The Microhematuria Nomogram: a simplified bladder cancer risk assessment tool
Microhematuria (MH) is a common laboratory finding that can be a sign of occult GU pathology. Currently, all patients with MH are recommended to undergo a cystoscopy and GU imaging which is associated with low diagnosis rates at a high cost to the health system. Identifying patients who are more likely to harbor pathology and focusing evaluation resources in a risk-stratified manner may reduce patient morbidity, lower costs, and improve diagnostic yield.
Using a multi-hospital data warehouse, all patients with a new diagnosis of MH were identified from 2010-2017. Separate training and validation cohorts were created via 1:1 randomization. The effects of clinical and demographic data on a bladder cancer diagnosis were evaluated via multivariate logistic regression to the training set and then fit to the validation set. Comparison of AUC values for each cohort was assessed by using the chi-squared test. A nomogram predicting risk of a bladder cancer diagnosis was developed based on the training set of this model in patients who underwent a complete MH evaluation.
In total, 52,321 patients were included and split among the training and validation cohorts. There were no differences between the two sets with regard to demographic or clinical factors. ROC AUC for the training and validation cohorts was 0.77 and 0.81, respectively, with no significant difference between the two (p=0.34) indicating adequate fit (Figure 1). In the training cohort, age, sex, race, smoking status, and RBC/hpf on initial UA were all significantly related to a bladder cancer diagnosis. After adjusting for all covariates, only age and RBC/hpf on UA were significantly related to a bladder cancer diagnosis and a nomogram was created using these factors (Figure 2).
A nomogram predicting risk of bladder cancer diagnosis in patients with a new diagnosis of microhematuria was created. After further validation, this may be used as a shared decision making tool in patients diagnosed with MH in deciding the risks and benefits of pursuing cystoscopy and GU imaging.
Funding: 2016-2017 AAMC Clinical Care Innovation Pilot Award