Article Text
PDF
Abstract
Introduction/Background The global COVID-19 impact surpasses 7.2 trillion dollars spent by May 2022, causing the postponement of 28 million surgeries and a surge in waiting lists. The UK and Brazil saw a 41.5% reduction in elective surgeries. Beyond immediate health crises, worsened conditions, prognosis, and financial impacts ensued. Cancer diagnoses globally dropped 40%-76%, affecting endometrial cancer prognosis. Pandemic-exacerbated structural issues led to a 20% decline in endometrial cancer diagnoses and a 43% reduction in precursor lesions in 2020. Brazil initiates a pioneering study to create a risk model for timely endometrial cancer diagnosis amid persistent hysteroscopy waiting lists, recognizing its crucial impact.
Methodology A case-control study at Hospital in Sao Paulo involved 2103 women undergoing hysteroscopy (March 2019 - 2022) for suspected endometrial cancer. Retrospective data extraction included demographic details, medical history, and ultrasound findings.
Results The study, encompassing 1945 primarily post-menopausal participants, identified 107 cases of endometrial cancer. Common comorbidities included hypertension (39.59%) and diabetes (17.3%), with 30% experiencing postmenopausal bleeding. Significant associations were observed with factors such as hypertension, diabetes, post-menopausal bleeding, and endometrial polyps. Multivariate linear regression revealed protective effects from the absence of high blood pressure, diabetes, and postmenopausal bleeding, reducing risks by 22.02%, 49.4%, and 55.51%, respectively. Increased pregnancies provided protective benefits (12.81% risk reduction per pregnancy), while higher BMI and age increased risks. The logistic regression model achieved a 75.7% AUC, with a cutoff adjustment enhancing sensitivity to 83.0%. Nomograms were developed for risk stratification, optimizing predictive capabilities in this ethically approved retrospective study utilizing medical records and R Studio analyses.
Conclusion A study with 1945 participants identified 107 cases of endometrial cancer. Factors like hypertension, diabetes, and postmenopausal bleeding showed significant associations. Protective effects were observed with absence of these conditions, and a logistic regression model achieved a 75.7% AUC. Nomograms were developed for risk stratification.
Disclosures .
