Risk factors for persistent pain after breast cancer surgery: a multicentre prospective cohort study
Howe‐Siang Tan, Jennifer K. Plichta, Amanda L. Kong, Chin Wen Tan, Sun Wook Hwang, Rehena Sultana, Melanie C. Wright, Alex Tiong Heng Sia, Ban Leong Sng, Ashraf S. Habib
Abstract
Identifying factors associated with persistent pain after breast cancer surgery may facilitate risk stratification and individualised management. Single-population studies have limited generalisability as socio-economic and genetic factors contribute to persistent pain development. Therefore, this prospective multicentre cohort study aimed to develop a predictive model from a sample of Asian and American women. We enrolled women undergoing elective breast cancer surgery at KK Women's and Children's Hospital and Duke University Medical Center. Pre-operative patient and clinical characteristics and EQ-5D-3L health status were recorded. Pain catastrophising scale; central sensitisation inventory; coping strategies questionnaire-revised; brief symptom inventory-18; perceived stress scale; mechanical temporal summation; and pressure-pain threshold assessments were performed. Persistent pain was defined as pain score ≥ 3 or pain affecting activities of daily living 4 months after surgery. Univariate associations were generated using generalised estimating equations. Enrolment site was forced into the multivariable model, and risk factors with p < 0.2 in univariate analyses were considered for backwards selection. Of 210 patients, 135 (64.3%) developed persistent pain. The multivariable model attained AUC = 0.807, with five independent associations: age (OR 0.85 95%CI 0.74-0.98 per 5 years); diabetes (OR 4.68, 95%CI 1.03-21.22); pre-operative pain score at sites other than the breast (OR 1.48, 95%CI 1.11-1.96); previous mastitis (OR 4.90, 95%CI 1.31-18.34); and perceived stress scale (OR 1.35, 95%CI 1.01-1.80 per 5 points), after adjusting for: enrolment site; pre-operative pain score at the breast; pre-operative overall pain score at rest; postoperative non-steroidal anti-inflammatory drug use; and pain catastrophising scale. Future research should validate this model and evaluate pre-emptive interventions to reduce persistent pain risk.