Summary

Imaging is assuming an increasing role at all levels of health care, including primary health care because of many reasons. The public is more aware of the various available technologies, their expectations are high, newer and better imaging products are available, and the imaging facilities are accessible to more people. However, due to the mirage of different imaging interventions, as family physicians, we have an unshakable role to guide our patients to make the appropriate choices for their individualised needs, balancing benefit with potential harm and also considering other factors like cost-effectiveness and accuracy. 

In view of the above issues and the relative inadequate availability of local imaging guidelines for primary care setting, our team searched and reviewed a few overseas and local imaging and clinical guidelines and developed our own statements of recommendation for the presenting problem of haematuria.

After appropriate history taking and physical examination, blood tests and urinalysis are performed. If glomerular bleeding with proteinuria or renal insufficiency is present, the patient should be referred to a nephrologist for evaluation.  If a non-glomerular source of bleeding is suspected, computed tomography urography can be performed as a first-line investigation. If it is unavailable or considered too expensive, ultrasound or intravenous urography can be done, with the understanding that further imaging may be needed. If the above imaging is unrevealing, cystoscopy is recommended for patients over 40 years old, in those with risk factors for urologic cancer or if persistent macroscopic haematuria is present. For those under 40, it would be performed only if urine cytology is positive or suspicious.  The patient should be referred to a urologist if suspicious lesions are found or if we are in doubt regarding the diagnosis and management.

摘要

影像檢查在涵蓋基層醫療在內的各個醫療層面日益普及，其原因有多種 ，包括公眾對各種現行技術的認識增加，他們的期望提高，出現更新和更好設備以及更容易得到相關服務。由于不同影像檢查可帶來的假像，家庭醫生有責任指引病人依據個人需要，平衡利弊，並考慮其他因素如成本效益和準確度，而作出適當的選擇。

鑑於上述問題而且本地為基層醫療而設影像檢查指引相對缺乏，我們搜尋和回顧一些海外和本地影像檢查的臨床指引，提出血尿檢查的建議。

採集病史和身體檢查後，進行血液檢測和尿液檢測。假如發現腎小球出血伴有蛋白尿或腎功能不全的情况，病人應轉介予腎臟專科醫生作評估。假如懷疑爲非腎小球出血，泌尿系統電腦素描造影可做爲第一綫檢查。假如無此檢查或考慮價格昂貴，可進行腎臟超聲波或靜脉泌尿系統造影術，但可能需要進一步造影檢查。假如上述造影檢查顯示不出病因，而病人年齡超過40歲、有泌尿系統癌症危險因素或有持續肉眼血尿的，建議作膀胱鏡檢查。假如40歲以下的病人，但尿液細胞學檢查呈陽性反應或令人懷疑，也建議作膀胱鏡檢查。假如發現可疑的病變或對診斷及治療產生疑問，須轉介泌尿科醫生作進一步診治。

---

Introduction

As first-line clinicians, we often manage clinical problems at its early stage.  Nowadays we receive more and more requests concerning imaging tests in our daily practice. According to an Australian study, there is a 16% increase in likelihood that a radiological investigation will be ordered in any primary care consultation compared to 5 years previously.¹

“But the days of general practitioners having one simple decision when investigating a patient’s symptoms ─ to take X-ray or not ─ are long gone.” said Steve Trumble.² Now we have different imaging tests of different modalities for different suspected pathology. Steve Trumble also said a useful investigation is one in which the result, positive or negative, will alter clinical management and/or add confidence to the clinician’s diagnosis.²

“Too many available choices ─ especially if imperfectly understood ─  create confusion and anxiety in the chooser,” said R.S. Heilman.³ One way to remedy the problem is to have guidelines. Guidelines are systematically developed statements to assist practitioner and patient making decisions about appropriate health care interventions for specific clinical circumstances.⁴

Our team searched and reviewed a number of imaging and clinical guidelines which are evidence-based or based on expert opinions and attempted to formulate statements of recommendation on imaging test ordering in the primary care setting. Hopefully this exercise can act as a platform for local imaging guideline development for primary care in the future.

Haematuria

Haematuria, either macroscopic or microscopic, is not an uncommon  encounter in the primary care setting.  The cause can be significant, such as a carcinoma of kidney.  It may be curable if detected in its early stage.

The American Urological Association (AUA) defines microscopic haematuria as:  “The presence of three or more red blood cells (RBC) per high-power field (HPF) on microscopic evaluation of urinary sediment from two of three properly collected urinalysis specimens”.⁵  However, no threshold number of RBC/HPF has been found that separates patients with clinically important disease from those with no detectable urinary tract abnormalities.⁶  Thus a number of authorities advocate that any amount of haematuria should be considered as a sign of urinary tract malignancy until proven otherwise.  Following this trend of thinking, all persons presented with haematuria should be thoroughly investigated.^{7, 8}

As urinary screening is not documented systematically in Hong Kong, the exact prevalence of asymptomatic microscopic haematuria is unknown.  In a conjoint study, 7,828 consecutive pregnant women were screened for microscopic haematuria in the antenatal clinic of a tertiary referral centre. Out of the 7,828 there were 207 women (2.64%) found with microscopic haematuria.⁹ In the four primary care clinics of the Professional Development and Quality Assurance Service, Department of Health, the prevalence of haematuria (ICPC coding U06) within the period 1 October 2005 to 30 September 2006 was 0.51%.

Worldwide, there had been only a few population-based studies addressing the prevalence of microscopic haematuria, and their results varied according to the age and sex distribution of the populations studied.  In six studies, the prevalence ranged from 0.18 to 16.1%.^10,11 Two studies reported a higher prevalence among women than among men.¹²  Some studies suggested that there might be an increased prevalence among older persons,¹² but others showed no difference according to age.¹³

In a report of the imagings ordered by general practitioners (GPs) in Australia, as represented by a random sample of 1,047 GPs participating in the BEACH (Bettering the Evaluation and Care of Health) programme over a one-year period in 1999 to 2000, haematuria was found to be among the top ten problems most likely to generate imaging orders.  An order for imaging was made at 29.8 % of contacts with haematuria.¹⁴

Methodology

A review of literature on haematuria, radiological investigations of haematuria and guidelines for haematuria was undertaken.  Clinical and radiological guidelines for haematuria were searched in the database available in the websites of the following professional bodies: NATIONAL GUIDELINE CLEARINGHOUSE, NICE GUIIDELINE, SIGN GUIDELINE and NEW ZEALAND GUIDELINE GROUP.  The Cochrane Library, Ovid and Medline databases were also searched for the following keywords and MeSH terms: haematuria/hematuria, presence of blood in the urine, microscopic haematuria, prevalence, radiological investigation*, diagnostic imaging, plain radiograph of kidneys, ureters and bladder (KUB), ultrasound of kidney, intravenous urography/urogram, computed tomography, computed tomography urography (CTU), cystoscopy, sensitivity, renal cell carcinoma, urothelial tumour*, transitional cell carcinoma of bladder, renal calculi, bladder calculi, urinary tract calculi, and guideline*.  In addition, the references in the articles obtained were scanned to identify published studies of potential interest, and other relevant and up-to-date articles.

Guidelines for investigations of haematuria

Five clinical/radiological guidelines on haematuria were identified as a result of searching. They are summarized and discussed in the following paragraphs.

1. Radiological guideline of The Royal College of Radiologists (RCR)

In the guideline produced by The Royal College of Radiologists (2003),¹⁵ classification of evidence levels has been translated into grades of recommendation [and the grade of available evidence] on whether or not the investigation is appropriate based on the system developed by the US Department of Health and Human Services, Agency for Health Care Policy and Research.¹⁶  

For the grades of recommendations, please see Appendix A.
For the grades of evidence levels of recommendations, please see Appendix B.

The recommendations and comment in the guideline for the different modalities of investigations for haematuria (macro- or microscopic) are:

1. Intravenous Urogram (IVU)

The grade of recommendation is ‘Indicated’ (evidence level: B).  IVU alone or Ultrasound (US) together with plain radiograph of Kidney, Ureter, Bladder (KUB) is not ideal for detecting upper urinary tract causes of haematuria.  In most patients, both IVU and ultrasound should be ordered, either concomitantly or in sequence.

2. Ultrasound and KUB/CT scan (CT)

The grade of recommendation is ‘Indicated’ (evidence level: B). In young patients with microscopic haematuria only, ultrasound and KUB may be ordered to evaluate the upper urinary tract, but this combination of imaging tests can miss some upper tract pathology, including small calculi.  Ultrasound alone is not as sensitive as either KUB or CT for detecting renal calculi but acceptable to detect urate calculi.

Bladder ultrasound can detect bladder tumours but cystoscopy remains the gold standard in the detection of bladder cancer.

2. Radiological guideline of American College of Radiology (ACR)⁶

An ACR Committee on Appropriateness Criteria and its expert panels (for haematuria: expert panel on urologic imaging) have developed criteria for determining appropriate imaging examinations for diagnosis and treatment of specified medical condition(s) in 2005.  The criteria are intended to guide referring physicians in making decisions regarding radiological imaging and treatment.  The Appropriateness Criteria Scale is ranked from 1 (=least appropriate) to 9 (=most appropriate). 

Some patients whose urinary tracts have no detectable pathology can release small amounts of blood into the urine. This has led some experts to suggest that minimal microscopic haematuria in an asymptomatic young adult requires no further investigation.  However, as mentioned above, there is no threshold number of RBC/ HPF that has been found to separate patients with clinically important disease from those with no detectable urinary tract abnormalities.  Thus, some authorities advocate that any amount of haematuria should be considered as a sign of urinary tract malignancy until proven otherwise, and thus all patients with haematuria need a complete work-up.  The evaluation will usually be accompanied by cystoscopy, since many bleeding urinary tract lesions arise in the lower tract and no imaging procedure is sensitive enough in diagnosing most of them.

Notwithstanding the above, there may be conditions in which complete radiological work-up is not necessary. For example, a young lady with clinically simple cystitis and whose haematuria completely resolved after successful therapy can probably be spared of any imaging work-up. For patients with proven glomerulopathy, extensive work-up to exclude a surgical lesion probably is also not necessary. An appropriate clinical assessment including history, physical examinations, urine analysis and appropriate biochemical tests should also be considered together with the imaging tests.

The following recommendations are for all patients with haematuria except those with generalized renal parenchymal disease or young females with haemorrhagic cystitis.  There was no agreement about the first imaging examination to choose.

1. Intravenous Urography (IVU)

The appropriateness rating is 8. IVU may miss small anterior and posterior renal masses and small bladder lesions.  If IVU is negative, CAT scan (or ultrasound) may be ordered.

2. CT kidney and urography

The appropriateness rating is 8.

3. Ultrasound (US)

The appropriateness rating is 6. Ultrasound is not likely to detect non-obstructing ureteric stones or small urothelial abnormalities. When ultrasound is used as the primary screening modality, adding KUB may increase the yield from imaging.  When ultrasound is negative and the source of haematuria remains obscure, IVU should be added.

4. KUB

The appropriateness rating is 2. KUB was assumed to be part of the indicated IVU.If an IVU is not performed, KUB may be performed together with ultrasound.

3. Clinical Referral Guidelines by Hong Kong College of Radiologists and Hospital Authority (1998) (HKCR&HA)¹⁷

The recommendations were presented in the form of algorithms.  Evidence-based approach was not found. Thus, the guideline could be considered as one based on expert opinion. 

In summary, for patients presenting with haematuria:

1. Appropriate clinical history, physical examination, urinalysis and blood tests should be performed.

2. If significant haematuria (more than 5 red cells per HPF) were found, further imaging would be necessary.

3. If glomerular bleeding and proteinuria are present, renal parenchymal disease is present.  Ultrasound for renal size and morphology is recommended.

4. For non-glomerular bleeding, the following is recommended:

a)  Intravenous urogram (IVU)
This is the imaging modality of choice because it is considered as the best        “catch all” method for the upper urinary tract.  It could be used to assess function, size and outline of kidneys and the collecting system.

b)  Ultrasound (US)
Considered to be more sensitive than IVU in detecting renal masses and bladder tumour.  Less sensitive than IVU in detecting renal or ureteric stone and urothelial tumour of the upper tract.

c)  CAT scan
More sensitive than ultrasound or IVU in detecting small renal mass or calculi. Performed if ultrasound or IVU is non-contributory.

d)  Cystoscopy
The gold standard for diagnosis of bladder pathology, including malignancy and indicated for patients over 40 years of age when other tests are negative.  For patients under 40, it should be performed only when cytology is positive.

5. If no underlying disorder is found, patients should be followed up at six-monthly intervals for up to three years if the haematuria persists.
4. Guideline for ‘Haematuria’ developed by the Finnish Medical Society Duodecim - Professional Association (FMS)¹⁸

The guideline for haematuria was part of the “EBM Guidelines. Evidence-Based Medicine” developed by the Finnish Medical Society Duodecim - Professional Association in 2004.
Since there was no definite correlation between the degree of haematuria and the severity of the underlying disease, the guideline recommended investigating scanty haematuria as thoroughly as more significant haematuria. 
Summary of the recommendations from the guideline:
1. Urinary tract infections and blood contamination (menstruation, sexual trauma, etc.) should first be excluded.
2. Further investigations should be carried out in all patients with confirmed haematuria that cannot be explained by the above causes  The level of evidence supporting the recommendations for investigations is graded as [C] (i.e. limited research-based evidence and at least one adequate scientific study). For full definitions of the levels of evidence, please see Appendix C.
   a)  Thorough medical history and physical examination should be performed.   
   b)  Appropriate blood tests and urinalysis are to be performed.  Blood tests should include renal function test, and prostate specific antigen.  If the red cell morphology in microscopic haematuria is suggestive of glomerular aetiology (dysmorphic erythrocytes or red-cell casts), and the patient has no proteinuria or renal impairment, no further investigations is needed.  However, the patient should be followed-up with occasional checks (six months and annually thereafter) for the possible development of proteinuria or renal impairment.
   c)  For all patients where the glomerular haematuria had not been verified with urinalysis or blood tests, an ultrasound examination of the kidneys and urinary tract should be performed.
   d ) Patients over 40 years of age should have urine cytology testing..
3. Cystoscopy is recommended for the following situations:
   a) patients over 50 years of age.  In younger patients, it should be done only if macroscopic haematuria is present or the patient had risk factors for bladder cancer (smoking, occupational exposure, history of cyclophosphamide treatment).
   b) with suspicious cells in their cytology report.
   c) the ultrasound result is suggestive of a bladder pathology.
4. Computed tomography is the investigation of choice for suspected urinary calculi or tumour of the upper urinary tract.

5. Despite meticulous investigations the cause of some patients’ haematuria could not be found.  In  these  cases  it  may  be  necessary  then  to  follow up  these patients, e.g. once a year, with a check-up of blood pressure and routine blood tests and urinalysis.
5. The American Urological Association (AUA) best practice policy recommendations on evaluation of asymptomatic microscopic haematuria (2001)5
The American Urological Association (AUA) convened a Best Practice Policy Panel to formulate recommendations for the evaluation of adult patients with asymptomatic microhaematuria.  The recommendations were based on extensive review of the literature and the panel members’ expert opinion.  In addition to urologists, the multispecialty panel included a radiologist, a family physician and a nephrologist.  Since no existing data showed the impact of IVU, ultrasound or CT on the management of patients with microscopic haematuria, the panel concluded that evidence-based guidelines could not be formulated.  Thus, the guideline would probably be described as one based on expert opinion.

Summary of recommendations from AUA was as follows:
1. The presence of significant proteinuria, red cell casts or renal insufficiency, or a predominance of dysmorphic red blood cells in the urine should prompt an evaluation for renal parenchymal disease or referral to a nephrologist.

2. If conditions suggestive of primary renal disease are not present, or if any of the following risk factors for significant urologic disease is present, complete urologic evaluation (includes upper tract imaging, voided urine cytology and cystoscopy) should be performed:

Age >40 years
Smoking history
Occupational exposure to chemicals or dyes (benzenes or aromatic amines)
Analgesic abuse
History of pelvic irradiation
History of gross haematuria
History of urologic disorder or disease
History of irritative voiding symptoms
History of recurrent urinary tract infection despite appropriate use of antibiotics

3. For the evaluation of the urinary tract by various imaging modalities, each has its own advantages and disadvantages.

a) Intravenous urography

This is considered by many to be the best initial study for the evaluation of the urinary tract.  However, it has a limited sensitivity in the detection of  small renal masses and could not distinguish solid from cystic masses.  In such situations, further lesion characterization by ultrasonography or computed tomography would be necessary.  It is better than ultrasonography for the detection of transitional cell carcinoma in kidney or ureter

b) Ultrasonography (US)

     Excellent for the detection and characterization of renal cysts but is limited in the detection of small solid lesions (<3 cm).

c) Computed tomography/CT urography

This is the preferred modality for the detection and characterization of solid renal masses. It is also considered to be the best modality for the evaluation of urinary stones, renal and perirenal infections, and associated complications.

4. Cystoscopy as a component of the initial office evaluation of microscopic haematuria is recommended in all adult patients over 40 years of age and in patients less than 40 years of age with risk factors for bladder cancer. This includes those patients in whom the upper tract imaging has shown a potentially benign source for the bleeding.
5. Because some patients with a negative initial evaluation for asymptomatic microhaematuria were found eventually to develop serious urologic disease, some form of follow-up for all patients is indicated. Repetition of urinalysis, voided urine cytology and blood pressure determination at six, 12, 24 and 36 months should be considered in all cases.  Further evaluation for renal parenchymal disease or referral to a nephrologist should be considered if haematuria persists and hypertension, proteinuria or evidence of glomerular bleeding develops.

Discussion

Haematuria is a common complaint presenting to doctors in both the primary and secondary settings.  Gross haematuria can create considerable anxiety and concern for the patient.  On the other hand, microscopic haematuria is often an incidental finding.  The bleeding can arise from any site along the urinary tract.  The cause can be benign or malignant.  

In general, recommendations of the guidelines start with history taking and appropriate physical examination and urinalysis.  Then, relevant investigations are performed according to the findings from the history, examination and urinalysis.  Most of the guidelines adopt the same approach for imaging of macro- and micro- haematuria. The imaging modalities mentioned differ in sensitivity, and all have their own merits and disadvantages.

Ultrasound is non-ionizing and therefore more convenient as no preparation is required. It is the investigation of choice for pregnant women. The disadvantage is that it is operator dependent and is suboptimal in obese patients. It is good at detecting and characterizing renal cystic masses. It is also good at assessing renal morphology, structure and vasculature, and detecting hydronephrosis. It can assess bladder wall morphology, detect large bladder tumours and assess bladder emptying.¹⁹ However, it has a low sensitivity for the detection of renal tumours less than 3 cm, as Grossfeld has commented.5 A prospective blinded study of 201 patients was performed by Warshauer et al in 1988 to determine the relative sensitivities of ultrasound and IVU for the diagnosis of renal parenchymal masses, using CT as a standard.  The sensitivity of ultrasound was 82 % for the detection of renal masses between 2 and 3 cm but only 60% for masses less than 2 cm.²⁰ A prospective study of 1,930 patients over 3.5 years (1994 to 1997) by Khadra et al found that 43% of renal tumours would have been missed if ultrasound had been used alone.  They found that only a combination of ultrasound and IVU detected all upper tract tumours.²¹

As commented by the Hong Kong College of Radiologists, ultrasound is less sensitive in detecting urothelial tumours than IVU.  The main disadvantage is its limited ability to evaluate thoroughly the urothelium for transitional cell carcinoma since the ureters are rarely visualized in its entirety, even when dilated.²²  The use of ultrasound alone is therefore not considered appropriate for the evaluation of microscopic haematuria in high-risk patients.²³

Ultrasound is not sensitive in detecting ureteric and small renal calculi.  Stones in the pelvicalyceal system can only be reliably identified if they are larger than five mm in size.²⁴ Ulusan et al evaluated the sonographic findings of 50 patients with renal stones. Using non-contrast CT as the gold standard, the sensitivity was only between 53 to 77%.²⁵  Fowler et al compared 123 ultrasound and CT examinations retrospectively for the presence of renal calculi. The sensitivity of ultrasound was only 24% with CT as a reference standard. 73% of calculi not visualized at ultrasound were less than 3.0 mm in size.²⁶ Vrtiska et al evaluated prospectively 83 patients with radiologically opaque renal stones with ultrasound and compared it to KUB with tomograms (KUB/T). Ultrasound could detect all of the stones in only 60% of these patients. 30% (80 of 269) of the papillary-calyceal stones seen on KUB/T were missed on ultrasound. 66% of the stones missed measured 2mm or less. They concluded that KUB/T was a more accurate imaging examination for determination of size and number of small stones.²⁷

Compared to ultrasound, intravenous urography is better at detecting transitional cell carcinoma, which usually manifests as a filling defect in the intrarenal collecting system, pelvis and ureter. However, it is poor for detecting small renal masses.  As demonstrated by Warshauer et al, the sensitivity of IVU for detecting renal masses 2-3 cm and <2 cm were 52% and 21% respectively.^{20, 28}  It could not distinguish solid from cystic masses.  Masses detected by IVU inevitably required further evaluation with ultrasound or CT scan for lesion confirmation and characterization.^{19, 23} 

IVU is more sensitive than ultrasound in detecting renal or ureteric calculi and in detecting urothelial lesions, but its role in the investigation of suspected urolithiasis is gradually replaced by CT.^{29, 30}  Worster A et al performed a meta-analysis of the accuracy of CT versus IVU in the diagnosis of suspected acute urolithiasis.  Four studies involving 296 patients were analysed.  The studies consistently demonstrated CT to be superior to IVU in accurately diagnosing acute urolithiasis.³¹

Another area that needs attention is the diagnosis of urothelial tumours.  The study done by Khadra et al showed that 27 % of upper tract tumours would have been missed if IVU had been used alone.²¹ Ultrasound or IVU on its own was likely to miss upper tract urinary transitional cell carcinoma. The comparison of CTU and IVU in the diagnosis of transitional cell carcinoma will be discussed later in this article.
 There is risk of adverse reaction from contrast agents and considerable radiation exposure in IVU.  It is time-consuming (requiring at least 30 minutes to perform), and occasionally requires delayed radiographs taken up to 24 hours after IV contrast administration to define the level of obstruction.²³  It is also potentially nephrotoxic.

CT urography (CTU) has emerged as an alternative to conventional urography by utilizing the advantages of modern multislice CT technique for the visualization of the entire upper renal tract. It may be defined as the examination of the urinary tract by CT in the excretory phase, following intravenous contrast administration. All the guidelines discussed above ranked the place of CT in the investigation for haematuria highly. RCR described it as being more sensitive than ultrasound in the detection of renal calculi; ACR gave it a high appropriateness rating; HK College of Radiologists commented it as being more sensitive than ultrasound and IVU in the detection of renal mass and calculi; FMS suggested it as investigation of choice for suspected urinary calculi or tumour of the upper urinary tract; and AUA described it as the preferred modality for the detection and characterization of solid renal masses and as the best modality for evaluation of urinary stones, renal and perirenal infections.

The advent of CT urography has made evaluation of the entire urinary tract possible during a single breath-hold.³²  Some described it as a comprehensive test, which can be performed as a substitute “one-stop” imaging test for a number of imaging studies, thereby saving time, hospital visits, and cost, and potentially shortening the duration of diagnostic evaluation for urinary tract pathology.^23,33-35 CT is currently the most sensitive and specific test for the detection and characterization of renal masses^20,23 and for the diagnosis of urinary tract calculi.²³ Neither IVU nor ultrasound plus KUB are considered appropriate modalities for the identification of radiolucent stones, which can be visualised by CT. Indirect signs of urinary tract stones such as ureteral dilation, renal and perirenal infections can be visible as well.  The extraurinary causes of microscopic haematuria and pain such as appendicitis and diverticulitis can also be detected.28  CT urography is now widely available in both public and private sectors in Hong Kong.

The major controversy regarding CTU is whether its accuracy is comparable to IVU in evaluating the urothelium in patients presenting with haematuria. Early evidence in small case series suggests that CTU is a promising technique in imaging the urothelium.²³ Caoili et al reviewed 370 CTU examinations performed over an 18-month period and reported 89 % sensitivity (24 out of 27) in detection of transitional cell carcinoma of the renal collecting system and ureter confirmed by endoscopic biopsy or surgery.³⁶ Dillman and colleagues examined 21 CTU examinations that contained one or more pathology-proven upper tract urothelial neoplasms.  The results were correlated with relevant endoscopic or surgical procedure notes as well as pathology results.  They found overall sensitivities of 91-94% by CTU in detection of the upper tract tumours.³⁷  Tsili and colleagues reviewed retrospectively the CT urographies of 75 patients, referred for painless haematuria. CTU detected all the seven upper tract malignancies with diameter equal or smaller than 5mm in diameter.³⁸  Further studies in larger groups of subjects are awaited to answer definitely whether CTU can replace the role of IVU in imaging the urothelium.²³

Cohen et al recommended that for non-glomerular source of bleeding, CTU should be performed first for the detection of occult stone disease or mass lesion (upper tract or bladder), unless it is unavailable or prohibitively expensive. In such case, ultrasound can be performed in place of CT (and is advised for patients with renal failure, pregnancy, or hypersensitivity to contrast medium), with the understanding that further imaging may be necessary.  Although CT is more expensive than IVU or ultrasound, studies involving the latter techniques are often followed by additional imaging and thus cannot be considered cost-saving in general.³⁹  The attractiveness of CTU over IVU is its potential to act as a “one-stop” imaging study assessing renal parenchyma and urothelium, whereas IVU will always need to be supplemented with ultrasound or CT for evaluation of the renal parenchyma.²³

A group of experts, the members of the CTU Working Group, from the European Society of Urogenital Radiology (ESUR) has systematically scrutinized papers published from 1995 to 2007.  They produced and discussed a summary document at 2006 Annual Meeting of the ESUR and the 2007 European Congress of Radiology with an aim to reach consensus.  True evidence-based guidelines could not be formulated because randomized clinical trials comparing different diagnostic modalities prospectively and outcome studies are not yet available.  They found that most of the currently published data falls within evidence categories III to IV.40    (As commented by O’Connor, the lack of studies with higher levels of evidence is a common problem in the radiology literature).²³ Thus, expert guidelines on indications and CTU examination technique were produced, which are based on an extensive literature review and expert opinion rather than evidence-based data.

The major concern, which may limit universal acceptance of CTU, is the radiation dose associated with the procedure, which is up to five times higher than that of IVU.^{40, 41} Pre-test probabilities for cancer are therefore taken into consideration.   CTU can be justified as a first-line test for patients who are at high-risk of transitional cell carcinoma.  Important risk factors include age above 40 years, macroscopic haematuria, smoking, history of genitor-urinary malignancy and occupational exposure.  The risk from the use of radiation is relatively less important in such high-risk groups.  For lower risk groups, CTU can be used as a problem-solving test if traditional work-up remains negative and significant undiagnosed symptoms persist.  In summary, the ESUR recommends the following diagnostic algorithm:

1.   Low risk patients: ultrasound and cystoscopy followed by IVU if both tests are negative and symptoms persist.
2. Medium risk patients: ultrasound and cystoscopy followed by IVU or CTU if both tests are negative and symptoms persist.
3. High risk patients: CTU and cystoscopy.

   The source of bleeding may remain obscure after imaging of the upper urinary tract. Cystoscopy is often needed to evaluate the lower urinary tract, especially for bladder cancer. Finnish Medical Society differed from AUA and Hong Kong College of radiologists in the recommendation on the age for performing cystoscopy.  In view of the possible serious consequence of bladder cancer, 40 years old of age is a safer choice.  Another indication for cystoscopy recommended by Cohen is the development of gross haematuria, because this finding is associated with a higher risk of urologic cancer than is microscopic haematuria.³⁹  

Summary of recommendations

In view of the above guidelines, and taking into consideration the context in Hong Kong, we would like to make a summary of recommendations for the investigations of patients with the presentation of haematuria in primary care. Relevant investigations are performed according to the findings from the history, examination and urinalysis:

1. When haematuria is present (gross or microscopic haematuria), urinary tract infection or blood contamination such as menstruation needs to be ruled out first. 

2. Relevant history should be obtained and pertinent physical examination is performed.  

3. The following laboratory tests are recommended:

a. Serum creatinine is measured for the detection of renal insufficiency. 

b. Urine cytology is performed for patients over 40 years of age.  If the result is positive, atypical or suspicious, cystoscopy should be arranged to rule out carcinoma of bladder. 

c. Microscopic urinalysis is performed for evidence of proteinuria or glomerular source of bleeding (dysmorphic red cells or red cell casts).  If glomerular bleeding with proteinuria or renal insufficiency is present, the patient should be referred to a nephrologist for evaluation. 

4. If a non-glomerular source of bleeding is suspected, the upper urinary tract should be imaged.  CTU may be performed as a first line investigation, especially for patients with risk factors for urothelial cancers.  If it is unavailable or considered too expensive, or for low-risk patients, ultrasound or IVU can be done as an initial investigation, with the understanding that further imaging may be necessary.   Ultrasound is advised as an initial test for patients with renal failure, pregnancy or hypersensitivity to contrast medium.  If a mass suspected to be malignant is identified, the patient should be referred to a urologist.

5. If the above imaging is non-conclusive, referral for cystoscopy is recommended for patients with positive cytology result, over 40 years old, in those with risk factors for urologic cancer or if macroscopic haematuria is present.  For those under 40, this should be performed only if the cytology is positive or suspicious.

6. If all the tests are normal and the haematuria is persistent, the patient should be followed up with regular check-ups of blood pressure, routine blood tests and urinalysis at 6-monthly intervals for up to three years.

Conclusion

There is increasing emphasis in the medical profession on an evidence-based approach to clinical decision-making.  This applies also in primary care service and in clinical problems that may involve radiological investigations.  There is, however, a paucity of evidence-based guidelines which advise on the appropriateness of diagnostic imaging tests.  It is hoped that the discussion and reviewing of the above imaging and clinical guidelines for haematuria would act as a starting point for further development of local imaging guideline for primary care service. 

Acknowledgement

We would like to give our sincere thanks to the clinic staff of Professional Development and Quality Assurance Unit, Department of Health for their support and precious comments in preparing this manuscript.

Key messages

1. Most adults with haematuria require urinary tract imaging.  Some patients can be spared from further work up if the cause is obvious, e.g. in young woman with simple cystitis and whose haematuria resolves completely after antibiotics.
2. IVU is more sensitive than ultrasound in the detection of renal or ureteric calculi and urothelial lesions.
3. Ultrasound is recommended as the initial test for patients with renal failure, pregnancy or hypersensitivity to contrast medium.
4. CT scan is currently the most sensitive and specific test for the detection and characterization of renal masses and for the diagnosis of urinary tract calculi.
5. If all relevant tests are normal and the haematuria is persistent, the patient should be followed up with check-up of blood pressure, routine blood tests and urinalysis regularly, e.g. at 6-monthly intervals for up to three years.

---

---

Reference

---

Appendix A - The grades of recommendations (for RCR) used are:

• Indicated. Investigation most likely contributes to clinical diagnosis and management.
• Specialised investigation. These are frequently complex investigations which will usually only be performed after discussion with the radiologist or in the context of locally agreed protocols.
• Not indicated initially. The clinical problem usually resolves with time.  RCR suggests deferring the study for three to six weeks (timescale may be shorter for children) and only performing it if symptoms continue. Acute back or neck pain are common examples.
• Indicated only in specific circumstances. These are non-routine studies which will only be carried out if there is convincing reason for performing the imaging intervention, e.g. plain radiography in a patient with backache in whom there are clinical indications to suggest something more than degenerative disease (e.g.osteoporotic vertebral fracture).
• Not indicated. The supposed rationale for the investigation is untenable (e.g. skull radiograph for dementia).

---

Appendix B - grades of evidence levels of recommendations (for RCR)

[A]

• High-quality diagnostic studies in which a new test is independently and blindly compared with a reference standard in an appropriate spectrum of patients.
• Systematic review and meta-analyses of such high quality studies
• Diagnostic clinical practice guidelines / clinical decision rules validated in a test set

[B]
Any of the following:

• Studies with a blind and independent comparison of the new test and reference standard in a set of non-consecutive patients or confined to a narrow spectrum of subjects
• Studies in which the reference standard was not performed on all subjects
• Systematic reviews of such studies
• Diagnostic clinical practice guidelines / clinical decision rules not validated in a test set

[C]
Any of the following:

• Studies in which the reference standard was not objective
• Studies in which the comparison between the new test and the reference standard was not blind or independent
• Studies in which positive and negative test results were verified using different reference standards
• Studies performed in an inappropriate set of patients
• Expert opinion

---

Appendix C - Definitions of Levels of Evidence(for FMS)

1. Strong research-based evidence. Multiple relevant, high-quality scientific studies with homogenic results.
2. Moderate research-based evidence. At least one relevant, high-quality study, or multiple adequate studies.
3. Limited research-based evidence. At least one adequate scientific study.
4. No research-based evidence. Expert panel evaluation of other information.