Computed tomography for preoperative shoulder arthroplasty planning: lifetime malignancy risk

Computed tomography (CT) is routinely used for preoperative planning in shoulder arthroplasty. However, this practice exposes patients to ionizing radiation. This study evaluates the potential lifetime cancer risks of conventional and low-dose shoulder CT protocols.

Effective dose, a metric accounting for absorbed radiation, organ sensitivity, and radiation type, was used to evaluate conventional (8.0 mSv) and low-dose (1.4 mSv) shoulder CT protocols. Effective dose was determined using institutional records and literature. The Biological Effects of Ionizing Radiation VII report was used to calculate lifetime attributable risk (LAR) of cancer and number needed to harm (NNH) based on age and sex.

The LAR of malignancy for a shoulder CT of a 50-year-old man was 0.0405% (NNH = 2,472) and 0.0071% (NNH = 14,124) for conventional and low-dose protocols, respectively. By age 80 in men, the NNH was 5,640 and 32,237 for conventional and low-dose protocols, respectively. The LAR of malignancy for a shoulder CT of a 50-year-old woman was 0.0541% (NNH = 1,849) and 0.0095% (NNH = 10,567) for conventional and low-dose protocols, respectively. By age 80 in women, NNH was 5,128 and 29,308 for conventional and low-dose protocols, respectively. There was a positive relationship between patient age and NNH.

The current reliance on preoperative shoulder CT for arthroplasty planning needs to be weighed against the potential lifetime cancer risks, especially among younger women. Risk-reduction strategies such as the widespread adoption of low-dose CT protocols should be considered.