The Radiological Investigation of Haematuria
Dr Phil Haslam
Consultant Interventional and Uroradiologist
The Freeman Hospital
Newcastle upon Tyne
Haematuria is an important finding as it may indicate the presence of serious pathology within the urinary tract, such as:
Renal cell carcinoma
Transitional cell carcinoma
Renal parenchymal disease
Urinary tract infection
Benign prostatic hypertrophy
Haematuria can be a symptom (visible or visible) or sign (non-visible or non-visible). Patients with visible haematuria tend to present to their doctor with this complaint, whereas non-visible haematuria patients are detected on urinanalysis. Visible haematuria is four times more likely to be due malignancy than non-visible haematuria. Non-visible haematuria has a very poor specificity and urinary tract infection should certainly be excluded prior to referring patients for imaging investigations.
The imaging strategy we use in the investigation of these patients therefore needs to detect calculi, renal parenchymal masses, upper tract urothelial tumours and of course bladder tumours. The single most effective modality for this is multiphase CT with the exception of bladder tumours, which are best seen at cystoscopy. Virtual CT cystoscopy is no substitute, unless cystoscopic access is impossible. Unfortunately haematuria is very common, not always significant and the demands on the service and radiation burden would be very high if all patients with haematuria were investigated in this way. A sensible strategy for investigation is needed.
In this article I intend to look at some of the different imaging modalities available to us, discuss their strengths and weaknesses and suggest how we should be investigating these patients.
The plain radiograph (xray KUB)
This still remains a useful examination for the assessment of the urinary tract. It has an excellent spatial resolution (0.03mm or 15 line pairs per mm). It will demonstrate many radio opaque calculi (approximately 90% are radio opaque) and has a relatively low radiation dose. It is an essential investigation in patients with calculus disease as it is important to know if the calculi can be seen on a plain film to enable follow up.
Plain abdominal film showing a single left ureteric calulus
Plain abdominal film with classical appearances of medullary sponge kidney
The Intravenous Urogram (IVU)
The intravenous urogram is a readily available, cheap and efficient means of assessing the upper urinary tract, especially the collecting system, but has now been largely superseded by the CTIVU. I read an excellent article in Radiology just after I was appointed as a consultant in 1999 ‘Epitaph for the urogram’ (1) in which Dr Amis declares ‘The excretory urogram is terminally ill. It survives only because of clinicians and radiologists who do not fully appreciate that more accurate delineation of the urinary tract can be gained with other techniques.’ A response in Radiology 2001 quoted Mark Twain ‘Reports of my death are greatly exaggerated.’(2) It has taken nearly 10 years or more for most modern hospitals to replace the IVU with CT. One of the main criticisms of the IVU is its poor sensitivity for small renal masses (<3cm). Even with the addition of nephrotomography the IVU will only identify approximately 85% of masses over 3cm in diameter. The presence of overlying bowel gas and poor bowel preparation all reduce its sensitivity further. Clearly the IVU in isolation is not sufficient for the investigation of haematuria.
Ultrasound remains an essential tool in the investigation of haematuria. It is of course radiation free and contrast free. It is sensitive for calculi and upper tract tumours, but small renal cell cancers can be isoechoic and be missed. Small transitional cell tumours can also be missed, unless the system is dilated. Small calculi are difficult to demonstrate when adjacent to sinus fat. It is however extremely good at differentiating solid from cystic masses. The presence of Doppler flow within a mass can differentiate it from a haemorrhagic cyst and the use of ultrasound contrast agents can also help in this respect.
A series of 3 ultrasound images with time post injection of ultrasound contrast agent. This demonstrates enhancement of the renal mass and ‘washout’ of contrast at 1:44 minutes.
Another difficulty is the small hyperechoic renal mass. These are not all angiomyolipomas, as 32% of renal cell cancers <3cm can be hyperechoic. These masses should be characterised with CT.
A small hyperechoic renal cell carcinoma proven at biopsy.
CT has rapidly become the investigation of choice for many urinary tract conditions. Its spatial resolution is less than the IVU at 0.25mm compared with 0.03mm but with the latest 64 and 128 slice scanners this is in all three plains giving true isotropic voxels. This coupled with its excellent contrast resolution (at least 10 times that of the best plain film) makes it the best test for renal calculae.
Multiphase CT is extremely good at evaluating renal masses. Any mass with non enhanced attenuation greater than 20 Hounsfield units (HU) may be a hyperdense cyst (due to haemorrhage or proteinacious fluid) or renal neoplasm. The enhancement characteristics of such a mass can distinguish the two. Its essential to perform both pre contrast and post contrast scans using the same scanning characteristics, to enable accurate measurement of CT attenuation. The best phase for detection of renal masses is the nephrographic phase that occurs at about 90-120 seconds post contrast injection.(4) At this point there is uniform enhancement of the cortex and medulla and maximal differentiation of relatively poorly enhancing renal masses.
A small low attenuation left renal mass proven at biopsy to be a small papillary RCC
If scanning is performed too early when the medulla has not enhanced sufficiently then a deeper mass may be missed. A mass is said to have enhanced when its mean attenuation increases by at least 20HU from baseline level. These must therefore be vascular and considered malignant until proven otherwise. Enhancement less than 10HU ‘pseudoenhancement’ can occur especially in smaller masses due to beam hardening and compensatory reconstruction algorithms.
The difficulty occurs with enhancement of 10-20 HU which is deemed indeterminate or when only a post contrast scan has been performed. In the former situation I often find that MRI helps. I usually perform a gradient echo sequence pre and post Gadolinium and find that most of these lesions do show definite enhancement. The alternative is ultrasound with contrast which can show wash in and washout of contrast confirming its vascularity. Sometimes only a post contrast scan has been performed. Providing the mass has been seen then a delayed scan at 10 minutes or more will show de-enhancement (washout) of at least 10 HU in vascular masses. (5)
CT is also very useful for detecting urothelial lesions such as TCC. It is most sensitive when a delayed scan is performed in the excretory phase (>180 seconds). Urothelial lesions can then be seen as filling defects in the contrast filled system.
This image shows a right lower pole TCC.
In practice we perform this scan at about 7 minutes post contrast after injection of 20mg of intravenous frusemide. This dilutes the contrast and provides a degree of urinary distension. The same effect can be achieved with good prehydration. It is important to roll the patient prior to the delayed scan to avoid layering of contrast posteriorly. Quite frequently upper tract TCC can be seen as an enhancing mass against the background of non opacified urine in the nephrographic phase. The delayed phase may then not be required as can be seen in this scan showing multifocal TCC of the ureter.
A full urinary CT examination should therefore require 3 phases. Pre contrast for calculi and to provide a baseline attenuation for masses. Nephrographic phase for detection of renal masses and a delayed phase for urothelial lesions. In addition a corticomedullary phase/vascular phase may be added to delineate vessels for surgical planning ! This is a significant amount of radiation.
There has been considerable recent interest in reducing the phases of the CTIVU and some perform a split bolus technique. An initial unenhanced scan is performed. Following this, contrast is injected followed by a second bolus of contrast two minutes later. The patient is scanned 90 seconds after this second bolus by which time the first bolus of contrast is being excreted. Thus the nephrographic and excretory phase images are obtained in one scan. (6) This has shown promising results with a high sensitivity and specificity in the detection of tumours.
MRI has the advantage of no radiation and no iodinated contrast. It is very sensitive when looking for enhancement in small renal masses or cyst walls. It is sensitive for urinary calculi but not as good as CT for the tiniest of stones. MR urography is an excellent technique to assess the upper tracts. This can be performed using a HASTE sequence in well-hydrated patients or post intravenous frusemide to distend the system.
A small right lower pole renal calculus on MRU.
Obstructed systems are well seen using this type of sequence with high resolution T2 images obtained through any area of abnormality. The degree of enhancement of any mass can be assessed post gadolinium. Excretory MR urography can then be perfomed using a T1 weighted gradient echo sequence in the excretory phase. This is complementary to the T2 weighted images and can produce IVU type images. Urothelial tumours and calculae are seen as filling defects.
The disadvantages of MRI are the time taken to perform a full evaluation of the urinary tract, its susceptibility to motion artefact and the lack of access for this large group of patients in most centres.
Clearly we cannot justify CT for every patient with haematuria. Non-visible haematuria has a poor specificity in screening for occult malignancy, but it is still reasonable to investigate these patients. The question is, how far should this be taken?
The risk of malignancy increases with age, male sex, smoking and occupational exposure. Khadra et al (8) looked prospectively at 1930 patients evaluated in a haematuria clinic and found no upper tract tumours in anyone less than 40 years of age.
Young patients with non-visible haematuria could therefore be imaged purely with ultrasound and a plain abdominal xray to look for calculus disease. They did however find 7 bladder carcinomas in patients under 40 years old, one of whom only had non-visible haematuria. They therefore advocate cystoscopy as part of the investigative algorithm for all types of haematuria. Non-visible haematuria can of course be a sign of renal parenchymal disease and some would argue that younger patients should be investigated by a nephrologist in the first instance and only have imaging if it is persistent and no nephrological cause has been found. (9)
Older patients have a higher incidence of significant pathology and a more thorough investigation strategy should be used. All should have cystoscopy as this is the most sensitive method to examine the bladder. The upper tracts should be imaged to look for renal masses and transitional cell tumours. In many centres this is done with the use of ultrasound and IVU.
Currently it is entirely possible to replace the IVU and ultrasound with a dedicated urinary tract CT (the modern IVU). This is at the expense of a higher radiation dose and we should therefore only do this in patients who are at higher risk of malignancy, the over 50s with negative ultrasound, abdominal xray and cystoscopy. MBUR 6 (10) produced by the Royal College of Radiologists in 2007 suggests that CTU be limited to patients over 45 years old. Edwards et al (9) take this even further in their discussion of IVU and suggest it is limited to men aged over 50 with visible haematuria and positive repeat dipstick urine anlaysis where first-line tests are negative.
In summary I would suggest CT and cystoscopy in the over 50s with visible haematuria. Younger patients under 50 with any type of haematuria should have ultrasound and a plain abdominal film in the first instance. If they have visible haematuria and other investigations are negative then they should proceed to CT. The IVU is not entirely dead. It has effectively transformed into the CTIVU.
1. Epitaph for the urogram. Amis ES Jr. Radiology 1999; 213:639–640.
2. Urography Survives Joshua A. Becker, MD, Howard M. Pollack, MD and Bruce L. McClennan, MD Radiology. 2001;218:299-300
3. Forman HP,et al.Hyperechoic renal cell carcinomas. Increase in detection at US. Radiology 1993;188:431-434.
4. Cohan RH, Sherman LS, Korobkin M, et al. Renal masses: assessment of cortico-medullary-phase and nephrographic-phase CT scans. Radiology 1995; 196:445–451.
5. Macari M, Bosniak MA. Delayed CT to evaluate renal masses incidentally discovered at contrast-enhanced CT: demonstration of vascularity with deenhancement. Radiology 1999; 213:674–680.
6. Lawrence C. Chow, Sharon W. Kwan, Eric W. Olcott, and Graham Sommer
Split-Bolus MDCT Urography with Synchronous Nephrographic and Excretory Phase Enhancement. Am. J. Roentgenol., Aug 2007; 189: 314 – 322.
7. Naoki Takahashi, Akira Kawashima, James F. Glockner, Robert P. Hartman, Bradley C. Leibovich, Anja C. S. Brau, Philip J. Beatty, and Bernard F. King. Small (<2-cm) Upper-Tract Urothelial Carcinoma: Evaluation with Gadolinium-enhanced Three-dimensional Spoiled Gradient-Recalled Echo MR Urography. Radiology 2008 247: 451-457
8. Khadra MH, Pickard RS, Charlton M, Powell PH, Neal DE. A prospective analysis of 1,930 patients with hematuria to evaluate current diagnostic practice. J Urol 2000;163: 524–7
9. Edwards, Tom J.; Dickinson, Andrew J.; Natale, Salvatore; Gosling, Jane; McGrath, John S. A prospective analysis of the diagnostic yield resulting from the attendance of 4020 patients at a protocol-driven haematuria clinic. BJU International. 97(2):301-305, February 2006.
10. Making the best use of clinical radiology services, Referral guidelines. The Royal College of Radiologists, September 2007