Internal validation and decision curve analysis of a preoperative nomogram predicting a postoperative complication in pheochromocytoma surgery: An international study.

OBJECTIVES

To develop a preoperative nomogram that would predict the risk of a postoperative complication for pheochromocytoma patients undergoing adrenalectomy using an international database.

METHODS

We retrospectively analyzed preoperative variables and postoperative outcomes in patients who underwent adrenalectomy for pheochromocytoma in three institutions from 2000 to 2017. Internal validation of a generated nomogram was carried out with receiver operating characteristics, calibration plots, and decision curve analyses.

RESULTS

A total of 153 patients who had undergone 166 adrenalectomies were included in the study. Overall, post-adrenalectomy complications were seen in 30% of patients, whereas 9.6% of patients sustained a Clavien ≥3a complication. Independent predictors of a complication were a history of hypertension, body mass index, tumor size, and Charlson Comorbidity Index score. On internal validation, the multivariable model generated a nomogram that predicted a postoperative complication or clinically hemodynamic event with an area under the curve of 0.86, showed good calibration and had an overall net benefit.

CONCLUSIONS

An internally validated nomogram combining body mass index, Charlson Comorbidity Index score and tumor size can predict the probability of a post-adrenalectomy complication in those with and without hypertension. The model, the first of its kind in pheochromocytoma surgery, identifies patients at risk of a postoperative complication at the time of their presentation with pheochromocytoma.