| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, CHUM–Notre-Dame Hospital, 1560 Sherbrooke East,
Montreal, Quebec, Canada H2L 4M1.
2 Department of Radiology, CHUM–Hotel-Dieu Hospital, Montrel, Quebec,
Canada H2W 1T8.
3 Department of Radiology, CHUM–St Luc Hospital, Montreal, Quebec, Canada
H2X 3J4.
Received February 17, 2004;
accepted after revision July 19, 2004.
This study was supported by an operating grant of the Canadian Health
Institute of Research, Grant Number MCT-52685. Gilles Soulez was supported by
a research scholarship of the Fonds de la Recherche en Santé du
Québec.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Estimations of positive and negative predictive values of baseline and captopril-enhanced renal scintigraphy and Doppler sonography examinations for predicting a favorable outcome after renal angioplasty were based on a previously published prospective study involving 74 patients who underwent this treatment. For gadolinium-enhanced MR angiography, predictive values were calculated from a subpopulation of 57 of these 74 subjects. The value of different combined strategies with these techniques for predicting clinical success after angioplasty was evaluated in this population. The costs of investigation and treatment per improved patient were calculated for each imaging technique and for combined strategies in a hypothetic 1,000-patient population with a 30% prevalence of renal artery stenosis, relying on the diagnostic performance reported in the literature for each technique in detecting renal artery stenosis.
RESULTS. The costs for each improved patient were $12,579 for patients selected on the basis of a positive finding on Doppler sonography (false-negative results = 12/1,000) and $10,149 for patients selected with criteria combining a positive finding on Doppler sonography with a bilateral resistive index of less than 0.75 (false-negative results = 32/1,000). Patient selection based on a positive finding on MR angiography cost $18,119 (false-negative results = 0), whereas the cost of patient selection based on a positive finding on renal scintigraphy was $12,939 (false-negative results = 29/1,000).
CONCLUSION. Doppler sonography is more cost-efficient but less sensitive than MR angiography for identifying patients with renovascular hypertension. MR angiography should be favored in hypertensive patients who are resistant to medical therapy to avoid false-negative examinations.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Hypertension is most often idiopathic, but 1–5% of patients with hypertension have renovascular disease [5]. Percutaneous transluminal renal angioplasty, according to most studies, cures hypertension in about 5–10% of patients or improves blood pressure values in approximately 40–60% [6–10]. Hence, there is an interest in screening and treating renovascular disease.
At present, revascularization is recommended for patients who have hypertension that cannot be adequately controlled with medications; patients with severe bilateral renal artery stenosis or stenosis in a solitary functioning kidney; or patients with congestive heart failure if no other cause can be found [11]. Recommendations for patients whose hypertension can be adequately controlled with two or three drugs are unclear.
Different diagnostic techniques are used to detect renal artery stenosis. Angiography, although acknowledged as the gold standard, is an invasive and expensive test. It provides solely anatomic data and does not provide any information on outcome after angioplasty. Several randomized studies recruiting hypertensive patients on the basis of angiographic stenosis without physiologic evaluation of the kidneys failed to show either a significant survival or clinical benefit of renal revascularization over medical treatment [12–14]. Therefore, it is important to find a screening test that is noninvasive and reflective of the therapeutic outcome after angioplasty at the lowest possible cost.
Positive results on renal scintigraphy have been linked to better clinical outcomes after angioplasty [15]. Recently, Radermacher et al. [16] reported good clinical outcomes after renal angioplasty in patients with a bilateral intrarenal resistive index of less than 0.80 as measured with Doppler sonography. In a recent study, we established that findings on Doppler sonography were a better predictor of therapeutic outcomes than those on renal scintigraphy [17]. In this cohort, most of the patients also were evaluated with MR angiography.
Using the same patient population as in a prior study [17], we sought to compare the cost–benefits of screening patients before renal angioplasty with Doppler sonography, captopril-enhanced renal scintigraphy, MR angiography, and catheter angiography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These 74 subjects were recruited from a population of 139 patients referred for suspected renal artery stenosis between January 1998 and January 2002. The reference pattern of our population was a clinical suspicion of renal artery stenosis in patients who either had no previous imaging study (35/139) or had a positive finding on proximal or intrarenal Doppler sonography (53/139), on captopril-enhanced renal scintigraphy (23/139), on MR angiography (11/139), or on catheter angiography (17/139). Of these 139 patients, 65 were excluded for the following reasons: the investigation was not completed before angioplasty (n = 5); the patient refused to undergo angiography (n = 4); no stenoses were found on noninvasive imaging (n = 13); the patient presented with clinical deterioration that contraindicated angiography (n = 8); renal artery stenosis was not significant on angiography (n = 28); the patient died (n = 3); or the patient was lost before clinical follow-up (n = 4).
The 74 patients (39 women and 35 men; mean age ± SD, 65 ± 9 years) in our study population had a mean duration of hypertension of 10 ± 11 years (range, 4 weeks–45 years). The mean number of regimen antihypertensive drugs taken was three (SD, ± one medication). All patients eventually underwent angiography, which disclosed a mean degree of stenosis of 72% ± 11%. All patients first had renal balloon angioplasty. In 52 patients (70%), angioplasty was followed by stent insertion. Technical success was obtained in all patients. There were no major complications after the procedure.
All patients had systolic hypertension above 160 mm Hg or diastolic hypertension above 90 mm Hg with one of the following clinical criteria of renovascular hypertension [18, 19]: hypertension despite treatment with three or more antihypertensive drugs; systolic or diastolic abdominal murmur; malignant hypertension (retinal hemorrhage, exudates, or papilledema); small unilateral kidney seen on previous imaging studies; recent onset of hypertension in a patient with known coronary, peripheral vascular, or cerebrovascular disease; onset of hypertension in a patient older than 50 years; unexplained azotemia or renal failure triggered by administration of angiotensin-converting-enzyme-inhibitor medication; or hypokalemia.
Because we reported the value of Doppler sonography, renal scintigraphy, catheter angiography, and renal angiography for predicting the therapeutic outcome in detail in a prior report [17], we have merely summarized them below.
Doppler examinations were performed in all but one patient 2–5 days after discontinuation of angiotensin-converting-enzyme-inhibitor medication. Doppler sonography was repeated 1 hr after oral administration of 25 mg of captopril. After measuring the length of both kidneys, we studied the proximal renal arteries and then conducted an intrarenal examination with measurement of the resistive index and early systolic acceleration obtained from the superior, middle, and inferior portions of the kidneys. The result of proximal Doppler analysis was considered positive if the maximum velocity in the proximal renal artery exceeded 180 cm/sec and if the ratio between the maximum velocity in the proximal renal artery and the maximum velocity in the aorta was greater than 3. Because the proximal arteries were examined both before and after captopril administration, the result of proximal renal assessment was considered positive if these criteria were met either at baseline or after the captopril-enhanced examination. The result of intrarenal Doppler sonography was deemed positive if either the quantitative (acceleration and time of acceleration) results or findings of morphologic pattern recognition were abnormal during the baseline or captopril-enhanced examination as defined previously [17]. Bilateral resistive index (mean of measurements taken in the upper, middle, and lower portions of both kidneys) and unilateral resistive index (mean of measurements taken from the stenosed kidney) were recorded systematically before and after captopril administration. In cases of bilateral stenosis, the unilateral resistive index was calculated from the kidney with the more severe stenosis on angiography, provided that the kidney length was greater than 8 cm.
Baseline and captopril-enhanced technetium-99m mercaptoacetyltriglycine (MAG-3) renal scintigraphy was performed in all patients following a 1-day 25-mg-captopril protocol, as recommended by the Working Party Group on Determining the Radionuclide of Choice [20], after cessation of angiotensin-converting-enzyme inhibitors in all but two patients. Results were interpreted according to the guidelines of the Society of Nuclear Medicine [21].
MR angiography was performed on 1.5-T scanner (Magnetom Vision, Siemens) using a phased-array body coil. We used a coronal 3D gradient-echo sequence centered on the aorta and the kidneys before and during dynamic IV administration of 0.2 mmol/kg of gadopentetate dimeglumine (Magnevist, Berlex Canada) at a rate of 1.5–2.0 mL/sec (FLASH 3D with interpolation; bandwidth, 390 mm; matrix, 512 x 180; TR/TE, 4/1; flip angle, 30°; matrix, 512 x 512). Before administration of the contrast material, we determined the transit time with a test bolus of 2 mL and a dynamic region-of-interest analysis of the signal intensity at the level of the renal arteries. Image acquisition was then started 2 sec before the signal intensity peak. After subtraction, maximum-intensity-projection reconstructions and multiplanar reformations were created. The threshold for renal artery stenosis was defined as a 50% reduction in the diameter of the artery.
Catheter angiography and renal revascularization procedures were conducted in accordance with the guidelines of the Society of Interventional Radiology [22]. All patients had atherosclerotic renal artery stenosis measuring at least 60% in diameter with a transstenotic systolic gradient greater than 20 mm Hg. Stenting was performed if a residual stenosis greater than 30% or a transstenotic systolic residual gradient greater than 10 mm Hg was observed after angioplasty.
Clinical outcome on hypertension was assessed for at least 3 months (mean, 4 months) after angioplasty and (in some patients) stenting using the criteria defined by the Society of Interventional Radiology [23].
Cost Calculation
The costs of investigation and treatment per improved patient were
calculated for each imaging technique and for combined strategies in a
hypothetic population of 1,000 patients with a 30% prevalence of renal artery
stenosis. A prevalence of 30% was chosen because it is easily reached after a
good clinical screening [3,
24]. Cost calculation included
the costs of detection and treatment of renal artery stenosis, depending on
the predictive value of the test or combination of tests that would have been
used.
The diagnostic performance of each screening test for detecting renal artery stenosis was based on a recent meta-analysis [5] (Table 1). From these results, we calculated the number of true-positive, false-negative, true-negative, and false-positive results in patients with renal artery stenosis for every diagnostic technique or diagnostic criteria combination. Then, we determined the number of patients with a positive result on a screening test who would have improved after angioplasty using the positive predictive value of clinical improvement in a population with renal artery stenosis established by our previous study [17]. For MR angiography, this calculation was based on the subgroup of 57 patients.
|
For example, knowing that renal scintigraphy, as published in the meta-analysis [5], has 80.8% sensitivity and 78.0% specificity in depicting renal artery stenosis, we assumed that the result of the test would be positive in 396 patients (true-positive and false-positive results) and negative in 604 patients (true-negative and false-negative results), given a 30% prevalence of renal artery stenosis in a 1,000-patient population. Patients with a negative screening test result would not have undergone further examination. The 396 patients with a positive screening test result—242 patients with a true-positive result (i.e., with stenosis: 80.8% x 1,000 patients x 30% prevalence) and 154 patients with a false-positive result (i.e, without stenosis: [1–78.0%] x 1,000 patients x [1–30% prevalence]) would have then undergone angiography. Knowing that angiography has 100% sensitivity and specificity in depicting renal artery stenosis, we predicted that the 154 false-positive patients would have a negative test result and would not undergo angioplasty, whereas the 242 true-positive patients would be treated. Given a 58% positive predictive value of renal scintigraphy [17] for predicting clinical improvement after renal angioplasty, we predicted that a total of 141 patients (242.4 x 58.3%) would show clinical improvement after the procedure.
Meta-analysis sensitivity (Se) and specificity (Sp) of
the combined diagnostic criteria to detect a renal artery stenosis were
calculated as follows:
 |
 |
The costs for each screening and therapeutic procedure were calculated in Canadian dollars, using the reimbursement values published by the Ministry of Health and Social Services of Quebec, Canada [25] (Table 2). The costs for each diagnostic procedure included the use of diagnostic equipment, supplies, technician time, and physician fees. If the patient underwent angioplasty according to the analytic model, the costs of catheters, balloons, stents, and the hospital stay were calculated. We assumed a 30% rate of angioplasty and a 70% rate of angioplasty with stenting, these being the proportions observed in our study group. The costs were added for each diagnostic step of the analytic scheme. We finally divided the total costs by the number of improved or cured patients to determine the cost of each therapeutic success.
|
In our example, 1,000 patients underwent renal scintigraphy (1,000 x $544). Of the 396 patients with a positive result, the 154 with a false-positive result had angiography only (154 x $1,107), and the 242 with a true-positive result had angiography followed by angioplasty alone in 30% of cases and angioplasty with stent insertion in 70% ([242.4 x 30% x $3,372.50] + [242.4 x 70% x $5,126]). The total costs were added and divided by the 141 clinically improved patients ($1,829,506/141.4), giving a cost per improved patient of $12,939.
We know that our calculation method favors screening tests with poor sensitivity values in depicting renal artery stenosis because of the large number of patients with a false-negative result created. Some patients who are falsely categorized as having normal arteries or as lacking a therapeutic response predicted by the screening examination will not undergo further investigation or treatment. To take such cases into account, we calculated the number of patients with a false-negative finding who could have responded to treatment if they had undergone angioplasty.
The number of patients with a false-negative result who could have responded to treatment was calculated by multiplying the number of patients with such a result to detect a renal artery stenosis in the general population (from the meta-analysis) by the proportion of patients with a false-negative result who, based on the findings of our previous series, would have responded to treatment despite having a negative test result (1 - negative predictive value) [17].
When we combined a diagnostic test with a functional criterion (e.g., Doppler sonography and resistive index), the number of false-negative examinations was calculated as follows: (number of patients with a false-negative result on Doppler sonography who would have responded to treatment) + (number of patients with a false-negative result using a resistive index threshold who would have responded to treatment). The number of false-negative examinations linked to a resistive index threshold was calculated using data from our prior study [17] as follows: (true-positive Doppler examinations) x (percentage of negative examinations using a resistive index threshold) x (1 - negative predictive value of resistive index criteria).
This work was approved by our institutional ethics and research committees. Written informed consent was obtained from every patient in our previous study.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table 3 lists the number of true-positive, true-negative, false-positive, and false-negative examinations obtained with each technique for detecting renal artery stenosis based on the meta-analysis [5]. Positive and negative predictive values of each technique for predicting the clinical response (cured or improved) in the 74 patients who underwent angioplasty (alone or with stenting) of their renal artery stenosis are also listed. We also calculated the costs per improved patient for each type of examination with the assumption that a 1,000-patient population was screened, with a 30% prevalence of renal artery stenosis. In addition, Table 3 provides the number of patients wrongfully excluded from successful treatment (false-negative test results) by each screening approach. Note that the total number of patients who would have responded to treatment (improved patients ± patients with a false-negative result) reported in Table 3 varies from row to row. As discussed in the Materials and Methods section, the data used to estimate the relevant proportions were obtained from different sources, which explains these discrepancies.
|
Catheter angiography and MR angiography had perfect sensitivity for detecting patients with clinical improvement; the cost per improved patient was $16,580 for catheter angiography and $18,119 for MR angiography. These costs are explained by the direct cost of examinations and their poor specificity for predicting the therapeutic response, leading numerous patients to undergo clinically unsuccessful renal angioplasty.
Lower costs were observed with Doppler sonography and renal scintigraphy ($12,579 and $12,939, respectively), but Doppler sonography had a higher negative predictive value for identifying patients who could have responded to treatment. Consequently, if Doppler sonography had been performed in a population of 1,000 patients with a clinical suspicion of renovascular hypertension, 12 patients would have been excluded from successful treatment. If the group had been screened with renal scintigraphy, 29 patients would have been excluded.
After adding a resistive index threshold value to a positive finding on Doppler sonography as the criteria for patient selection, we observed a decrease in the negative predictive value of Doppler sonography for detecting therapeutic response but a proportionally greater increase in its positive predictive value, thus lowering the cost per improved patient. Combined criteria with the best cost-efficiency were positive findings on Doppler sonography and a bilateral resistive index of less than 0.75. According to these criteria, 32 patients would have been excluded from successful treatment, but the cost per improved patient was $10,149. The combination of unilateral measurement of the resistive index (stenosed kidney) with a positive finding on Doppler sonography was less cost-effective than bilateral measurement of the resistive index. Measurement of a unilateral resistive index after captopril administration combined with a positive finding on Doppler sonography improved the negative predictive value of the test compared with the baseline resistive index, but this improvement did not translate into a greater cost-efficiency.
The screening approach with the lowest cost per improved patient ($9,405) was achieved using the combination of renal length exceeding 9 cm and a positive result on Doppler sonography, but this strategy resulted in a high number of patients (71/1,000) being wrongfully excluded from successful treatment.
When MR angiography was combined with Doppler sonography or scintigraphy, there was little effect on the predictive values for the clinical response, but the cost increase was high ($19,413 for Doppler sonography and $21,288 for renal scintigraphy). The combination of resistive index criteria (bilateral index before captopril administration < 0.75) with a positive finding on MR angiography was less expensive ($17,592) than the last two combinations, but it was less cost-efficient than the combination of a positive Doppler finding with a resistive index threshold.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The 50% rate of clinical success observed in our series is within the lower range of the success rates reported in the literature on renal angioplasty of atherosclerotic stenosis, which varies between 40% and 70% [6–10]. It is well known that the presence of renal artery stenosis on catheter angiography is not sufficient to guarantee a good therapeutic response after renal angioplasty, as confirmed by the recently published results of randomized studies [12–14]. Furthermore, there is no correlation between the degree of angiographic stenosis and the therapeutic response after angioplasty [17].
MR angiography is an excellent screening test with which to detect renal artery stenosis, with a sensitivity and specificity of 96.4% and 92.2%, respectively. However, MR angiography does not provide information on the outcome after angioplasty, leading numerous patients to undergo unnecessary revascularization. We found that the positive predictive value of MR angiography for predicting clinical success after angioplasty was very low. Because MR angiography is an expensive examination and may lead to more unsuccessful revascularization procedures, its systematic use for screening patients with a clinical suspicion of renovascular disease results in a high cost-per-improved-patient ratio. The main advantage of MR angiography, due to its high negative predictive value, is the absence of false-negative examinations.
Doppler sonography has a relatively high sensitivity and specificity for detecting renal artery stenosis. In our series, its negative predictive value for detecting patients who would respond to angioplasty was very good, but its positive predictive value was quite low [17]. This can be explained by the fact that the criterion of a positive finding on a Doppler examination is based on the detection of renal artery stenosis and not on the prediction of clinical response after treatment. Even if the positive predictive value of Doppler sonography for predicting a clinical failure was low, this examination is not expensive. Consequently, the cost per improved patient was lower than that for MR angiography. The number of patients excluded from successful treatment in a population of 1,000 patients with a 30% prevalence of renal artery stenosis was acceptable (12/1,000). However, a high level of expertise in renal Doppler sonography is required to reproduce these results.
As reported previously, resistive index measurements play an important role in patient selection before angioplasty [16, 17, 25, 27]. Bilateral measurement of the resistive index was slightly more predictive than unilateral measurements, which emphasizes the importance of evaluating both sides. Interestingly, captopril administration did not bring any additional value when compared with baseline studies. Captopril lowered resistive index values and increased negative predictive values for predicting the clinical response. This effect was counterbalanced by a lower positive predictive value that was not cost-efficient.
The combination of a positive result on Doppler sonography with an averaged bilateral resistive index threshold value of less than 0.75 was a good compromise in terms of cost-efficiency, with a cost per improved patient of $10,149 and a false-negative rate of 32 per 1,000 patients. The threshold value of the resistive index recommended by Radermacher et al. [16] is even higher, but such a high value was not discriminative in our study. Indeed, only six (8%) of 74 patients in our series had a resistive index greater than 0.80 compared with 35 (27%) of 131 patients in their series [16, 17]. This discrepancy could be explained by different patient populations.
Patient selection based on a positive result on Doppler sonography combined with renal length exceeding 9 cm was the least expensive approach, but the drawback was a higher proportion of patients excluded from successful treatment.
In the literature, the sensitivity of renal scintigraphy for detecting renal artery stenosis varies between 34% and 93%, with a mean rate of 80.8% reported in the meta-analysis of Boudewijn et al. [5]. The predictive value of captopril-enhanced renal scintigraphy in identifying patients who will respond to angioplasty is also controversial. Correlation between a positive result on captopril-enhanced scintigraphy and improvement of hypertension after angioplasty has been reported in several noncontrolled studies [15, 28]. Other studies have suggested that captopril-enhanced renal scintigraphy was not useful for patient selection before angioplasty [9, 17, 29]. In our series, renal scintigraphy showed a poor negative predictive value (50%) for predicting clinical success. For a physiologic examination such as renal scintigraphy, we should expect a better clinical predictive value. In our calculation, after using the 80.8% sensitivity value for the first screening step of renal artery stenosis detection, the cost of renal scintigraphy was similar to that of Doppler sonography, but more patients would have been excluded from successful treatment (29 vs 12). If the costs were calculated as observed in our series, the number of false-negative examinations would have been much higher.
Treatment of hypertension has been proven to reduce long-term cardiovascular morbidity and renal function deterioration [30, 31]. The cost savings associated with endovascular treatment of renovascular hypertension are difficult to estimate. The limitations of current economic analyses have been stressed by several researchers and health economists [2]. Even the costs saved by reducing drug administration are hard to establish because of variations in the price of medications and market fluctuations [32]. Because of this lack of consensus, we did not calculate the costs associated with poorly controlled hypertension in patients with true renovascular hypertension and a false-negative examination. We preferred to calculate only the direct costs of screening tests and revascularization treatment to provide a comparison of the different screening techniques studied. The costs were calculated according to the reimbursements of the Canadian Ministry of Health and Social Services in Canadian dollars. These costs are lower than those in the United States. However, because the relative cost values for each examination in Canada and in the United States are similar, this would not alter our conclusions [33].
We are aware that our method of calculating costs favors examinations that are not very sensitive, mostly because of the high number of false-negative findings that results in excluding patients from the investigation process after the first screening test. Not treating patients who had false-negative results on screening tests involves costs that unfortunately cannot be well evaluated. For this reason, we calculated the number of patients with false-negative results who could have responded to treatment in a hypothetic 1,000-patient population with clinical criteria of renovascular hypertension.
We did not calculate the costs linked with potential complications of diagnostic and interventional procedures. Our sample size was calculated to evaluate the performance of each examination as a predictor of the therapeutic response after angioplasty. Considering the low incidence of complications, this sample size was not large enough to provide an accurate estimation of costs linked with complications. For this reason, we preferred to calculate only the costs directly related to the diagnosis and treatment of renovascular disease. Because complications are mainly related to renal angiography and angioplasty, this omission should not significantly influence the comparison of scintigraphy, Doppler sonography, and MR angiography.
For each screening procedure, we attempted to estimate both the proportion of patients who would have improved in a given scenario and the proportion of those who could have improved but who would have been denied treatment because of misclassification at the screening stage. This estimation was calculated by combining the performance of each diagnostic examination as a tool to detect a renal artery stenosis (on the basis of a meta-analysis [5]) and the predictive therapeutic values of these examinations (on the basis of our previous series [17]). The population of patients who had a positive screening result for the presence of renal artery stenosis according to the data of the meta-analysis probably is not exactly the same population that we enrolled in our study. However, given the limitations of the available data, we believe this approach is justified. First, by relying on the meta-analysis, we obtain a more precise estimate of the expected false-negative rate of a given screening test. On the other hand, our series provides the only available empirical evidence regarding the expected rate of response to treatment among subjects whose stenosis is "missed" by the test. However, the precision of the latter estimates is limited by the small number of patients in our series with a false-negative result, which was used to estimate the expected rates of treatment response for these patients. The benefits of combining and modeling data from different sources in the context of cost-effectiveness evaluations of interventions have been extensively reviewed by Buxton et al. [34].
In a cost-efficiency analysis published in 1998, Radermacher and Brunkhorst [35] reported cost-per-patient savings of 15,820 Deutsch marks after successful renal revascularization in patients with normal renal function. This calculation was based on a 10-year life expectancy and included cost savings in antihypertensive drug treatments and the prevention of end-stage renal artery disease. After conversion to Canadian dollars and adjustments for a mean inflation rate of 2% per year, the actual equivalent is $21,222. Compared with the cost per patient, screening for renovascular disease could be a good investment, especially if the most cost-effective imaging techniques are chosen. All the imaging strategies except one (MR angiography combined with renal scintigraphy) involve direct costs of less than $21,222. Using this cost-saving per patient, if we integrated the additional cost of patients with false-negative results into our calculation ($21,222), the cost per improved patient would be $14,832 for Doppler sonography; $16,728 for renal Doppler sonography with a resistive index threshold less than 0.75; $17,261 for renal scintigraphy; $16,580 for angiography; $18,119 for MR angiography; and $25,425 for Doppler sonography with a renal length exceeding 9 cm. These data suggest that in a population with resistance to drug therapy, Doppler sonography is still more cost-efficient than renal scintigraphy and MR angiography, but the cost differences are relatively small. In this population, criteria that include a resistive index threshold are not really cost-efficient.
Because there was no long-term follow-up in our population, we compared only direct costs without adjustments for quality of life and calculation of indirect costs. Recently, Carlos et al. [36] published a cost-effectiveness study that included quality-of-life adjustments. Enhanced medical treatment without imaging was compared with MR angiography, CT angiography, and angiography in a hypothetic cohort of individuals with medication-resistant hypertension. Performance of each examination in detecting renal artery stenosis was based on reports in the literature. They found that MR angiography, CT angiography, and angiography were superior to enhanced medical therapy alone. MR angiography was superior to CT angiography, but angiography was superior to MR angiography if indirect costs were included. MR angiography remained the preferred strategy if only the direct costs were calculated. Doppler sonography and renal scintigraphy were not included because they had no data on the predictive value of these tests [36].
In a population of patients who are resistant to drug therapy (hypertension that does not respond adequately to three or more drugs), we think that MR angiography can be justified as the first-choice option to revascularize the kidney. The presence of significant renal artery stenosis requires revascularization, and in this setting, it is important to minimize false-negative examinations. CT angiography or even catheter angiography can be an alternative to MR angiography, but many referring physicians are reluctant to use iodine contrast material in this population.
In a renovascular population in whom hypertension is controlled with medical therapy (up to three drugs), the benefit of renal angioplasty is more controversial [37]. In this setting, a patient can be treated medically and investigated with MR angiography only in the case of blood-pressure deterioration. If the referring physician prefers to examine the patient to prevent further deterioration of blood pressure and renal function or to reduce the medication, a functional imaging test, especially Doppler sonography combined with the resistive index threshold, can be a good alternative because such a test can be used to identify patients who will respond to angioplasty, and thus the test could be more cost-efficient. For those with negative results on screening tests, we suggest clinical follow-up and MR angiography in cases of deterioration of the hypertensive status.
Doppler sonography is the most cost-effective examination for screening renovascular hypertension. It could be used as an initial screening tool for medically controlled hypertensive patients with a clinical suspicion of renovascular disease. The rationale for using Doppler sonography in this setting would be to identify patients who could benefit from revascularization. In our experience, the sensitivity of renal scintigraphy does not seem sufficiently high to warrant its use as an alternative to Doppler sonography. MR angiography should be performed in hypertensive patients who are resistant to medical therapy to maximize the detection of renal artery stenosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. J. White Catheter-Based Therapy for Atherosclerotic Renal Artery Stenosis Circulation, March 21, 2006; 113(11): 1464 - 1473. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
