Does Obesity Affect Pathologic Agreement of Initial and Final Tumor Grade of Disease in Endometrial Cancer Patients?
The objectives of this study were to compare preoperative and postoperative tumor grade to determine if surgical staging decisions for endometrial cancer based on preoperative biopsy are feasible and whether obesity affects the agreement.Methods
A retrospective cohort study of women with endometrial cancer between January 2010 and December 2011 was performed. Demographics, stage of final pathology, biopsy method, preoperative and postoperative tissue grade, and histology were abstracted and stratified by patient body mass index (obese ≥30 kg/m2 and nonobese <30 kg/m2). Patients with incomplete records or uterine sarcoma were excluded. The agreement between preoperative and postoperative tumor grade for all patients and in obese and nonobese patients was determined using weighted κ statistics.Results
Four hindered forty-five patients were included: 161 nonobese patients and 284 obese patients. The proportion of preoperative sampling via office biopsy and dilation and curettage was similar in each cohort. Overall, the agreement between preoperative and postoperative pathology was only fair (weighted κ = 0.21). Stratified by body mass index, the agreement between preoperative and postoperative grade remains fair in obese and slight in nonobese patients (weighted κ = 0.21 and 0.19, respectively). Substantial increases in tumor grade from preoperative to postoperative pathologic specimens occurred in both cohorts.Conclusions
Obesity does not appear to significantly alter the correlation between preoperative biopsy and final tumor grade. With only fair correlation between preoperative and postoperative pathologic evaluation, utilization of preoperative biopsy pathology results as a triage tool for surgical staging should be avoided. However, the discordance between preoperative and postoperative pathology in favor of a higher grade on final pathology in both groups may cause some surgeons to favor staging.