Review article

National Practice Recommendations for Hematuria:

How to Evaluate in the Absence of Strong Evidence?
Ronald Loo, MD
Joel Whittaker, MPH
Violeta Rabrenivich, MHA, CHIE
Introduction
Abstract Background
Hematuria is one of the most common conditions
Quality improvement requires physicians to system-
confronting clinical urologists and is present in many
atically explore new scientific evidence, integrate this
genitourinary pathology conditions. Although research-
information into practice, and evaluate their perfor-
ers have studied hematuria symptoms in an effort to
mance. In addition, Kaiser Permanente (KP) clinicians
determine the best diagnostic pathway, the existing
need to effectively leverage our integrated delivery
lack of scientific evidence has created variations
system in providing preventive care, improving early
in clinical practice. The literature does not provide
detection, and managing complex clinical conditions.
enough evidence to significantly alter the need to as-
Daily, clinicians face situations that require a com-
sess these patients. Consequently, many patients with
prehensive understanding of the complex variables
microscopic or gross hematuria undergo low-yield
involved in patient care, aggressive decision making,
workups that include invasive testing and imaging
and prioritization of work and of resources.
with radiation. In 2007, a national group of Kaiser
The Interregional Chiefs of Urology Service (IRCUS)
Permanente (KP) urology chiefs agreed that national
is one of many KP groups that embrace quality-
practice recommendations were needed to address
improvement methods and activities. With support
existing variations in the management and workup
from the regional Clinical Practice Guidelines team, the
of hematuria. Using a KP guideline methodology, the
national KP HealthConnect team, and The Permanente
group reached a consensus agreement on the follow-
Federation, these clinicians have elected to work on
ing recommendations: 1) referral to urology is recom-
several areas of focus as a national quality-improvement
mended for all people with gross hematuria or high-
agenda. On the basis of an identified need and many
grade hematuria (>50 red blood cells per high-power
years of clinical practice, the group decided to focus
field [RBCs/HPF]) on a single urinalysis (UA); 2) referral
on a standardized hematuria evaluation.
to urology and urologic evaluation is recommended
for men or women with asymptomatic microscopic
Lack of Scientific Evidence
hematuria or symptomatic hematuria that produces
Adult microhematuria is an example of a clinical
>3 RBCs/HPF on two of three properly performed and
symptom for which the lack of scientific evidence has
collected urinalyses; and 3) voided urinary cytology
created variations in clinical practice. Hematuria is one
should be eliminated from asymptomatic hematuria
of the most common conditions confronting clinical
screening protocol. The test is not sensitive enough
urologists and is present in a number of genitourinary
to obviate further workup if findings are negative, and
pathology conditions. According to KP experts, it is
elimination of this screening test is estimated to save
estimated to account for 20% of all urologic visits and
millions of dollars across the US. Hematuria on a UA
up to 13.9% of urologic hospitalizations.
should be reported as 0 to 3 RBC/HPF, 4 to 10 RBC/
Similar efforts to address hematuria symptoms
HPF, 11 to 25 RBC/HPF, 26 to 50 RBC/HPF, >50 RBC/
have been initiated by professional associations and
HPF, or gross hematuria. This approach will also reduce
individual clinicians. In 2001, the American Urological
radiation exposure.
Association (AUA) convened the Best Practice Policy

Ronald Loo, MD, is Regional Chief of Urology for the Southern California Permanente Medical
Group and Chair of the Kaiser Permanente Interregional Urology Chiefs. E-mail: ronald.k.loo@kp.org.
Joel Whittaker, MPH, is a Senior Consultant for the Technology Assessment and Guideline Unit for
the Southern California Permanente Medical Group. E-mail: joel.l.whittaker@kp.org.
Violeta Rabrenovich, MHA, CHIE, is Director of Medical Group Performance Improvement for The
Permanente Federation, LLC in Oakland, CA. E-mail: violeta.rabrenovich@kp.org.

The Permanente Journal/ Winter 2009/ Volume 13 No. 1 37

REVIEW article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

Table 1. The Kaiser Permanente Interregional management and workup (Table 2). Evaluation of
Chiefs of Urology Service recommendations issued by the AUA and input were
Colorado obtained both from national and regional KP Guideline
Edward Swartz, MD Development Units.
Georgia Although the AUA recommendations represented a
David Starr, MD consensus statement of urologists from across the US,
Group Health Permanente; Seattle, WA our clinicians believed the KP guideline methodology
Marc A Lowe, MD (Common Methodology),2 developed by interregional
Hawaii guideline experts, to be more rigorous. (In accordance
Howard Landa, MD with KP’s Common Guideline Methodology, consensus-
Albert Mariani, MD based recommendations are developed when an
Michelle Aspera, MD
important clinical question needs to be answered and
Ohio
the evidence is insufficient to support evidence-based
Nabil Chehade, MD
recommendations.) The group believed that the existing
Northern California
situation warranted development of national practice
Gary Nicolaisen, MD
recommendations. In addition, they believed that KP,
Northwest
Stephen Lieberman, MD
Southern California
Table 2. Core clinical questions and evidence
Ron Loo, MD
search strategy
James Murphy, MD
Marguerite Koster Core clinical questions
Thomas Vandergast, MD 1. For patients with microhematuria, what threshold
Joel Whittaker, MPH of red blood cells per high power field (RBC/HPF) is
The Permanente Federation associated with a sufficient probability of urologic
Jed Weissberg, MD pathology to warrant a referral to urology?
Violeta Rabrenovich, MHA, CHIE 2. How should hematuria be reported on the
urinalysis?
3. What is the role of urine cytology and/or bladder
Panel on Asymptomatic Microscopic Hematuria to tumor markers in the detection of urologic
formulate policy statements and recommendations pathology among patients with hematuria?
for the evaluation of asymptomatic microhematuria in 4. For patients with hematuria, what imaging
tests (ie, CT urogram, modified CT urogram,
adults.1 As a result of these efforts, the AUA recom-
intravenous pyelogram, helical CT, and/or renal
mended that an appropriate renal or urologic evalua- ultrasonography) should be employed for the
tion be performed for all patients with asymptomatic detection of urologic cancers?
microscopic hematuria who are at risk for urologic 5. Is routine urinalysis screening effective for reducing
disease or primary renal disease; however, there was no urologic pathology in the asymptomatic population?
consensus on when to test for microscopic hematuria To answer the group’s clinical questions, a literature
search included the following databases and specialty
in the primary care setting, and screening was not ad- sites:
dressed in this report. • Kaiser Permanente Clinical Library
In addition, the current literature does not provide • Clinical Evidence via OVID
enough evidence to significantly alter the need to assess • PUBMED
these patients. Consequently, many patients undergo • Hayes
low-yield workups that include invasive testing and • Blue Cross
imaging with radiation. • Blue Shield TEC, Health Tech Assessment database
• Southern California Permanente Medical Group
Medical Tech Assessment Database
Methodology • Turning Research Into Practice database (Bandolier,
IRCUS is a multidisciplinary group (Table 1) that Agency for Healthcare Research and Quality,
New Zealand Guideline Group, Monash,
works to ensure that KP provides safe, effective, and National Institute for Clinical Excellence, Scottish
high-quality care; to reduce practice variation; and to Intercollegiate Guideline Network)
create organizational improvement in urologic care. • American Urological Association
In 2007, they sponsored a review of the literature to • American College of Radiology
address core clinical questions relating to hematuria CT = computed tomography

38 The Permanente Journal/ Winter 2009/ Volume 13 No. 1

 Review article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

as an integrated delivery system, offers a unique op- Results

portunity to manage patient care across care settings Kaiser Permanente National
in a more effective way. Finally, as with many other Practice Resource
specialties, KP Urology Departments have a long and After reviewing the literature, the group concluded
outstanding history of research and quality-improve- that evidence relating to the diagnostic follow-up care
ment efforts that can be leveraged. of hematuria was insufficient for developing “evidence-
This initiative was sponsored by the Associate based” practice recommendations. Despite insufficient
Executive Medical Directors—Quality, Southern evidence, the clinicians agreed that consensus-based
California Permanente Medical Group Technology national practice recommendations were nonethe-
Assessment and Guideline Unit, and The Permanente less warranted to reduce the variation in hematuria
Federation. management.

Figure 1. Adult hematuria workup algorithm.

C + S = culture and sensitivity; e-gfr = estimated glomerular filtration rate; HPF = high-power field; rbc = red blood cells; IVP = intravenous pyelogram; KUB =
kidneys, ureter, bladder; NSAIDs = nonsteroidal anti-inflammatory drugs; RUS = renal ultrasound; U/A = urinalysis.
1
Urine specimens should be collected >48 hours after exercise. The U/A should be analyzed fresh if possible, by a standardized methodology to avoid the lysus of
formed elements from heat for chemical breakdown.
2
After urologic evaluation is completed, re-referral for persistent microhematuria is not needed unless there is a change in clinical situation, such as the occur-
rence of gross hematuria or another sign or symptom suggestive of possible urologic pathology.
3
CT Urogram is defined as a two-phase study (noncontrast followed by postcontrast delay) and KUB reconstruction. When IVP is ordered, clinicians should take
into consideration patient history of chronic illness (diabetes, heart failure, and other comorbidities), as well as a patient being on certain medications (metformin,
NSAIDS, and others).
4
Patients receiving contrast should have a serum e-gfr testing performed prior to the procedure.

The Permanente Journal/ Winter 2009/ Volume 13 No. 1 39

REVIEW article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

The Standardized Hematuria Evaluation Practice which identified 18 studies evaluating hematuria and
Resource3 describes the evidence and steps for screen- the risk of urologic disease. Eleven of the 18 studies
ing adult patients, the making of risk assessments, and did not provide the data needed to evaluate urologic
summarizes suggested diagnostic follow-up treatment. cancer rates at specific cutoff points below 8 RBC/
The goals of this work—in the face of insufficient HPF. Among the seven remaining studies where cutoff
evidence—are to standardize and optimize a proper points of >2 to 5 RBC/HPF were used to define mi-
workup for patients with hematuria and to minimize ra- crohematuria, urologic or renal cancers were detected
diation exposure from unnecessary testing among those 1.3% to 8.3% of the time among patients who were
patients unlikely to have serious disease (Figure 1). older than 33 years.5–7 The review also examined
An additional goal is to provide clinicians with laboratory case series studies to determine the upper
adequate background and resources to increase their limit of normal (95th percentile) among “healthy”
comfort in evaluating patients with asymptomatic patients receiving microscopic UA. These studies
microscopic hematuria. This practice resource is not show a strong trend toward “normal” limits, ranging
intended to replace a clinician’s judgment or to establish from 0 to 2 RBC/HPF in men and from 0 to 5 RBC/
a protocol for all patients with this clinical issue. HPF in women. It is important to note that although
The clinicians agreed on and supported the following the normal limits varied between men and women,
consensus-based recommendations. the studies did not provide information regarding the
actual presence or absence of urologic disease in the
Recommendation 1 populations studied. Ultimately, the SCPMG review
Referral to urology is recommended for all patients concluded that there is insufficient evidence to deter-
with gross hematuria or high-grade hematuria (>50 mine the “optimal” RBC/HPF cutoff point for detecting
RBC/HPF) on a single urinalysis (UA). clinically significant asymptomatic microhematuria. A
Referral to urology and urologic evaluation is recom- subsequent literature search was conducted in 2007 to
mended for men or women with asymptomatic micro- update the 2003 SCPMG review. No additional system-
scopic hematuria or symptomatic hematuria (unilateral atic reviews, meta-analyses, or randomized, controlled
flank pain, lower irritative voiding symptoms, recurrent trials (RCTs) were identified.
urinary tract infections despite appropriate use of antibiot- In the absence of high-quality RCTs or systematic
ics, etc) that produces >3 RBC/HPF on two of three prop- reviews, the Interregional Urology Chiefs Group agreed,
erly performed and collected UAs. (Note: Urine specimens on a consensus basis, that referral to urology for further
should be collected >48 hours after exercise. The UA diagnostic workup is recommended for asymptomatic
should also done when the urine is fresh if possible, by a patients whose microscopic UA yields >3 RBC/HPF on
standardized methodology, to avoid the lysis of formed two of three properly urinalyses, regardless of patient
elements from heat or chemical breakdown.) sex. The chiefs also agreed that urine samples should
Evidence review and rationale: A review of the be collected after avoiding strenuous physical exercise
evidence identified one evidence review by Southern for >48 hours to avoid glomerular or urothelial exer-
California Permanente Medical Group (SCPMG),4 cise hematuria; urine should also be analyzed fresh
if possible, by a standardized methodology, to avoid
the lysis of formed elements from heat or chemical
National recommendations
breakdown. The chiefs recommend evaluating three
1. Referral to urology is recommended for all patients with
urine specimens because of evidence from one study
gross hematuria or high-grade hematuria (>50 RBC/HPF)
on a single urinalysis (UA). showing that 18% of patients with a life-threatening
2. Hematuria on a UA should be reported out as 0 to 3 RBC/ lesion had negative findings on at least one UA within
HPF, 4 to 10 RBC/HPF, 11 to 25 RBC/HPF, 26 to 50 RBC/ six months of the diagnosis.8
HPF, >50 RBC/HPF, or gross hematuria. Hematuria evaluation: This evaluation should not
3. There was no consensus on the role of urine cytology and/ be performed if the risk of the testing exceeds the risk
or bladder tumor markers in the evaluation of patients with
hematuria.
of the medical condition that is diagnosed. Thus, if the
4. A modified computed tomography (CT) urogram or IVP
life-threatening risk of a hematuria evaluation (instru-
with concurrent renal ultrasound is recommended for mentation urosepsis, contrast anaphylaxis, radiation
patients with significant hematuria. risk, contrast nephropathy) is greater than the yield
5. There is insufficient evidence to recommend routine UA of the evaluation for a defined population, then the
to screen for asymptomatic hematuria in the absence of evaluation should not be performed.
clinical indicators.

40 The Permanente Journal/ Winter 2009/ Volume 13 No. 1

 Review article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

Follow-up care: Hematuria is likely to persist in the ogy and/or bladder tumor markers in the evaluation
majority of patients who are monitored. The evidence of patients with hematuria.
for the risk of cancer developing within two to five years Evidence review and rationale: A literature
in patients with hematuria who have been evaluated is search was conducted to identify studies evaluating
scanty, but it is in the range of 0% to 3%. the effectiveness of urine cytology and bladder tumor
A retrospective study (which did not distinguish markers for the detection of urologic cancer among
between gross hematuria and microhematuria) of patients with hematuria. Several systematic reviews
823 patients who did not have a malignancy and were identified.
whose medical charts were available found that the One systematic review11–13 identified 15 studies
average follow-up monitoring was 14.7 years. Using evaluating urine cytology and NMP22 BladderChek Test
intravenous pyelogram (IVP) as the lone imaging (Matritech, Inc, Newton, MA, USA) as tests for detecting
modality, transitional cell carcinoma (TCC) was found urinary tract malignancy. Pooled data from these stud-
in 7 of 740 (0.95%) patients at an average of 14.3 ies showed that sensitivity for the urine cytology test
years after diagnosis (range, 5.3–23.9 years). Renal ranged from 3% to 100%, whereas specificity ranged
cell carcinoma (RCC) developed in 5 of 740 (0.68%) from 62% to 100%. The review also found inconsistent
of patients at an average of 15.7 years after diagnosis data from five heterogeneous studies regarding the
(range, 2.5–23.2 years). Overall, 1.5% of the evaluated sensitivity (58%–91%) and specificity (60%–84%) of the
patients developed TCC or RCC. Gross hematuria or NMP22 BladderChek Test for the detection of urologic
a smoking history was present in 77%. Although the disease. The authors caution against drawing definitive
data on which to base recommendations for hematuria conclusions, given that the studies included were het-
follow-up care is limited, consideration may be given erogeneous, methodologically flawed, and
to reevaluating any patient with gross hematuria or subject to potential bias. Ultimately, the study
These studies
persistent microhematuria and a smoking history at authors agreed with the AUA statement that
show a strong
two to five years.9 the available data are insufficient to recom-
trend toward
mend routine use of voided urinary markers
“normal”
Recommendation 2 in patients with microscopic hematuria.
limits ranging
Hematuria on a UA should be reported out as 0 to 3 In 2005, the KP Southern California Tech-
from 0 to 2
RBC/HPF, 4 to 10 RBC/HPF, 11 to 25 RBC/HPF, 26 to nology Assessment and Guidelines Team
RBC/HPF in
50 RBC/HPF, >50 RBC/HPF, or gross hematuria. reviewed the literature for the use of the
males and
Evidence review and rationale: Literature sources NMP22 BladderChek Test to detect primary or
from 0 to 5
that specifically compared the effect of reporting UA recurrent TCC of the urinary tract. No RCTs
4
RBC/HPF in
results according to varied cutoff points were not identi- were identified. They did, however, find 22
females.
fied. One large prospective study10 of 1000 consecutive uncontrolled studies evaluating the accuracy
patients with asymptomatic hematuria found that the of the test. The NMP22 BladderChek Test had
incidence of urologic pathology was greater for people a sensitivity ranging from 30% to 100%, specificity of
with high-grade hematuria vs low-grade microhema- 60% to 90%, and positive predictive value (PPV) of 34%
turia; no difference between low (4–10 RBC/HPF) and to 76%. The team concluded that the sensitivity of the
intermediate grades of hematuria was found. NMP22 BladderChek Test suggests that it may help to
To gain a better understanding of how hematuria detect low-grade primary carcinomas, but the specificity
correlates with the presence or absence of urologic and PPV of the NMP22 BladderChek Test also suggest
disease, IRCUS agreed to standardize the reporting of that the test would result in an increased number of
UA results according to the following cutoffs: 0 to 3 unnecessary cystoscopic procedures. However, most
RBC/HPF, 4 to 10 RBC/HPF, 11 to 25 RBC/HPF, 26 to 50 results highlight increased specificity and sensitivity.14
RBC/HPF, >50 RBC/HPF, or gross hematuria. Examina- Another systematic review15 pooled data from 42
tion of UA data that are reported in a standard fashion studies (n = 5706) and compared the diagnostic accu-
may provide insight to clinicians on how to best to racy of urine cytology vs other tests (BTA [Polymedco,
stratify hematuria workups on the basis of the yield of Inc, Cortlandt Manor, NY, USA], BTA stat [Polymedco,
urologic disease in each category. Inc], BTA TRAK [Polymedco, Inc], telomerase, or NMP22
BladderChek Test) against the reference standard of
Recommendation 3 cystoscopy and/or histopathology. They found that
There was no consensus on the role of urine cytol- cytology had a pooled specificity of 94%, which was

The Permanente Journal/ Winter 2009/ Volume 13 No. 1 41

REVIEW article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

significantly higher than for the other tests evaluated in The KP Hawaii Region did an analysis of the current
the study. The authors also stated that none of the tests well-established bladder tumor markers. A frank malig-
evaluated in the studies reached levels of sensitivity that nancy reading for urine cytology (cost, $60.25) had a 41%
are acceptable in lieu of cystoscopy for clinical practice. sensitivity but a 97.2% specificity in 17 studies encom-
In addition, 22 of 42 studies used a case-control design, passing 4,685 patients. In four BTA (cost, $98.00) studies
which provides greater potential for bias. encompassing 455 patients, there was a 78% sensitivity
A cross-sectional study14 (n = 668) compared the NMP22 and a 80% specificity. In five NMP22 BladderChek Test
BladderChek Test with urine cytology and with reference (cost, $15.50) studies encompassing 846 patients, there
standard (cystoscopy and pathology findings) for detection was a 80% sensitivity and a 77% specificity. In a study of
of recurrent bladder cancer. They found that the NMP22 the FISH test encompassing 456 patients, Sarosdy et al17
BladderChek Test had a sensitivity of approximately 49% found a sensitivity of 68% and a specificity of 80%.
and specificity ranging from 83% to 91%. The degree to For a clinical test to be useful, it must change what
which these findings can be applied to primary prevention the clinician does. A specificity of 97.2% (2.8% false
in populations with bladder cancer is unclear. positive rate) for cytology would likely cause a urolo-
The purpose of a bladder tumor marker is to increase gist to have a lower threshold for ordering a biopsy of
the clinician’s index of suspicion for TCC of the urinary indeterminate bladder or prostatic urethral lesions and
tract. Questions have been raised about the appropriate- might prompt ureteroscopy.
ness of urine cytology as part of a hematuria study. This
was studied from the KP Hawaii Hematuria 1000-patient Recommendation 4
hematuria database. We found a sensitivity of 55% and A modified computed tomography (CT) urogram or
a specificity of 99.3%. Unique information that led to IVP with concurrent renal ultrasound is recommended for
a diagnosis of urinary tract TCC was found in four pa- patients with significant hematuria (as already defined).
tients. The cost to diagnose a cancer by this test and no As long as the renal ultrasound is done concurrently
other (unique information) in the hematuria evaluation with IVP, there is no need for renal tomography. This
was $8367 vs $5616 for IVP, $3235 for cystoscopy, and approach will reduce radiation exposure (Table 3). One
$3291 for creatinine. The cost of the test to diagnose a caveat: the radiation exposure associated with the modi-
life-threatening lesion (in support of other tests whose fied CT urogram has been reported to be 12 to 24 times
findings might also have made the diagnosis) was $1521 higher than with IVP. The modified CT urogram should
for cytology, $1695 for IVP, $3044 for cystoscopy, and be conducted with a protocol capable of visualizing
$3291 for creatinine. This study supported the use of any collecting-system lesions using the lowest radiation
urine cytology in that it diagnosed TCC not diagnosed dose possible. Patients receiving contrast should have
by other tests, and the cost of the test was comparable a serum estimated glomerular filtration rate (eGFR) test
to other well-established costs.15,16 performed before the procedure. When IVP is ordered,
clinicians should take into consideration the patient’s
history of chronic illness (diabetes, heart failure, and
Table 3. Summary of radiation exposure by other comorbidities), as well as whether the patient
imaging test takes certain medications (metformin, nonsteroidal
Imaging (CPT code) Millisieverts (mSv) anti-inflammatory drugs, and others).
Intravenous urogram or 1.6 Evidence review and rationale: A review of the
intravenous pyelogram (76497)
literature was conducted to identify studies that evaluate
Renal and bladder ultrasound 0
(76775) the effectiveness of CT urogram and/or IVP for detect-
KUB (radiograph plain film) 0.07 ing urologic disease. A complementary search of the
(74000) literature was also conducted to identify studies that
CT without contrast, abdomen 10 compare the relative differences in radiation exposure
and pelvis (74150, 72192) that may exist between the two imaging modalities.
CT with contrast, abdomen and 14 One systematic review and two cohort studies (reports
pelvis (74160, 72193)
about which were published subsequently to the sys-
CT with and without contrast, 24
abdomen and pelvis (74170,
tematic review) were identified. A brief summary of
72194) this evidence is provided below.
CT = computed tomography; CPT = current procedural terminology; One systematic review by Rogers et al11 identified
KUB = kidneys, ureter, bladder three studies evaluating the use of the CT urogram to

42 The Permanente Journal/ Winter 2009/ Volume 13 No. 1

 Review article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

identify any abnormality that may cause hematuria. not performed in the same patients and that the in-
According to that review: creased radiation exposure provided by two tests could
One study combined CT with IVP as the reference not be justified.
standard and reported a sensitivity of 100% and a The two cohorts were unmatched, but analysis
specificity of 97%. indicated that there were no statistically significant
A second study used histopathology as the refer- differences in patient characteristics.
ence standard and reported a sensitivity of 92% and a Patients were not stratified by risk of
specificity of 94%. disease, and the authors believe that this … the radiation
A third study evaluated the CT as a method to detect contributed to the relatively low overall exposure
filling defects or strictures in the urinary tract and re- detection rate. associated with
ported a sensitivity of 82% and a specificity of 97%. Rogers et al11 also identified seven non- the modified
The authors concluded that there is some evidence randomized studies evaluating IVP (also CT urogram has
to support the use of CT to determine the cause of known as intravenous urography, or IVU) been reported
hematuria. However, they also reported that the evi- as an index test for the detection of urologic to be 12 to 24
dence base is limited evidence to three diagnostic ac- cancer among people with hematuria. They times higher
curacy studies, one of which was poorly reported and reported the following results: than with IVP.
not designed for the purpose of detecting significant “Seven studies evaluated IVU as an index
urologic pathology. test … . [Four] studies evaluated IVU against
Turney et al18 conducted a cohort study (n = 200) final diagnosis, but for different target conditions: upper
comparing CT urogram findings with those for cystos- urinary tract tumors (sensitivity 89%, specificity 95%),
copy and pathology to determine the diagnostic accu- lower tract tumors (sensitivity 56%, specificity 98%),
racy of CT urography (CTU) for detection of bladder any upper tract pathology (sensitivity 67%, specificity
cancer. They reported a sensitivity of 93%, a specificity 91%), any renal abnormality (sensitivity 90%, specificity
of 98%, a PPV of 98%, and a negative predictive value 98%) or any filling defect or structure in the urinary
of 97%. In this publication’s introduction, the study tract (sensitivity 68%, specificity 98%). Across the IVU
authors claimed that CTU is becoming recognized as studies, specificity values (range 91%-100%) appeared
the diagnostic tool of choice for many urologic condi- to be more consistent than sensitivity values (range
tions and represents the “gold standard” for examining 55%-90%), although it is difficult to estimate the overall
upper urinary tracts. This explicit bias suggests caution value of IVU as a test owing to the clinical and statistical
when interpreting the study results. heterogeneity between studies.”11
Another nonrandomized cohort study (n = 512), Radiation exposure: Several studies evaluating the
conducted by Albani et al,19 examined the diagnostic radiation exposure levels from CT urography and IVP
accuracy of CTU vs IVP in adults with hematuria. among adults with hematuria and flank pain (suspected
For the identification of upper tract lesions, CTU had renal colic) were identified (Kim et al,20 Homer et al,21
a sensitivity of 94% and a PPV of 89%, whereas IVP Thomson et al,22 and others23–30). The data suggest that
had a sensitivity of 50% and a PPV of 40%. Owing to radiation may be higher for noncontrast CT (range,
the lack of a gold-standard examination for upper tract 1.4–10.0 millisieverts [mSv]) and noncontrast helical CT
evaluation, specificity could not be calculated. (range, 2.806–5.004 mSv) than for IVP (range, 1.48–4.46
For the identification of lower urinary tract lesions, mSv). With CT, exposures were consistently higher for
CTU had a sensitivity of 40% and specificity of 93%, women than for men. (Table 3 provides a summary of
whereas IVP had a sensitivity of 37% and a specificity radiation exposures by imaging test.)
of 97%. Both imaging modalities failed to detect more Studies that explicitly evaluated the health impact
than 60% of bladder lesions smaller than 2 cm. of different levels of radiation exposure from the CT
The overall detection rates were 25.5% for CTU and urogram versus IVP among patients with hematuria
19.4% for IVP. were not identified (summary of average doses from
Methodologic issues: The authors identified two American College of Radiology and Radiological Society
cohorts and included in the analysis only those study of North America).31
subjects who could make the required follow-up visits. There is no clear consensus that CTU is superior to
The effect of this design in reaching definitive conclu- IVP for a hematuria evaluation; however, there is emerg-
sions is uncertain. ing evidence that this may be the case. Although radia-
The authors acknowledged that CTU and IVP were tion exposures are higher for CTU than for IVP, newer

The Permanente Journal/ Winter 2009/ Volume 13 No. 1 43

REVIEW article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

CT protocols and technologic advances are reducing Recommendation 5

radiation dose while increasing the anatomic detail of There is insufficient evidence to recommend routine
images in addition to identifying pathology in other UA to screen for asymptomatic hematuria in the absence
organ systems that would not be noted on IVP. The of clinical indicators.
interregional urology chiefs agreed, on a consensus Evidence review and rationale: Hematuria screen-
basis, that CTU can be used to evaluate patients with ing for cancer in the asymptomatic population has
significant hematuria according to a protocol capable not been clinically established. RCTs and high-quality
of visualizing collecting-system lesions using the low- epidemiologic studies supporting the use of routine
est radiation dose possible. Patients receiving contrast UA screening among asymptomatic adults are lacking.
CT should have a serum creatinine test performed Furthermore, in a 2006 report, the US Preventive Ser-
before the procedure. Alternatively, a concurrent IVP vices Task Force recommended against routine bladder
and renal ultrasound would also provide acceptable screening among asymptomatic persons.34,35
imaging of significant renal masses and collecting- Routine screening for bladder cancer with urine
system lesions with less radiation but less standardiza- dipstick, microscopic UA, or urine cytology is not
tion (more operator-dependent) (Mariani AJ, personal recommended in asymptomatic persons. All patients
communication, 2007 May 19).a,32 who smoke tobacco should be routinely counseled to
The interregional urology chiefs also took the fol- quit smoking.
lowing into consideration: In patients without significant urologic symptoms,
A CT of the abdomen and pelvis with and without microscopic hematuria is occasionally detected on
contrast exposes the patient to about 20 times the ra- routine UA. At present, routine screening of adults for
diation dose of an IVP. (Note: renal ultrasound has no microscopic hematuria with UA is not recommended
It has been associated radiation exposure.) because of the intermittent occurrence of this finding
estimated An IVP will detect only 10% of 1-cm lesions and 52% and the low incidence of significant associated urologic
that … a of 2- to 3-cm lesions. Fortunately <4% of renal masses disease.
significant that are <3 cm in size will behave malignantly, even
number though ≤90% are RCCs. This may be the reason why Discussion
of future IVP served urology as well as it did for so long as the Implementation: Collaboration and Tools
cancers will standard imaging for a hematuria evaluation. Implementation of a nationwide adult asymptomatic
be caused by CT is superior to renal ultrasound for the detection of microhematuria screening and management program
iatrogenic small renal masses, but renal ultrasound detected 100% is a result of the collaborative efforts of clinicians
unnecessary of lesions >2.5 cm and the majority of lesions >1.5 cm representing multiple areas of medical care: urology,
imaging.39 in one well-designed study.33 Again, most small lesions primary care, radiology, and laboratory. Many other
do not behave malignantly. departments—including guideline development, re-
Fine-cut CT images can approach the collecting- gional continuing medical education, regional labo-
system detail of an IVP and provide additional func- ratory, and national and regional KP HealthConnect
tional information. On IVP, tumors present as negative implementation teams—were consulted in both the
filling defects (as do clots and radiolucent stones). On planning and implementation.
CT, a tumor will usually opacify after contrast, and a Standardized implementation tools were developed:
radiolucent stone is easily distinguished from a blood • A hematuria-management standardized presentation
clot. Renal ultrasound can also easily distinguish a clot that is used during chiefs’ Departments of Urology,
from a radiolucent stone. Primary Care, Radiology, and other meetings to
The cost of a CT urogram (~$282) would be ap- provide an overview of the recommendations and
proximately the same as the cost of an IVP plus a renal to educate clinicians and staff
ultrasound (~$228 + ~$87 = ~$315), according to data • Hematuria practice resource pocket cards to assist
from KP Hawaii Region 2007 (Mariani AJ, personal clinicians during patient care
communication, 2007 May 19).a • Continuing medical education materials
CT scans account for 70% of all medical x-ray ex- • KP HealthConnect hematuria diagnosis SmartSet
posure even though they represent 20% of diagnostic list
imaging studies. It is estimated that a single dose of 10 • Standardization of reporting of hematuria by labora-
mSv (<1 CT scan) has a lifetime cancer risk of 1/1000 tory departments
and a death rate of 1/2000. • Implementation and adherence to national practice

44 The Permanente Journal/ Winter 2009/ Volume 13 No. 1

 Review article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

recommendations will be evaluated in the future, a

Albert J Mariani, MD, Associate Medical Director of Spe-
and the lessons that this provides will be used to cialty and Hospital Services, Hawaii Permanente Medical
Group.
modify practice recommendations, provide feedback
to clinicians, and support ongoing performance
Disclosure Statement
improvement efforts. The author(s) have no conflicts of interest to disclose.

Eliminate One-Quarter of Future Workups Acknowledgment

In an effort to minimize variations in reporting and Katharine O’Moore-Klopf, ELS, of KOK Edit provided editorial
to collect definitive evidence to completely eliminate assistance.
the need to assess the lowest-risk patients with hema-
References
turia in the near future, a KP HealthConnect SmartSet
1. Grossfeld GD, Wolf JS Jr, Litwan MS, et al. Asymptomatic
list data collection tool has been developed to allow microscopic hematuria in adults: summary of the AUA best
concurrent electronic data analysis of hematuria workup practice policy recommendations. Am Fam Physician 2001
outcomes. It will be used nationwide by KP urolo- Mar 15;63(6):1145–54.
gists to document their workup findings; we estimate 2. Evidence-based clinical guideline development (“Common
Methodology”) [monograph on the Intranet]. In: National
that within one year, enough data will be captured to
Guideline Initiative: Policies and procedures. Oakland (CA):
eliminate the need to assess 25% or more of the pa- Kaiser Permanente Care Management Institute; 2002
tients currently being evaluated. A valuable outcome Jun, updated 2008 Apr [cited 2008 Nov 24]. Available
goal is to demonstrate the power and capability of KP from: http://dms.kp.org/docushare/dsweb/Get/Document-
HealthConnect in population-based clinical research. 751827/3_P_P_Common_Methodology.doc. (Password
protected.)
To reinforce continuity of care, the KP HealthConnect
3. Kaiser Permanente Interregional Chiefs of Urology Service
tool will generate patient-care instructions for further
Group. Interregional practice resource: standardized
follow-up treatment. hematuria evaluation, December 2007 [monograph on
the Intranet]. Oakland (CA): Kaiser Permanente: 2007 Dec
High Radiation Risk [cited 2008 Nov 24]. Available from: http://cl.kp.org/pkc/na-
It has been estimated that acute radiation exposures tional/cmi/programs/practice_resources/hematuria_practice
resource.htm. (Password protected.)
as low as 10 mSv pose significant cancer risk, so much
4. Kaiser Permanente Southern California Medical Technology
so that a significant number of future cancers will be Assessment [monograph on the Intranet]. Pasadena (CA):
caused by iatrogenic unnecessary imaging.36 Southern California Permanente Medical Group; 2000 Jun;
updated 2002 Dec 10; 2005 Mar (cited 2008 Nov 24].
Conclusions Available from: http://cl.kp.org/pkc/national/ref/ebwebsite/
evidence_synopses/urol_op_microhematuria_nl.htm. (Pass-
More than 62 million CT scans are performed an-
word protected.)
nually in the US, a large number of which are due to
5. Mariani AJ, Luangphinith S, Loo S, Scottolini A, Hodges
screening and assessment of asymptomatic patients CV. Dipstick chemical urinalysis: an accurate cost-effective
with microhematuria. Clearly, major efforts to curtail screening test. J Urol 1984 Jul;132(1):64–6.
unnecessary radiation exposure are sorely needed. As 6. Mariani AJ. The evaluation of adult hematuria: a clinical
advocated by the KP IRCUS, an immediate reduction update. In: AUA Update Series; Houston (TX): AUA Office of
Education; 1998;17(Lesson 24):186–91.
in radiation exposure by collectively switching to a
7. Mariani AJ. Hematuria evaluation [letter]. J Urol 2001
modified CT urogram and a commitment to support
Jul;166(1):545.
the collection of evidence through KP HealthConnect to 8. Hofland CA, Mariani AJ, Sutherland RS. The long-term inci-
completely eliminate unnecessary workup underscore dence of transitional cell carcinoma and renal cell carcinoma
our dedication to the KP promise.37 in patients who have undergone a negative hematuria evalu-
The KP National Hematuria Guideline has been a ation. Proceedings of the 51st Annual James C Kimbrough
Urological Seminar; 2004 Jan 11-16; San Antonio, TX.
tremendous inspiration to all participants. The work
9. Carcinoma detected by a standardized hematuria evaluation
illustrates the potential that KP possesses in effecting
[abstract]. Paper presented at the 99th Annual Meeting of
safer and more reliable evidence-based care as well as the American Urological Association; 2004 May 8–13; San
its obligation as a health care leader in striving to an- Francisco, CA.
swer previously unanswerable questions and to change 10. Mariani AJ, Mariani MC, Macchioni C, Stams UK, Hariharan
the way that medicine is practiced for our patients and A, Moreira A. The significance of adult hematuria: 1000
hematuria evaluations including a risk-benefit and cost-
worldwide. v
effectiveness analysis. J Urol 1989 Feb;141(2):350–5.
11. Rodgers M, Nixon J, Hempel S, et al. Diagnostic tests and al-

The Permanente Journal/ Winter 2009/ Volume 13 No. 1 45

REVIEW article
National Practice Recommendations for Hematuria: How to Evaluate in the Absence of Strong Evidence?

gorithms used in the investigation of haematuria: systematic 25. Martin DR, Semelka RC. Health effects of ionizing radiation
reviews and economic evaluation. Health Technol Assess from diagnostic CT. Lancet 2007 May 27;367(9524):1712–
2006 Jun;10(18). 4. Erratum in: Lancet 2006 Oct 28;368(9546):1494.
12. Hofland CA, Mariana AJ. Long-term survival of patients 26. Liu W, Esler SJ, Kenny BJ, Goh RH, Rainbow AJ, Stevenson
with bladder transitional cell carcinoma detected by a stan- GW. Low-dose nonenhanced helical CT of renal colic:
dardized hematuria evaluation. Paper presented at the 99th assessment of ureter stone detection and measurement of
Annual Meeting of the American Urological Association effective dose equivalent. Radiology 2000 Apr;215(1):51–4.
2004 May 8-15; San Francisco, CA. 27. Mendelson RM, Arnold-Reed DE, KuanM, et al. Renal
13. Hofland CA, Mariani AJ. Is cytology required for a hematu- colic: a prospective evaluation of non-enhanced spiral CT
ria evaluation? J Urol 2004 Jan;71(1):324–6. versus intravenous pyelography. Australas Radiol 2003
14. Grossman HB, Soloway M, Messing E, et al. Surveillance for Mar;47(1):22–8.
recurrent bladder cancer using a point-of-care proteomic 28. Smith RC, Rosenfield AT, Chloe KA, et al. Acute flank pain:
assay. JAMA 2006 Jan 18;295(3):299–305. comparison of non-contrast-enhanced CT and intravenous
15. Glas AS, Roos D, Deutekom M, Zwinderman AH, Bossuyt urography. Radiology 1995 Mar;194(3):789–94.
PM, Kurth KH. Tumor markers in the diagnosis of pri- 29. Tack D, Sourtzis S, Delpierre I, De Maetelaer V, Gevenois
mary bladder cancer. A systematic review. J Urol 2003 PA. Low-dose unenhanced multidetector CT of patients
Jun;169(6):1975–82. with suspected renal colic. AJR Am J Roentgenol 2003
16. Gilbert J. CMI Study. Comparison of sensitivities and Feb;180(2):305–11.
specificities for malignancy detection in hematuria tests 30. Van Beers BE, Dechambre S, Hulcelle P, Materne R, Jamart J.
[monograph on the Intranet]. Oakland, CA: Kaiser Perma- Value of multislice helical CT scans and maximum-intensity-
nente Care Management Institute; 2007 Feb 21 [cited 2008 projection images to improve detection of ureteral stones
Nov 24]. Available from: http://cl.kp.org/pkc/national/ref/ at abdominal radiography. AJR Am J Roentgenol 2001
ebwebsite/specialty_index.html. (Password protected.) Nov;177(5):1117–21.
17. Sarosdy MF, Schellhammer P, Bokinsky G, et al. Clinical 31. Safety: Radiation exposure in x-ray examinations [mono-
evaluation of a multi-target fluorescent in situ hybridiza- graph on the Internet]. Oakbrook (IL): Radiological Society
tion assay for detection of bladder cancer. J Urol 2002 of North America; 2008 [cited 2008 Nov 24]. Available
Nov;168(5):1950–4. from: www.radiologyinfo.org/en/safety/index.cfm?pg=sfty_
18. Turney BW, Willatt JM, Nixon D, Crew JP, Cowan NC. xray&bhcp=1.
Computed tomography urography for diagnosing bladder 32. CMI Evidence Synopsis—Hematuria, CT urogram: radiation
cancer. BJU Int 2006 Aug;98(2):345–8. exposure for selected diagnostic procedures [monograph
19. Albani JM, Ciaschini MW, Streem SB, Herts BR, Angermeier on the Intranet]. Oakland (CA): Kaiser Permanente Care
KW. The role of computerized tomographic urogra- Management Institute; 2005 Dec 29 [cited 2008 Nov 24].
phy in the initial evaluation of hematuria. J Urol 2007 Available from: http://cl.kp.org/pkc/national/ref/eb/website/
Feb;177(2):644–8. evidence_synopses/urol_op_hema_radexpos_nl.htm. (Pass-
20. Kim BS, Hwang IK, Choi YW, et al. Low-dose and standard- word protected.)
dose unenhanced helical computed tomography for the 33. Jamis-Dow CA, Choyke PL, Jennings SB, et al. Small (≤ 3cm)
assessment of acute renal colic: prospective comparative renal masses: detection with CT versus US and pathologic
study. Acta Radiol 2005 Nov;46(7):756–63. correlation. Radiology 1996 Mar;198(3):785–8.
21. Homer JA, Davies-Payne DL, Peddinti BS. Randomized 34. Screening for Bladder Cancer, Topic Page [monograph on
prospective comparison of non-contrast enhanced helical the Internet]. Rockville, MD; US Preventive Services Task
computed tomography and intravenous urography in the Force. Agency for Healthcare Research and Quality; 2004
diagnosis of acute ureteric colic. Australas Radiol 2001 Jun [cited 2008 Nov 24]. Available from: www.ahrq.gov/
Aug;45(3):285–90. clinic/uspstf/uspsblad.htm
22. Thomson JM, Glocer J, Abbott C, Maling TM, Mark S. 35. Barrett PH, Okawa G, Bowman J. Advanced/policy track—
Computed tomography versus intravenous urography in evidence-based guidelines. Perm J 2005 Spring;9(2):83–8.
diagnosis of acute flank pain from urolithiasis: a random- 36. Brenner DJ, Doll R, Goodhead DT, et al. Cancer risks at-
ized study comparing imaging costs and radiation dose. tributable to low doses of ionizing radiation: assessing
Australas Radiol 2001 Aug;45(3):291–7. what we really know. Proc Natl Acad Sci U S A 2003 Nov
23. Diel J, Perlmutter S, Venkataramanan N, Mueller R, Lane MJ, 25;100(24):13761–6.
Katz DS. Unenhanced helical CT using increased pitch for 37. Brenner DJ, Hall EJ. Computed tomography—an increas-
suspected renal colic: an effective technique for radiation ing source of radiation exposure. N Engl J Med 2007 Nov
dose reduction? J Comput Assist Tomogr 2000 Sep- 29;357(22):2277–84.
Oct;24(5):795–801.
24. Hamm M, Knopfle E, Wartenberg S, Wawroschek F,
Weckermann D, Harzmann R. Low dose unenhanced helical
computerized tomography for the evaluation of acute flank
pain. J Urol 2002 Apr;167(4):1687–91.

46 The Permanente Journal/ Winter 2009/ Volume 13 No. 1