Summary of findings 2. Lithium compared to valproate for acute mania.
| Lithium compared to valproate for acute mania | ||||||
| Patient or population: acute mania Setting: inpatients and outpatients attending mood disorders clinics Intervention: lithium Comparison: valproate | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with valproate | Risk with lithium | |||||
|
Efficacy: response (categorical) YMRS/SADS‐C decrease ≥ 50% by end of study |
Study population | OR 1.22 (0.87 to 1.70) | 607 (5 RCTs) | ⊕⊕⊕⊝ Moderate1 | ||
| 438 per 1000 | 487 per 1000 (404 to 569) | |||||
|
Efficacy: response (continuous) Change in YMRS (ITT‐LOCF) from baseline to end of study |
The mean efficacy: response (continuous) change in YMRS (ITT‐LOCF) from baseline to end of study in the valproate group ranged from −23.8 to −7.4 | The mean efficacy: response (continuous) change in YMRS (ITT‐LOCF) from baseline to end of study in the lithium group ranged from −23.55 to −6.1 | ‐ | 398 (5 RCTs) | ⊕⊝⊝⊝ Very low2,3 | |
|
Efficacy: remission (categorical) YMRS ≤ 12 and no increase in MADRS at end of study |
Study population | OR 0.78 (0.46 to 1.32) | 257 (1 RCT) | ⊕⊕⊕⊝ Moderate4 | ||
| 713 per 1000 | 660 per 1000 (533 to 766) | |||||
| Acceptability: total withdrawals | Study population | OR 1.20 (0.86 to 1.69) | 629 (5 RCTs) | ⊕⊕⊕⊕ High | ||
| 330 per 1000 | 372 per 1000 (298 to 455) | |||||
| Adverse event: tremor | Study population | OR 10.51 (1.96 to 56.48) | 449 (2 RCTs) | ⊕⊕⊕⊕ High5,6 | ||
| 4 per 1000 | 44 per 1000 (9 to 200) | |||||
| Adverse event: somnolence | Study population | OR 0.47 (0.29 to 0.76) | 575 (4 RCTs) | ⊕⊕⊕⊕ High | ||
| 221 per 1000 | 118 per 1000 (76 to 177) | |||||
| Adverse event: nausea | Study population | OR 1.53 (0.97 to 2.40) | 583 (4 RCTs) | ⊕⊕⊕⊝ Moderate7 | ||
| 158 per 1000 | 222 per 1000 (154 to 310) | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; ITT: intention‐to‐treat; LOCF: last observation carried forward; MADRS: Montgomery‐Åsberg Depression Rating Scale; OR: odds ratio; RCT: randomised controlled trial; SADS‐C: Schedule for Affective Disorders and Schizophrenia‐change; YMRS: Young Mania Rating Scale | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
1Downgraded one level due to imprecision: two small studies with very wide confidence intervals; overall result is precise which is probably an overestimate of true precision. 2Downgarded two levels due to imprecision and inconsistency. Six studies, three missing standard deviation, which could not be imputed. Three studies giving highly variable results. Overall estimate is therefore much more consistent than heterogeneity of studies should suggest. 3Downgraded for imprecision. Three studies all with differing, imprecise results. 4Downgraded one level due to suspicion of reporting bias. Single study reported this outcome when multiple other larger studies will have had this data available; strongly suggests few participants met remission criteria by end of study. 5Both studies have very wide confidence intervals. 6Likely publication bias, but not downgraded as the included studies are of high quality. Four large studies examined this question but did not report this common adverse event with lithium. 7Downgraded one level due to imprecision. All values fairly imprecise, especially Kowatch 2000, may be overestimating true effect.