Applied Radiology

Dr. Fine is Professor of Clinical Nuclear Medicine at Albert Einstein College of Medicine and Chairman of the Department of Nuclear Medicine at Jacobi Medical Center. Dr. Blaufox is Professor of Nuclear Medicine and Chairman, Department of Nuclear Medicine, Albert Einstein College of Medicine, Bronx, NY.

In the past several years there have been a number of important advances in renal nuclear medicine. In general, while the improvements have been evolutionary rather than revolutionary, the cumulative effect over 10 years has been substantial. Nuclear medicine techniques have become procedures of first choice in the diagnosis of renovascular hypertension and in the differential diagnosis of hydronephrosis. Substantial consensus now exists for many indications, protocols, and general methodology.

Quantitation of renal function has become easier, more practical, and more accurate than ever using nuclear techniques. The voiding cystourethrography (VCU) has become widespread in the evaluation of reflux, and renal morphology studies have become invaluable in children with pyelonephritis. Full descriptions of all these techniques are beyond the scope of this article, however. Many excellent reviews are cited here, including consensus reports and practice guidelines and these should be consulted for greater detail. Our objective is to provide selected comments regarding critical features of the most important and useful renal radionuclide procedures currently in use.

Tracers

The most significant advance in radiotracer development in the past 10 years has been the clinical application of Tc-99m mercaptoacetyltriglycine (MAG3) as an all-purpose renal scinti-imaging, function, and excretion agent. ¹ Older function and excretion tracers continue to retain a role, although to a lesser extent than previously. MAG3 effectively eliminated the use of I-123 orthoiodohippurate due to the expense of cyclotron production of the latter isotope. I-131 orthoiodohippurate retains a role in emergency studies due to its long shelf life, although it is of value only in centers using older gamma cameras still capable of imaging the 364 keV photopeak. Tc-99m diethy-lenetriaminepentaacetic acid (DTPA) continues to be used in many renal function/excretion studies, although it is considered a second-line tracer in the presence of renal dysfunction, and in the evaluation of obstruction. Tc-99m glucoheptonate (GHA) and Tc-99m dimercaptosuccinate (DMSA) remain useful in the evaluation of acute and chronic pyelonephritis.

Principle procedures and indications

Several renal radionuclide exams have become the first-line procedures in the evaluation of specific clinical conditions. These are indicated in Table 1. Diuretic renography ^2-4 and captopril renography ^5-7 are the subjects of several recent consensus reports, and practice guidelines, whose major features are discussed below. Table 1 also includes renal scintigraphy and cystourethrography in view of our strong perception of substantially increased usage in recent years.

Other renal radionuclide procedures of significance are included in Table 2. These include procedures that have achieved substantial utilization in many institutions, but whose indications have not yet achieved the level of widespread consensus.

Specific procedures addressed by major consensus studies

Captopril renography--Captopril renography has emerged as the procedure of choice to identify renovascular hypertension in patients with high blood pressure. Data have accumulated confirming the specificity of the procedure in a variety of clinical circumstances. Specificity is maintained using strict interpretative criteria, even in renal insufficiency, ⁸ although sensitivity is reduced. Of great importance is the procedure's value in predicting blood pressure improvement after renovascular surgery (i.e., the hallmark of renovascular hypertension [RVH] and its distinction from renal artery stenosis [RAS]). ^9-12 An important consensus report ⁵ was followed in print by Procedure Guidelines ⁷ by many of the same authors and investigators. Recommendations from these can be taken together as the definitive current state of the art. They include:

Preparation:

a) Adequate hydration (250 to 500 mL fluid by mouth, at minimum, 30 minutes before tracer administration).

b) Discontinuation of diuretics for 1 week and preferably (although not absolutely) ACE inhibitors (captopril 1 to 2 days; enalapril, lisinopril and other longer-acting inhibitors 4 to 5 days). Other medications may be maintained without jeopardizing interpretability of the exam.

Study methodology: a) 1- or 2-study using Tc-99m MAG3, or DTPA. I-123 OIH and I-131 OIH are not as widely available, but are acceptable tracers as well.

b) Baseline exam with 2 mCi (75 MBq) MAG3 or 3 to 5 mCi DTPA.

c) 25-50 mg oral crushed captopril. (IV furosemide optional; IV enalaprilat instead of oral captopril also optional); monitor blood pressure for 1 hour.

d) Post-captopril study with same tracer as baseline, using at least double dose.

Intepretation:

a) A post-captopril unilateral reduction in renal uptake (seen mostly with DTPA) or prolongation in parenchymal retention (any tracer) are the most reliable criteria for identification of renal artery stenosis causing renovascular hypertension. Qualitative interpretation may be refined into probabilities of disease using the Oei/Nally criteria (figures 1 and 2).

b) The only quantitative criteria that perform respectably are:

1) a unilateral increase by at least 0.15 in the 20/Pk value post-captopril for MAG3 or or OIH (and probably for DTPA as well);

2) Tpk increase by >=2 minutes

3) a unilateral reduction in relative renal function by >10%.

No other quantitative criteria are definitely helpful.

Renal dysfunction (glomerular filtration rate [GFR] <50 mL/min), a small kidney, and bilateral disease all reduce the sensitivity of the test, although specificity appears preserved if a unilateral change in the renogram after captopril is strictly adhered to as a diagnostic requirement. ⁸ An important pitfall to note is a bilateral and symmetric change in the renogram after captopril. This finding is sometimes contributed to by patient dehydration, although the cause is often not determined. Regardless of circumstances, this must be considered a non-diagnostic result, as most exams with this result do not turn out to have bilateral renal artery disease.

Variations: The most common variation is to skip the baseline exam under the premise that a normal post-captopril exam ends the search for RVH. While this is true, an abnormal exam usually mandates recalling the patient for a "baseline" study on another day. Therefore, any pre-test circumstance that increases the likelihood for an abnormal renal exam pushes examiners toward performing baseline and post-captopril on the same day (Table 3). The baseline exam should definitely be performed if the pre-test risk for RVH reaches 30% or more. ¹² In patients for whom there are very few risk factors for RVH, the convenience of omitting the baseline exam may be practical.

Diuresis renography--Twenty years after its introduction, diuresis renography continues to be a highly valuable clinical standard for assessment of renal outlet obstruction among patients with incidentally discovered hydroureteronephrosis. It is important to distinguish true outlet obstruction from among the non-obstructive conditions that may be associated with hydronephrosis (Table 4). Limitations on the procedure exist in patients with renal dysfunction and severe hydronephrosis in whom an insufficient diuretic response may be misleading or non-diagnostic. In addition, an optimal technique has evolved over the years owing to attention to many well documented technical details. ²

Preparation: Adequate hydration is very important. In adults, the dehydration that commonly accompanies an early morning study (particularly in patients who mistakenly think that all medical tests should be performed after an overnight fast) generally should be addressed by administration of 250 to 500 mL of fluids 30 minutes before the study. In small children, additional intravenous hydration is recommended. ³ Bladder catheterization also is needed in small children (i.e., those who cannot be relied on to void on command) to permit adequate urine flow.

Tracer: The tracer of choice has become Tc-99m MAG3 due to rapid clearance from the blood and consequent prompt filling of the renal pelvis. DTPA, in many patients, is cleared from the plasma too slowly and, therefore, fills the pelvis too gradually to permit adequate assessment of pelvic drainage.

Diuretic: The consensus recommendation calls for intravenous administration of 0.5 mg/kg furo-semide (1 mg/kg up to 2 years old) at 20 or 30 minutes after administration of the radiotracer (often identified as F+20 or F+30, respectively).

Interpretation: A brisk response to furosemide usually can be gauged scintigraphically (figure 3). In addition, the response should be assessed by the renographically determined renal pelvic time to half-emptying, using a renal pelvic region of interest. A value of 12 to 15 minutes or less may be interpreted as a normal value. One should bear in mind, however, that slavish reliance on this number may be misleading. For example, renal dysfunction, dehydration, and severe pelvic dilatation all may contribute to a prolongation in the half-emptying time to furosemide without the presence of obstruction. It's best to recognize that <10 minutes is clearly normal, 10 to 15 minutes is probably normal, and >20 minutes is abnormal. Values between 15 to 20 minutes should be interpreted with caution, and may require sequential studies to diagnose obstruction by the deterioration in renal function. Other pitfalls in technique that can also misleadingly suggest obstruction include infiltration of either the tracer or the diuretic. The former can be detected by an image of the injection site, or a rising background curve.

Variations: The peak action of intravenous furosemide is usually about 20 minutes after administration. Therefore, the absence of a brisk diuretic response following furosemide should prompt a repeat study with furosemide administered 15 minutes prior to the radiotracer (so-called F-15 study; figure 4). Unobstructed kidneys will have normal renograms using this approach. (A minority recommendation in a consensus paper was to dispense with the F+20 study altogether in favor of the F-15 study for all patients. ² This approach has merit, although it should probably be reserved for experienced practitioners). Recently there has been support for a study giving furosemide simultaneous with the radiotracer. ¹⁴

Detection of renal infection and scar--Both GHA and DMSA have been shown to be of value in demonstrating regions of acute and chronic renal infection in children and adults. ¹⁵ Evidence has accumulated showing the progression of repeated bouts of acute pyelonephritis with the development of scars. ^16,17 The value of scinti-imaging is in early detection of abnormalities, especially since data are available demonstrating healing of scintigraphic defects associated with successful early antibiotic treatment. ^17-23 Acute pyelonephritis is associated with regions of hypofunction manifested as intraparenchymal defects. Renal contour abnormalities or other morphologic distortions due to scar most commonly reflect chronic pyelonephritis. In acute or chronic presentations, scintigraphy has very high sensitivity (>90%) for detection of infection or scar. The relative value of planar vs SPECT imaging is somewhat controversial, as some investigators have found increased sensitivity of SPECT to introduce artifactual defects. ²⁴ Nonetheless, even with planar imaging, the studies show remarkable interobserver reproducibility ²⁵ and sensitivity easily exceeds IVU, sonography, and CT. Spiral CT may be competitive, but data are still lacking.

Voiding cystourethrography--The voiding cystourethrogram remains valuable to document and follow ongoing reflux, a known contributor to renal infection and scarring. ^26,27 The radiologic and radiotracer versions of this procedure are quite similar, with a slight edge in sensitivity going to contrast VCU for grade I and II reflux and to the radionuclide study for grades III to V. Institutional practices usually determine whether a child's first diagnostic VCU is performed with contrast or tracer. Regardless, follow-up usually is performed with the radionuclide procedure due to much lower radiation absorbed dose. Identification of reflux in a child with scarring signals the need for corrective surgery to stop the reflux of urine.

Other procedures

Measurement of renal function­­ Nephrologists have become increasingly interested in accurate evaluation of GFR in a variety of clinical situations. A common example is in the sequential evaluation of renal transplant patients on renotoxic drugs. Creatinine clearance determinations are generally compromised by urinary collection difficulties. Clinically useful formulas to determine GFR have been developed based only on serum creatinine concentrations together with the patient's age and weight, but more accurate still have been single sample DTPA clearance determinations.

Measurement of renal function is more difficult to standardize in children. This difficulty is reflected in the lack of consensus that still exists regarding measurement methods.

Total renal function can be quite easily measured using single blood sample methods. These techniques are approaching consensus for measurement of GFR using Tc-99m DTPA. In adults, formulas have been developed that accurately determine GFR from a single blood sample drawn at about 3 hours after injection. ^28,29 In children, consensus has been harder to reach, principally because it is more difficult to establish norms in children.

Individual renal function remains the unique province of nuclear medicine among non-invasive techniques. ^30-32 The simplest procedure that has gained consensus, using MAG3 or OIH is

R (%) = R/(R + L)

Where R is the summed counts within the right kidney region of interest from 60 to 120 seconds and L is the same for the left kidney. Using DTPA, the same formula applies, but the counts are best obtained from 90 to 150 seconds.

Critical to correct performance of this simple equation is that the correct time of integration is: a) performed before 150 seconds to minimize contribution of pelvic excretion; b) begun at no sooner than 60 seconds to minimize the contribution of blood background; and c) The use of background subtraction is generally done with lateral semi-lunar crescent regions of interest normalized to the area of the respective renal region of interest.

Exercise renography: The association of abnormal renal function during upright bicycle exercise has been found in up to 50% of individuals with early essential hypertension with no other evidence of renal disease. This suggests a possible causative role of the kidney in some individuals with essential hypertension. This line of evidence, while provocative, has yet to be fully evaluated. ^33-35

Aspirin renography: Aspirin, a prostaglandin E2 inhibitor, has an enormous range of uses; one of which is to cause reduced urine production in a kidney affected by RVH. The effect on the renogram is similar to that of captopril, and the sensitivity and specificity may be comparable, although these data are available only in small series. ³⁶

Evaluation of impotence: Techniques have been developed using gamma camera scinti-images with computer quantitation of blood washout as well as with lightweight highly sensitive bedside probes. ³⁷ These have not yet become routine, although a great deal has been learned about the physiology of erectile incompetence.

Conclusion

It is clear that renal radionuclide procedures have assumed major roles in the evaluation of several important clinical conditions affecting children and adults. In most instances, the aspect of the radionuclide test that sets its value apart from a competitive imaging exam is the functional information revealed. This can be seen explicitly in captopril renography, where the physiologic significance of a given anatomic stenosis is assessed by its effect on the renin angiotensin system. Similarly, diuretic renography tests the physiologic significance of a given anatomic dilatation. Somewhat less obvious is the remarkable superior sensitivity of GHA and DMSA renal morphology studies in pye-lonephritis compared with most anatomic imaging; although here spiral CT may end up playing a larger role in the future. Additional functional tests on the horizon maintain attention on the value of renal radionuclide procedures well into the foreseeable future. AR