes. Combination therapy resulted in no deaths.
Comparison 5: Pharmacological interventions with non-invasive brain stimulation
Non-invasive brain stimulation with pharmacological intervention reduced the number of people meeting study criteria for depression at end of treatment (RR 0.77, 95% CI 0.64 to 0.91; P = 0.002; 3 RCTs; 392 participants; low-certainty evidence) but not the number of people with inadequate response to treatment (RR 0.95, 95% CI 0.69 to 1.30; P = 0.75; 3 RCTs; 392 participants; very low-certainty evidence) compared to pharmacological therapy alone. Very low-certainty evidence from five trials suggest no difference in deaths between this combination therapy (RR 1.06, 95% CI 0.27 to 4.16; P = 0.93; 487 participants) compared to pharmacological therapy intervention and sham stimulation or usual care.
Comparison 6: Non-invasive brain stimulation with psychological therapy
No trials of this combination reported on the primary outcomes.