Missing observations in bone histomorphometry on osteoporosis: implications and suggestions for an approach

Area of Science:

• Bone Biology
• Histomorphometry
• Osteoporosis Research

Background:

• Bone histomorphometry relies on tetracycline double labeling for turnover indices.
• Missing readings in double labels can lead to information loss and selection bias.
• Exclusion of biopsies with missing data may skew results towards high-turnover patients.

Purpose of the Study:

• To analyze the impact of excluding iliac crest bone biopsies with missing readings in women with spinal crush fracture osteoporosis.
• To evaluate the utility of mineral apposition rate (MAR) lower limits for recoding missing data.
• To investigate potential selection bias in bone histomorphometric trials.

Main Methods:

• Analysis of iliac crest bone biopsies from 158 women with spinal crush fracture osteoporosis.
• Examination of two lower limits for MAR to recode missing interlabel width (Ir.L.Wi) values.
• Comparison of histomorphometric indices with and without exclusion of cases with missing readings.

Main Results:

• Excluding biopsies with missing data increased mineralizing surface (MS/BS) by 60%.
• A recoding procedure using a minimal MAR of 0.1 microm/day (Ir.L.Wi = 1 microm) decreased average MAR by 4% and prolonged remodeling period by 19%.
• Biopsy exclusion was more prevalent in older and more osteopenic individuals, suggesting bias.

Conclusions:

• Ignoring missing double labels in bone histomorphometry introduces selection bias, favoring high-turnover patients.
• Recoding procedures are necessary to accurately represent low-turnover patients.
• Exclusion of older, osteopenic patients poses a risk of bias in osteoporosis trials.