How many patients are needed? Variation and design considerations in bone histomorphometry

Area of Science:

• Orthopedics and Bone Biology
• Pharmacology and Drug Development
• Biostatistics and Study Design

Background:

• Bone histomorphometry is vital for assessing drug effects in osteoporosis research.
• Accurate sample size calculation is essential to avoid Type II errors (failing to detect real differences).

Purpose of the Study:

• To determine optimal sample sizes for bone histomorphometric studies.
• To compare the statistical power of single versus paired biopsy designs for detecting a 15% difference between groups.

Main Methods:

• Calculated sample sizes based on variance components from three distinct patient groups.
• Compared statistical power of single biopsy (post-treatment) and paired biopsy (post-treatment minus baseline) designs.
• Utilized a significance level of 0.05 and a Type II error risk of 0.20.

Main Results:

• Mineral apposition rate, wall thickness, and erosion depth require small sample sizes (n=25) in single biopsy designs.
• Bone volume and other surface parameters need larger sample sizes (100-600), but bisphosphonates can reduce this for activation frequency.
• Paired biopsy designs generally require larger sample sizes than single biopsy designs due to higher within-individual variation.

Conclusions:

• Single biopsy designs are more cost-effective for randomized clinical trials in bone histomorphometry.
• Study scaling should be based on the statistical power of specific histomorphometric indices.
• Mineral apposition rate, wall thickness, and erosion depth are powerful indices for detecting treatment effects.