Chronic Post-Surgical Pain at 3 Months After Breast Cancer Surgery: Incidence, Predictors, and an Internally Validated Risk Model.

Background/

Objectives

Chronic post-surgical pain (CPSP) affects 25-60% of women after breast cancer surgery. Prospective data from the Middle East and North Africa (MENA) region remain scarce. This study estimated 3-month CPSP incidence in a Moroccan cohort, developed two internally validated prediction models, and quantified the mediated effects of regional anaesthesia and ketamine through prevention of severe acute postoperative pain.

Methods

Prospective cohort of 862 women undergoing curative breast cancer surgery (Institut National d'Oncologie, Rabat; NCT04676438). CPSP was defined as a visual analogue scale (VAS) ≥ 3/10 AND Brief Pain Inventory (BPI) interference ≥ 3/10 at three months. Two logistic regression models were developed (Model A: 7 preoperative variables; Model B: 11 perioperative variables) and internally validated by 1000-bootstrap optimism correction. Counterfactual mediation analysis quantified effects mediated through severe acute pain.

Results

CPSP incidence was 39.1% (95% confidence interval (CI) 35.9-42.4%), with 22.6% neuropathic-predominant. Model A showed limited discrimination (area under the curve (AUC) 0.58), with pain catastrophising as the dominant predictor. Model B demonstrated superior performance (AUC 0.694; Brier score 0.209) with net clinical benefit across the 15-55% range of clinically plausible decision thresholds. In counterfactual mediation analysis, an estimated 58.2% and 53.7% of the RA and ketamine effects, respectively, were mediated through severe acute pain, under the framework's no-unmeasured-confounding assumption. Perioperative ketoprofen independently reduced chronification risk (adjusted odds ratio (aOR) 0.66).

Conclusions

In this single-centre Moroccan cohort, 39% of women developed CPSP at three months. Two validated models enable risk stratification at distinct clinical decision points. Regional anaesthesia and ketamine were associated with lower CPSP risk, in a pattern consistent with mediation through prevention of severe acute pain under the model's assumptions; ketoprofen was independently associated with lower chronification risk. These observational findings are hypothesis-generating and require confirmation before informing changes to analgesic protocols.