To determine whether high-risk patients with hematuria receive evaluation according to guideline recommendations.
To determine whether high-risk patients with hematuria receive evaluation according to guideline recommendations.
We recently performed a screening study for bladder cancer using a urine-based tumor marker in 1502 subjects at high risk based on aged ≥50 years, ≥10-year smoking history, and/or a 15-year or more environmental exposure. We evaluated use of urinalysis (UA) within 3 years preceding the screening study. Chart review was performed to determine if this subset with microhematuria received any additional evaluation.
Of 1502 study participants, routine urinalysis was performed in 73.2% and 164 (14.9%) subjects had documented hematuria (>3 red blood cells / high-power field) before inclusion. Of these, 42.1% had no further evaluation. Additional testing included repeat urinalysis (36%), urine culture (15.2%), cytology (10.4%), imaging (22.6% overall: 15.9% computed tomography, 4.3% intravenous pyelography; 2.4% magnetic resonance imaging), and cystoscopy (12.8%). Three subjects with microscopic hematuria (2%) were subsequently found to have bladder cancer during the screening study but were not referred for evaluation based on their hematuria. The source of hematuria was unknown in 65%, infection in 22%, benign prostatic enlargement in 10%, and renal stone disease in 4%, but these results are based on incomplete evaluation since only 12.8% underwent cystoscopy.
Subjects at high risk for bladder cancer based on ≥10 years of smoking or environmental exposure with microscopic hematuria are rarely evaluated thoroughly and only 12.8% were referred for urologic evaluation. Further studies are needed to evaluate both the utilization and effectiveness of guidelines for hematuria. Cancer 2010. © 2010 American Cancer Society.
Bladder cancer is the fourth most common cancer in men and fifth most common overall, with an estimated 68,800 new cases of bladder cancer and 14,100 deaths from the disease are expected in 2008 in the United States.1 Most patients are diagnosed after presenting with hematuria. A urinalysis is a common test obtained by primary care physicians and the prevalence of microhematuria is between 9% and 18% in apparently normal individuals.2-4 Approximately, 10% of patients with gross hematuria and 2% to 5% of patients with microscopic hematuria have bladder cancer.2-5
The American Urologic Association best practice policy states that patients at high-risk for bladder cancer (especially those with a history of smoking or chemical exposure) should be considered for a full urologic evaluation after 1 properly performed urinalysis documenting the presence of at least 3 red blood cells (RBCs) per high-power field (HPF).2 An initial finding of microscopic hematuria on urinary dipstick should be confirmed by microscopic evaluation of the urinary sediment.2 Patients of whom a carefully performed history suggests a “benign” cause of their microscopic hematuria should undergo a repeated urinalysis after 48 hours. The evaluation for high-risk patients should include cystoscopy, cytology, and upper tract imaging.6 A separate review of clinical practice recommendations for microhematuria defined microscopic hematuria as 2 or more red cells per HPF on microscopic examination.5 This study in the New England Journal of Medicine recommended a repeat urinalysis for patients and no further evaluation if microscopic hematuria is absent on repeated testing in low-risk patients. If patients have risk factors for bladder cancer such as cigarette smoking or exposure to toxins, then further evaluation with upper tract imaging and cytology and cystoscopy are recommended especially for patients over the age of 50. There are few studies evaluating uptake of these recommendations by primary care physicians.
Two recent studies found suboptimal referral patterns for microscopic hematuria or any hematuria.7, 8 A survey of 788 primary care physicians including internists, family practitioners, primary care physicians, and gynecologists found that only 36% of clinicians routinely refer patients to urologists.7 Similarly, the observed rate of urologic referrals was 27% for women and 47% for men, when a health plan database of 926 patients with hematuria was queried.8 These studies offer evidence for inconsistent referral, but there may be differences between real life practice and response to a survey. Furthermore, the individual patient scenarios may be obscured during database searches. As such, we determined the evaluation for hematuria performed on a cohort of patients at high risk for bladder cancer enrolled in large screening study performed at our institution before inclusion in the screening study. Between March 2006 and November 2007, we screened 1502 subjects for bladder cancer using a urine-based tumor marker, Nuclear Matrix Protein 22 (NMP22).9
This previous screening study solicited subjects from well patient clinics aged ≥50 years, with a 10-year or greater smoking history (any number of cigarettes), and/or a significant (15 or more years) high-risk occupation, such as working in the dye, petroleum, or chemical industry.9 The study excluded those with known cancer or hematuria; therefore, participants were an asymptomatic cohort, who would not otherwise require urologic evaluation and yet were at risk for bladder cancer based on age and smoking and/or environmental factors. We performed a chart review to determine the use of routine urinalysis by primary care physicians and assess what evaluation was performed if hematuria was identified.
All studies were performed with the approval and oversight of the Institutional Review Board for the Protection of Human Subjects. The study population included 1502 participants, including 925 from the Dallas Veterans Affairs Hospital (VA) and 577 from the University of Texas Southwestern (UTSW) recruited into our NMP-22 bladder cancer screening study between March 2006 and November 2007.9 This previous screening study solicited subjects from well patient clinics aged ≥50 years with a 10-year or greater smoking history (any number of cigarettes) and /or a significant (15 or more years) high-risk occupation, such as working in the dye, petroleum, or chemical industry.The study excluded subjects with a history of urologic malignancy, gross hematuria, active urinary tract infection, or urolithiasis. The charts for each participant were reviewed for the 3 years before study entry to determine if there were any evaluations of the urine in the form of urine dipstick testing or urinalysis. Microscopic hematuria was defined as 3 or more red blood cells (RBCs)/HPF. In those with evidence of hematuria, we checked transcripts, progress notes, imaging results, referrals, and laboratory results for subsequent work up. All records at these institutions are electronic medical records. Based on the American Urological Association (AUA) guidelines for evaluation of high-risk patients with hematuria, we defined patients who underwent cystoscopy and upper tract imaging as undergoing guidelines recommendations and patients who did not have complete evaluation as undergoing “guideline inappropriate” evaluation. The patient records at our institution are computerized, allowing easy access to pertinent data.
Univariate and multivariate logistic regression analyses were performed to identify factors associated with further evaluation or referral for cystoscopy. Factors examined included age, sex, tobacco use, number of pack years of tobacco smoked, exposure to occupational/environmental risks, and location of evaluation. All statistical tests were performed with STATA/SE version 10.1 (Stata Corp. LP, College Station, Tex).
There were 1502 participants in the original NMP-22 screening study, including 925 from the VA and 577 from UTSW. Of those, 73.2% had at least 1 urinalysis with or without microscopy during the 3 years before inclusion in the screening study. The proportion of subjects who had a urinalysis was 86.7% at VA hospital and 52.3% at UTSW. Microhematuria was detected in 164 participants (46% UTSW, 54% VA) during the 3 years before inclusion.
Of the 164 participants found to have microhematuria, 69 (42.1%) had no further evaluation (48.3% VA and 34.7% UTSW). Of the remaining 95 participants who had additional testing, 36% had a repeat urinalysis, 15.2% a urine culture, 10.4% cytology, 22.6% received imaging (15.9% computed tomography, 4.3% intravenous pyelography, 2.4% magnetic resonance imaging, and 12.8% received a cystoscopy. Table 1 and Figure 1 (flow chart) show the evaluation and description of subjects included in this study. Of subjects with repeat urinalysis, 28 (47.9%) of 59 were still positive for hematuria.
|Evaluationa||Overall N=164||VA n=89||UTSW n=75|
|No. (%)||No. (%)||No. (%)|
|Repeat UA||59 (36.0)||26 (29.2)||33 (43.8)|
|Urine culture||25 (15.2)||0 (0.0)||25 (33.3)|
|Cytology||17 (10.4)||8 (9.0)||9 (12.0)|
|All imaging||37 (22.6)||27 (30.3)||10 (13.3)|
|CT||26 (15.9)||20 (22.5)||6 (8.0)|
|US||0 (0)||0 (0)||0 (0)|
|IVP||7 (4.3)||5 (5.6)||2 (2.7)|
|MRI||4 (2.4)||2 (2.2)||2 (2.7)|
|Referral to urology/cystoscopy||21 (12.8)||14 (15.7)||7 (9.3)|
Table 2 shows findings from hematuria evaluations performed. Three participants with microscopic hematuria (2%) were subsequently found to have bladder cancer during the screening study but were not referred for evaluation based on the hematuria that predated study inclusion. The source of hematuria was unknown in 65%, infection in 22%, benign prostatic enlargement in 10%, and renal stone disease in 4% but these results are based on incomplete evaluation since only 12.8% underwent cystoscopy.
|Diagnosis||Total N=164||VA n=89||UTSW n=75|
|No. (%)||No. (%)||No. (%)|
|Bladder cancera||3 (2)||3 (3)||—|
|Unknown||107 (65)||57 (64)||50 (67)|
|Urinary tract infection||36 (22)||8 (9)||28 (37)|
|Benign prostatic hypertrophy||17 (10)||10 (11)||7 (9)|
|Renal cyst(s)||9 (5)||8 (9)||1 (1)|
|Renal stone(s)||7 (4)||3 (3)||4 (5)|
|Renal disease||5 (3)||1 (1)||4 (5)|
|Prostate cancer||1 (1)||1 (1)||—|
|Birt-Hogg-Dube||1 (1)||1 (1)||—|
|Urethral stricture||2 (1)||2 (2)||—|
|Ureteral stone(s)||1 (1)||1 (1)||—|
|Bladder diverticulum||1 (1)||—||1 (1)|
Univariate and multivariate analyses were performed to identify factors associated with further evaluation or referral for cystoscopy. When evaluating factors, including age, gender, tobacco use, number of pack years of tobacco, exposure to occupational/environmental risks, and location of evaluation, only tobacco use was associated with a statistically significant decreased likelihood of evaluation. Of the 90 subjects who received some sort of evaluation, 29 (32%) were smokers, while 38 (51%) of subjects who did not get evaluated were smokers (n = 74). Similarly, of the 21 subjects who underwent a cystoscopic evaluation, 6 (29%) were smokers, while 61 (74%) of 143 subjects who did not get a cystoscopy were smokers. On multivariate analysis, in a model including age, gender, occupational exposure, and pack years of tobacco, tobacco use was associated with a nearly 2.5-fold increased risk of no evaluation (odds ratio [OR], 2.43; 95% confidence interval [CI], 1.12-5.26, P = .025) and a nearly 4-fold increased risk of no cystoscopy (OR, 3.84; 95% CI, 1.06-14.3, P = .04).
Microscopic hematuria is a common condition with a prevalence of 9% to 18% of apparently normal individuals.2-4 Mariani et al evaluated 1000 consecutive patients with hematuria and 9.1% had life-threatening lesions but disease was localized in 92% of cases.4 Khadra and colleagues evaluated 1930 hematuria patients and found bladder cancer in 11.9%, including 4.8% and 19.3% of patients with microscopic and gross hematuria, respectively.10 Additional findings included urinary tract infections (13%), medical renal disease (9.8%), urolithiasis (3.6%), renal cancers (0.6%), and prostate cancer (0.4%). Although significant diseases were less common in patients with microscopic hematuria (31.8%) as opposed to gross hematuria (47.5%), it still represented a considerable disease burden because the absolute number of patients with microscopic hematuria exceeds those with gross hematuria. Because of this risk, both the AUA guidelines and the clinical practice recommendations state that patients at high risk for bladder cancer should undergo evaluation with cystoscopy, cytology, and upper tract imaging.5, 6 It is concerning that 87% of subjects in our study did not undergo appropriate evaluation according to guidelines panels.
Our study identified subjects who were selected for bladder cancer screening based on aged ≥50 years and ≥10 year history of smoking and/or ≥15 year environmental exposure. Similar to other studies, we found microscopic hematuria in 10.9% of our subjects with 73.2% having at least 1 urinalysis within 3 years before inclusion in our screening study. The etiology of most urothelial cancers has been associated with increasing age, tobacco exposure, occupational exposures to aromatic amines, and exposures from the chemical and rubber industry.11 Cigarette smoking accounts for a large proportion of the disease in men (50%-75%) and women (14%-35%), while the attributable risk for occupational exposures is only 10%.12, 13 Despite the finding that subjects in our population with microscopic hematuria clearly met the criteria for high risk of bladder cancer, only 12.8% were referred to see a urologist for cystoscopic evaluation. Furthermore, only 10.4% had urine cytologic evaluation and 22.6% had imaging. As such, the vast majority of the population with microscopic hematuria at high risk did not have a complete evaluation. Smokers, who are particularly at increased risk for bladder cancer, were paradoxically less likely to be referred for guideline-appropriate evaluation than nonsmokers. Furthermore, 3 subjects (2%) who were diagnosed with bladder cancer at time of the screening study had hematuria that had not been previously evaluated.
Our study confirms prior reports, which suggest that many patients with microscopic hematuria do not get referred for evaluation by urologists.7, 8 There are no available studies that explore the rationale for referring some patients in favor of others, but our study suggests that at 2 separate hospitals, a similar pattern exists with regard to evaluation (or lack thereof) of microscopic hematuria. Future studies will be necessary to evaluate barriers to appropriate evaluation and to test interventions to improve proper workup of evaluation. Interventions might include education or structured, reflexive protocols when hematuria is noted at laboratory testing. Problems with inconsistent referrals have been blamed for delays in diagnosis of bladder cancer.14 At this time, 25% of bladder cancers are diagnosed with advanced stage and these patients have a significantly lower survival than patients with noninvasive disease. Of note, a greater proportion of women with bladder cancer die from their disease likely secondary to delay in diagnosis.15
One possibility for low referral rates is a perceived low yield of cancer in patients with microscopic hematuria. Because over 60% of patients who are evaluated have no explanation for the microscopic hematuria, some clinicians may believe that there is little benefit to referring patients and ordering expensive imaging.10 Interestingly, there is significant pressure on clinicians to perform other tests of relatively low yield such as evaluation for colon cancer. As part of Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial, a total of 64,658 subjects underwent screening flexible sigmoidoscopy.16 The yields for colorectal cancer per 1000 screened, depending on 5-year age group, was 1.1-2.5 in women and 2.4-5.6 in men. This cancer yield is nearly 10-fold less than the likelihood of bladder cancer in patients with microscopic hematuria, yet screening for colorectal cancer is clearly accepted and promoted by primary physicians. Problems with appropriate evaluation of fecal occult blood are common and there are several studies that have attempted to improve physician management approaches,17-19 including education and structured reflexive protocols; these may serve as a model for improvement of management of individuals with hematuria.
This study has several limitations. Data used for this study were originally collected for clinical, not research purposes, leading to the possibility of misclassification of the presence or absence of hematuria, or subsequent workup. However, the central computerized medical records used for clinical care at both institutions in our study setting make the likelihood of misclassification unlikely. It is a retrospective chart review, but performing a prospective study on compliance to guidelines would be self-defeating because clinicians would be biased to conform to guidelines if they knew they were being studied. Also, subjects with microhematuria, without subsequent work up, may have been aware of microhematuria, concerned about the finding, and perhaps more likely to participate in our screening study. This may have lead to an overestimation of the prevalence of subjects with inappropriate work-up of hematuria. Nonetheless, the raw frequency of inappropriate (n = 143 of 164 subjects) work-up was quite high, emphasizing that even if our prevalence estimates are high, that this is an important clinical problem that deserves further study. Finally, it is not clear why patients were not referred for evaluation, and it is possible that some patients were referred but were noncompliant with recommendations to see a urologist.
Subjects at high risk for bladder cancer based on ≥10 years of smoking or environmental exposure with microscopic hematuria are rarely evaluated thoroughly and only 12.8% were referred for urologic evaluation. Further studies are needed to evaluate both the utilization and effectiveness of guidelines for hematuria.
Samir Gupta was supported by National Institutes of Health grant number 1 KL2 RR024983-01, titled, “North and Central Texas Clinical and Translational Science Initiative” (Milton Packer, M.D., PI) from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH) and NIH Roadmap for Medical Research, and its contents are solely the responsibility of the authors and do not necessarily represent the official view of the NCRR or NIH. Information on NCRR is available at http://www.ncrr.nih.gov/. Information on Re-engineering the Clinical Research Enterprise can be obtained from http://nihroadmap.nih.gov/clinicalresearch/overview-translational.asp.