Can systematic reviews with sparse data be trusted?

Afshari, Arash; Wetterslev, Jørn; Smith, A. F.

doi:10.1111/anae.13730

Cited by 49 publications

(38 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…dexamethasone after single‐shot spinal anaesthesia, we set out to collate the available studies on this topic and perform a meta‐analysis. Moreover, we used trial sequential analysis to control for the risk of spurious findings, a problem which is not only associated with single randomised controlled trials but also with meta‐analyses . This statistical approach adjusts for the risks of false positive and false negative findings, thereby increasing the validity of the results.…”

Section: Introductionmentioning

confidence: 99%

Effect of intravenous dexamethasone on postoperative pain after spinal anaesthesia – a systematic review with meta‐analysis and trial sequential analysis

et al. 2019

View full text Add to dashboard Cite

There are data suggesting that intravenous dexamethasone has an effect on postoperative analgesia when given during single-shot spinal anaesthesia. However, the research literature is equivocal. We performed a systematic literature search followed by conventional meta-analysis (random effects model). We used trial sequential analysis to control for type-1 and -2 statistical errors. We also performed a leave-one-out metaanalysis for our primary outcome, the consumption of intravenous morphine in the first 24 postoperative hours. We applied the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system to rate the level of evidence. We obtained data from 1133 patients, reported in 17 trials. Reporting quality was high, with low risk of bias. Dexamethasone use was associated with a significant reduction in 24-h morphine consumption, the mean difference (95%CI) being À4.01 (À5.01 to À3.01) mg, 6 trials, 326 participants, I 2 = 0%. Trial sequential analysis showed that there was firm evidence for the primary outcome, and leave-one-out metaanalysis showed that our result was not driven by one single trial. The GRADE evaluation showed a high level of evidence, suggesting that further studies are unlikely to alter the result. The time to first analgesic request (95% CI) was significantly prolonged by 86.62 (10.62-162.62) min, I 2 = 93%, in the dexamethasone group. For other secondary outcomes including number of patients requiring rescue analgesia, or visual analogue scale pain scores, we found no evidence of a significant difference between the treatment arms. We report a high level of evidence that intravenous dexamethasone improves postoperative analgesia after spinal anaesthesia.

show abstract

Section: Introductionmentioning

confidence: 99%

Effect of intravenous dexamethasone on postoperative pain after spinal anaesthesia – a systematic review with meta‐analysis and trial sequential analysis

et al. 2019

View full text Add to dashboard Cite

show abstract

“…We should also remember that the systematic review, much vaunted as the highest level of clinical research evidence, has a retrospective character too. While it is prospective in intent, and the process of preparation contains many elements to guard against some of the possible biases that can bedevil it, the systematic review is at its heart a backward‐looking, observational exercise whose subjects are not patients, but rather other researchers' trials .…”

Section: Retrospective Studiesmentioning

confidence: 99%

Major trauma and the need for massive transfusion

Smith

Choi

2017

Anaesthesia

Self Cite

View full text Add to dashboard Cite

“…On the other hand, one of the key advantages of a systematic review and meta‐analysis is the increase in power and precision of intervention effects. Systematic reviews are, however, prone to multiple testing due to multiple outcomes, multiple intervention groups, sub‐group analyses, or repeated testing over time, which may happen with regular updating . Trial sequential analysis has been suggested as a method for dealing with the increased likelihood of chance findings after repeated analyses .…”

Section: Primary Vs Secondary Outcomesmentioning

confidence: 99%