Estimating the sample size of sham-controlled randomized controlled trials using existing evidence

Area of Science:

• Clinical Trial Design
• Medical Interventions
• Systematic Reviews

Background:

• Traditional randomized controlled trials (RCTs) often determine sample size based on achievable participant numbers.
• An alternative is planning trials based on existing evidence, particularly relevant for sham-controlled RCTs (sham-RCTs).
• Renal sympathetic denervation for hypertension is examined within this context.

Purpose of the Study:

• To explore sham-RCT design using sequential meta-analysis and conditional planning.
• To evaluate renal sympathetic denervation for hypertension efficacy.
• To assess the impact of evidence-based conditional sample size calculations.

Main Methods:

• Systematic review of six sham-RCTs (981 participants) on renal sympathetic denervation for hypertension.
• Sequential meta-analysis to determine early rejection points for the null hypothesis.
• Fixed-effect meta-analysis for evidence-based conditional sample size calculations.

Main Results:

• Six sham-RCTs were identified, with the first trial being significantly larger than subsequent ones.
• Individual trial sample sizes were based on overestimated intervention effects, leading to low power.
• Sequential meta-analysis of the first four trials provided strong evidence against the null hypothesis (mean difference -2.75).
• Conditional planning suggested larger sample sizes than those used, potentially due to overoptimistic effect expectations and time-effect associations.

Conclusions:

• Sequential meta-analysis allows for earlier conclusions in sham-RCTs, minimizing patient exposure to sham procedures.
• Conditional planning for new sham-RCTs is challenging due to anticipated improvements in surgical techniques and increasing intervention effects over time.
• Failure to account for these factors in conditional planning may not improve sham-RCT design.