| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, University of Michigan, 1500 E Medical Center Dr.,
Ann Arbor, MI 48109-0030.
2 Department of Surgery, University of Michigan, Ann Arbor, MI 48109-0030.
3 Department of Internal Medicine, University of Michigan, Ann Arbor, MI
48109-0030.
Received April 30, 2003;
accepted after revision June 23, 2003.
Address correspondence R. C. Carlos.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A decision analytic model compared the life expectancy and incremental cost per life year using three initial diagnostic tests in a cohort of hypothetical individuals with medication-resistant hypertension over a range of renal artery stenosis probabilities: CT angiography (sensitivity, 96%; specificity, 96%; cost, $865); MR angiography (98%, 94%, $850); and conventional angiography (99%, 99%, $2,627). All imaging strategies were compared with a base case scenario mimicking the natural history of medication-resistant hypertension and with a scenario immediate enhanced medical therapy without prior imaging. Individuals without evidence of renal artery stenosis on initial testing underwent conventional angiography if enhanced medical therapy failed to control hypertension. Individuals diagnosed with renal artery stenosis on MR angiography required conventional angiography for definitive stent treatment ($11,1223). Blood pressure response to renal artery stenting or enhanced medical therapy varied according to blood pressure, as did the incidence of myocardial infarction and stroke resulting from hypertension. Patients who progressed to end-stage renal disease received dialysis ($60,000 per year). Quality-of-life adjustments were made for patients with hypertension, end-stage renal disease, myocardial infarction, and stroke. Short-term disutilities from undergoing an imaging test were included. The analysis accounted for direct costs derived from Medicare reimbursements and total costs derived from the literature.
RESULTS. All imaging strategies were cost-effective compared with enhanced medical therapy alone or with natural history. When only direct costs were considered, MR angiography was the preferred strategy, with conventional angiography as a cost-effective alternative to MR angiography. When total costs were considered, conventional angiography dominated all other strategies. Adjusting for quality of life decreased the incremental cost-effectiveness ratios, making an already competitive strategy a more favorable alternative to the base case. Adjusting for test-related disutility did not significantly influence the cost-effectiveness of any of the imaging tests. Despite marked variation in the key clinical and cost variables, MR angiography remained the most cost-effective strategy.
CONCLUSION. In the evaluation and treatment of medication-resistant hypertension, strategies that included preliminary imaging saved more lives than did the immediate institution of enhanced medical therapy at a lesser cost.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Conventional angiography is the standard of reference for diagnosis of renal artery stenosis and has been referred to as the gold standard. Besides morphologic assessment, gradient pressure measurements obtained during the procedure can be used to estimate hemodynamic significance of the stenosis. Furthermore, treatment with balloon angioplasty or stent placement may be undertaken, often within the same procedure. However, despite its status as the gold standard, to our knowledge, conventional angiography has not been validated in the literature. It is invasive, requiring direct puncture of the femoral artery to gain access to the arterial system, and intravascular complications, although rare, can be severe. In addition, the contrast agents required for opacification of the relevant vessels are known nephrotoxins.
Alternatives to conventional angiography include MR angiography, CT angiography, captopril scintigraphy, and renal sonography. These alternatives are noninvasive with variable sensitivities and specificities. MR angiography has been reported to have sensitivities of 88–100% and specificities of 75–100% [4–9] and is performed with nonnephrotoxic contrast agents. However, a failure rate of up to 10% has been reported because of technical factors and claustrophobia. CT angiography has reported sensitivities of 95–100%, although it requires iodinated contrast administration [10–14]. Another group of researchers found that conventional angiography was cost-effective compared with MR angiography [15]. However, that study omitted the quality of life component from the cost-effectiveness analysis.
Standard valuation analysis recommends incorporation of quality-of-life measurements, if available, in cost-effectiveness analyses [16]. Quality-of-life estimates adjust primarily for morbidity and mortality rates. Contributions to quality of life may also include the reassurance value of a test or the discomfort associated with undergoing a test, although these contributions are subordinate. For example, previous work determined that diagnostic MRI performed in patients with equivocal neurologic symptoms became cost-effective, compared with a no-test strategy, only after incorporation of the reassurance value of a negative result on MRI [17]. Swan et al. [18] and Swan and Langlotz [19] have studied the short-term discomfort that the patient experiences during the performance of a given diagnostic examination. This discomfort has been termed "short-term disutility" and has been directly quantified in individuals undergoing both MR angiography and conventional angiography for peripheral vascular disease. We revisit the issue of cost-effectiveness of imaging in individuals with suspected renovascular hypertension. The purpose of our study was to evaluate the contribution of quality of life and short-term disutilities to cost-effectiveness.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Decision Analytic Model
To determine the preferred initial diagnostic strategy of evaluation of
renal artery stenosis, we created a decision-analytic computer model (Figs.
1 and
2) simulating the diagnosis and
treatment of renal artery stenosis in a hypothetical cohort of patients with
suspected renovascular hypertension using DATA 3.5 (TreeAge Software,
Williamstown, MA). The analysis was conducted from the perspective of the
health care system, which would bear the additional cost of diagnosis, initial
treatment, and consequences of renovascular hypertension, as well as from the
societal perspective, including estimates of indirect costs primarily due to
lost productivity from morbidity and mortality.
|
|
Base case scenario.—The natural history of medication-resistant hypertension without treatment was modeled as the base case scenario. The hypothetical medication-resistant hypertension cohort did not receive additional diagnostic testing or enhanced medical therapy. Individuals entered a Markov model that accounted for blood-pressure-specific risks of myocardial infarction, stroke, chronic renal insufficiency requiring dialysis, and death.
Alternatives to the base case.—We considered the following alternatives to the base case scenario: enhanced medical therapy without preceding diagnostic testing and diagnostic testing followed by treatment (percutaneous transluminal angioplasty with stent placement) for patients who tested positive for renal artery stenosis or enhanced medical therapy in the absence of renal artery stenosis.
We considered three diagnostic testing options at initial presentation (Table 1). The physician could choose to perform an initial CT angiography, MR angiography, or conventional angiography. The results of the diagnostic examination were considered positive if a greater than 50% renal artery stenosis was found. If the results of the initial diagnostic test were positive, the individual underwent renal artery stenting, either in the same procedure as the diagnostic conventional angiography or in a conventional angiography performed after the CT angiography or MR angiography. If the results of the initial diagnostic test were negative, a third antihypertensive medication was prescribed, and the individual was reevaluated clinically after enhanced medical therapy.
|
Response to treatment.—Response to treatment (renal artery stenting or enhanced medical management) was stratified as improved (DBP < 90 mm Hg), somewhat improved (DBP, 90–110 mm Hg) or not improved (DBP > 110 mm Hg) [15]. Individuals who showed at least some improvement required no additional diagnostic testing or intervention. Individuals who showed no improvement received a third drug after renal artery stenting or a fourth drug after a trial of enhanced medication. If there was no response after the fourth antihypertensive medication, the individual's hypertension was considered treated but uncontrolled. At the end of year 1 (defined as the treatment year), the individuals entered into a Markov model that accounted for blood-pressure-specific risks of myocardial infarction, stroke, chronic renal insufficiency requiring dialysis, and death. The cohort was followed for life after renal artery stenting or initiation of enhanced medical therapy. The benefits incurred from renal artery stenting persisted for the follow-up period. We accounted for direct costs of diagnostic tests, antihypertensive medication, renal artery stenting, and the follow-up and consequences of comorbid conditions such as myocardial infarction, stroke, or chronic renal failure requiring dialysis.
Diagnostic Test Characteristics
Performance characteristics for diagnostic tests and other clinical
probabilities used in the simulation were abstracted from critical analysis of
the literature (Tables 1 and
2). The principal values for CT
angiography and MR angiography were derived from a meta-analysis comparing CT
angiography and MR angiography with the reference standard of conventional
angiography [21]. Morbidity
due to noninvasive testing results from iodinated-contrast–induced
nephropathy requiring dialysis. Incidence of nephropathy requiring dialysis
was derived from an analysis of the Blue Cross Blue Shield Percutaneous
Coronary Interventions database for southeast Michigan
[22].
|
Because conventional angiography has not been validated in the literature, we assumed a sensitivity of 99% and specificity of 99%. Current mortality rates of 0.04% and morbidity rates of 0.45% for conventional angiography have been estimated [15]. Morbidity after conventional angiography relates mainly to renal artery dissection [15], transient elevation of renal function tests requiring observation, and nephropathy requiring dialysis relating to the exposure to contrast agents [22–24]. Renal artery dissection in patients without renal artery stenosis, as a result of a diagnostic (but not therapeutic) conventional angiography, incurred an additional cost that was equivalent to that of renal artery stenting, because stent placement is the treatment of choice for dissections. Patients with renal artery stenosis and dissection after diagnostic conventional angiography did not incur the additional cost of stenting because this cost had already been incurred for treatment.
Treatment Effectiveness and Survival
A 5% incidence of renal artery rupture or dissection from renal artery
stenting was estimated from the study by Morris et al.
[25]. The restenosis rate
after renal artery stenting during the first year was 0.074%
[26]. Response to renal artery
stenting was derived from Yutan et al.
[26]; mortality rates after
renal artery stenting were derived from Ramsay and Waller
[27]; nephropathy requiring
dialysis after renal artery stenting was estimated from the study by Freeman
et al. [22]. Response to the
addition of a third medication in individuals with renal artery stenosis was
assumed to be slightly worse than the response to renal artery stenting, such
that improvement occurred in 10%, some improvement in 40%, and no improvement
in 50%. Response to the addition of a fourth medication in individuals with
renal artery stenosis was assumed to be worse still, with improvement
occurring in 5%, some improvement occurring in 10%, and no improvement in 85%.
These assumptions were derived from a previous cost-effectiveness analysis
[15].
Blood pressure–specific incidences of myocardial infarction, stroke, and chronic renal failure in the setting of renal artery stenosis are not empirically derivable from the literature. We derived estimates using the technique of Nelemans et al. [15]. Existing stroke or myocardial infarction increased the probability of new myocardial infarction or stroke threefold [28, 29]. Literature estimates provided the probabilities of immediate mortality from myocardial infarction and stroke [15]. Mortality rates from chronic renal failure were obtained from survival estimates of individuals on dialysis for renovascular disease [15].
Quality of Life and Short-Term Disutility
The time trade-off is a method of measuring patient preferences for
individual health states. The subjects are asked to consider the relative
amounts of (life) time they would be willing to trade to survive in various
health states. For example, how many years of life with myocardial infarction
are they willing to trade for 1 year of normal health. This preference is
termed the "utility for the health state."
Utilities for life-years lived were derived from a time trade-off study performed in a population of patients with myocardial infarction or chronic renal failure in Beaver Dam, WI [30], and from a time trade-off study performed in a population of stroke patients [31]. In combinations of conditions, the lowest quality adjustor was used because it was assumed that quality of life is unlikely to exceed the lowest adjustor given multiple medical conditions and is likely to be lower than the lowest individual value. For example, for an individual who had both a myocardial infarction (quality adjustor, 0.729) and a stroke (quality adjustor, 0.78), subsequent life-years lived would be adjusted using the rate for a myocardial infarction.
Test-dependent short-term disutilities of a hypothetical hypertensive patient undergoing MR angiography and conventional angiography were derived from a cohort of patients with peripheral vascular disease who underwent both MR angiography and conventional angiography [18, 19]; risk-based assessment using the time trade-off method was used to determine disutility. "Short-term disutility" was defined as the discomfort experienced by a patient, associated with having to undergo a diagnostic test that results in a transient decrease in quality of life, with a more invasive test resulting in a larger decrement. For MR angiography, test-related disutility was 0.0441; for conventional angiography, test-related disutility was 0.1153. Test-related disutility for CT angiography was estimated as being equal to MR angiography, likely underestimating its true disutility but avoiding bias against CT angiography in the absence of literature values. To incorporate the test-related disutility into a formal cost-effectiveness analysis, we calculated the willingness-to-pay as described by Swan et al. [18] and multiplied the test-related disutility by the difference between the qualities of life before and after treatment for hypertension.
Cost Estimates
Direct costs for each diagnostic and therapeutic procedure were based on
actual 1999 Medicare reimbursements for southeast Michigan
(Table 3) or from the
literature [32,
33]. Cost estimates included
all hospital, technical, and professional fees. Indirect costs for morbidity
due to myocardial infarction, stroke, and chronic renal failure were derived
from the 2003 Heart and Stroke Update from the American Heart Association
[33] and from Rodby et al.
[34]. Indirect cost for loss
of future earnings from premature death was derived from a human capital
valuation of a statistical life year by Hirth et al.
[35]. All costs were expressed
in 1999 United States dollars. All costs incurred and all benefits accrued in
the future were discounted by an assumed inflation rate of 3% per year.
|
Cost-Effectiveness Analysis
In each strategy, the clinical outcomes considered included life-years
lived, quality-adjusted life-years lived, and test-disutility adjusted
quality-adjusted life-years. Cost-effectiveness analysis compares two
strategies by calculating the incremental cost-effectiveness ratio, defined as
the incremental cost for each additional outcome of interest. In our analysis,
we calculated the incremental cost-effectiveness ratio by comparing the base
case scenario with enhanced medical therapy and with each of the three
diagnostic testing strategies, using each of the clinical outcomes described
above. A strategy was considered cost-effective compared with the base case
scenario if the incremental cost-effectiveness ratio was less than $50,000 per
additional quality-adjusted life year
[35]. A strategy was
considered dominant if it yielded more beneficial outcomes at a lesser cost
than the alternative.
Cost-effectiveness analysis was initially performed by accounting for direct costs only because the data regarding direct costs were more robust. However, because the true cost of illness includes indirect costs, such as lost wages due to morbidity or lost future earnings due to premature death, we also performed a cost-effectiveness analysis accounting for total costs.
Sensitivity Analysis
The cost-effectiveness model relies on best estimates that have some
uncertainty, either because of sampling variation or lack of published
studies. Sensitivity analyses allowed us to test the best- and worst-case
scenarios and to evaluate whether the preference for one strategy over another
would change on the basis of the values used
[36]. Because we initially
examined a variety of clinical outcome measures, the sensitivity analyses were
performed after adjusting for quality of life and test-related disutility.
Sensitivity analysis examined the impact of varying the performance
characteristics and costs for diagnostic techniques over a range of values
(Tables 1,
2,
3). If values were unavailable
in the literature, we conservatively estimated 50% of the base case value as a
lower limit and 200% of the base case value as the upper limit. Model
sensitivity to key clinical variables and to indirect costs was also
tested.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We initially accounted for direct costs only. Adjusting for health-related quality of life, we found that immediate enhanced medical therapy without preliminary imaging costs approximately $7,600 for each additional quality-adjusted life year gained (Table 4). Similarly, all three of the imaging strategies tested were cost-effective compared with the base case, given the commonly acknowledged threshold of $50,000 per quality-adjusted life-year for a "cost-effective" intervention. In the imaging arms, we subsequently assigned a quality-of-life cost to performing a diagnostic test by accounting for short-term disutilities experienced by the hypothetical cohort. Incorporation of these test-related disutilities contributed little to the analysis, increasing the incremental cost-effectiveness ratio by an average of $50 for each imaging strategy for each additional quality-adjusted life-year. Of all the alternatives to the base case, MR angiography was the most cost-effective (Fig. 3) with an incremental cost utility of approximately $6,000 per additional quality-adjusted life-year, although the maximum difference in the incremental cost-effectiveness ratio between imaging strategies was less than $2,000. Even medical therapy, as the most costly alternative to the base case, cost only an additional $11,000 for each additional quality-adjusted life-year.
|
|
In the evaluation of drug therapy, the investigational drug is best compared with the standard of care, rather than with a placebo. Similarly, although initial comparison with the base case scenario was performed, with the natural history analogous to that of a placebo, we further compared the three imaging strategies with enhanced medical therapy, which represented the standard of care. Conventional angiography costs an additional $1,500 for each additional quality-adjusted life-year gained over enhanced medical therapy alone. Furthermore, both CT and MR angiography dominated enhanced medical therapy alone, meaning that CT angiography and MR angiography strategies independently resulted in a longer life expectancy at a lower cost compared with enhanced medical therapy alone.
Comparing the imaging strategies, we found that MR angiography dominated CT angiography. Conventional angiography was cost-effective at $9,200 for each quality-adjusted life year saved over MR angiography.
Total-Cost Accounting
Comparing the alternative strategies to the base case scenario, all
alternatives dominated the base case after inclusion of indirect costs.
Comparing the three imaging strategies with the strategy of enhanced medical
therapy alone, we found that CT angiography and MR angiography were
cost-effective, with each additional quality-adjusted life-year costing less
than $16,000. Conventional angiography dominated enhanced medical therapy. In
fact, conventional angiography dominated all other strategies.
Sensitivity Analyses
When we accounted for direct costs, sensitivity analysis showed that the
model was robust despite significant variation in most of the key variables.
Preliminary analysis stratified the variables in descending order of impact.
We selected the variables that had the greatest influence on the
cost-effectiveness analysis, namely, probability of stenosis and cost of
dialysis. We also selected the most influential cost associated with
management of hypertension and its sequelae, which included the costs of
uncontrolled hypertension, yearly stroke management, and yearly myocardial
infarction management, as well as the indirect costs associated with
myocardial infarction, stroke, chronic renal failure, and loss of future
earnings due to premature mortality. We further evaluated the effects of the
cost and rate of response to renal artery stenting and the effect of
diagnostic test characteristics on the decision model.
No clinically relevant variable altered the results of the cost-effectiveness analysis. MR angiography remained the most cost-effective alternative to the base case scenario and continued to dominate the strategy of enhanced medical therapy alone, regardless of the variable examined. Conventional angiography continued to be a cost-effective alternative to MR angiography across all variables, except when the probability of stenosis was conservatively valued at less than 3%.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although screening for renal artery stenosis in the general hypertensive population may be impractical because of the low prevalence of disease, screening for renal artery stenosis in targeted populations is more feasible because of an increased incidence rate of renal artery stenosis associated with medication-resistant hypertension, renal failure, and a clinical history of atherosclerotic disease. Incidence rates of as much as 40% have been reported in these populations [36–38]. Sensitivity analyses of the current model show that at these prevalences, MR angiography remains the preferred strategy, with conventional angiography remaining cost-effective as an initial diagnostic test. Only at prevalences of less than 0.05% does conventional angiography exceed the cost-effectiveness threshold of $50,000 per quality-adjusted life-year. Preference for MR angiography continues at any tested disease prevalence, even at prevalences reported in the general hypertensive population. More recent data suggest that the incidence of renal artery stenosis in "asymptomatic" individuals has been markedly underestimated. In a prospective cohort undergoing coronary angiography, individuals with essentially normal renal function and excellent blood pressure control (clinical features not typically associated with a high prevalence of renal artery disease) showed 19% incidence of at least unilateral renal artery stenosis of greater than 50% [38].
The utility of screening individuals with renal artery stenosis depends on the utility of treatment of stenosis and subsequent prevention or delay of end-stage renal disease, stroke, and myocardial infarction. Whether earlier renal artery intervention in patients with identified renal artery stenosis would improve blood pressure control, preservation of renal function, survival, or other cardiovascular end points is currently not known. However, a decision analysis assessing the benefit of treating asymptomatic renal artery stenoses identified during coronary arteriography suggests that the incremental cost-effectiveness of prophylactic renal artery stenting is $12,466 per quality-adjusted life-year [39]. This analysis suggests that there may be an economic rationale for screening for renal artery stenosis, particularly in individuals with a high prevalence of disease, although our analysis suggests that MR angiography screening is cost-effective even at low disease prevalence. Decreased down-stream expenditure relating to management of cardiovascular sequelae from hypertension recoups the initial increased expenditure for instituting preliminary MR angiography before enhanced medical treatment.
Residual lifetime risk for hypertension for middle-aged and elderly individuals derived from the Framingham Heart Study is 90%, indicating a huge public burden [40]. The best noninvasive method for detecting underlying renal stenosis in a high-risk hypertensive individual is debated. A recent meta-analysis summarized and compared the validity of CT angiography, gadolinium-enhanced MR angiography, sonography, captopril renography, and the angiotensin-converting enzyme captopril challenge test for the diagnosis of renal artery stenosis in patients with suspected renovascular disease [21]. Summary receiver operating characteristic (ROC) analysis showed that CT angiography and MR angiography performed the best with areas-under-the-ROC curve (Az) of 0.99 were compared with sonography (Az = 0.93), captopril renal scintigraphy (Az = 0.92), and angiotensin-converting enzyme captopril challenge test (Az = 0.72). Previous decision analysis further suggested that despite the diagnostic effectiveness of MR angiography, at prevailing costs, a diagnostic strategy with initial MR angiography is not cost-effective compared with a strategy of initial angiography [15]. Indeed, the MR angiography strategy was dominated by other strategies—that is, the MR angiography strategy resulted in fewer life-years saved at an increased cost. However, this decision analysis only evaluated prolongation of life and did not account for differences in quality of life of patients with controlled hypertension, compared with patients who have suffered a cardiovascular event resulting from renovascular disease. A widely recommended approach is to measure health outcomes in terms of quality-adjusted life-years to incorporate both the prolongation and quality of life [16, 41]. Quality-adjusted life-years capture both reduced morbidity and reduced mortality rates, and they incorporate the value (or preference) that patients have for different outcomes [42]. In our analysis, incorporation of quality-of-life adjustors resulted in a competitive MR angiography strategy. Nonetheless, conventional angiography first remained a costeffective alternative.
Preferences for health states usually refer to long-term permanent health outcomes, such as stroke. However, individuals place value even on temporary events, including discomfort (or disutility) associated with diagnostic testing. Swan et al. [18] derived the relative disutility of MR angiography and conventional angiography in a group of patients with underlying peripheral vascular disease. As expected, patients preferred MR angiography over conventional angiography and rated a higher disutility with conventional angiography [18]. In a cost-effectiveness analysis comparing MR angiography and conventional angiography in patients with peripheral vascular disease, the result without incorporation of relative disutilities showed that MR angiography was not cost-effective compared with conventional angiography (given performance characteristics of MR angiography in 1995). However, after incorporation of disutilities, the cost-effectiveness ratio decreased to a reasonable $12,967 per quality-adjusted life-year [19]. We used the same disutilities as those measured by Swan et al. Thus, it became more costly to perform conventional angiography first, but adjusting for test-related discomfort did not change the model conclusions, and conventional angiography remained cost-effective. Of course, test-related disutilities in patients with peripheral vascular disease likely underestimate test-related disutilities in patients with renovascular hypertension, because hypertension is often clinically silent with little or no change in quality of life until a resultant catastrophic cardiovascular sequela occurs. In fact, the current analysis shows that under a wide range of assumptions, identification on conventional angiography of individuals with renal artery stenosis increases the number of additional life-years saved at a surprisingly affordable incremental cost compared with MR angiography.
Gold et al. [16] advocate performing the analysis from the societal perspective accounting for the true cost of illness. The cost of illness reflects not just the direct costs incurred by the patient or the health care system, but the costs to society due to lost productivity from morbidity or early mortality. We incorporated indirect costs in the latter iteration of the decision model, resulting in the dominance of conventional angiography compared with all other strategies. This dominance highlights the inordinate cost of hypertensive cardiovascular sequelae to society, particularly the cost of early mortality.
The model does have the following limitations: other diagnostic modalities such as duplex Doppler sonography or captopril scintigraphy were not considered because we incorporated a clinical decision rule with a predictive value similar to that of duplex Doppler sonography or captopril scintigraphy that substituted for these diagnostic tests. The use of surgical intervention was not examined because a recent analysis of percutaneous transluminal angioplasty compared with surgery showed no significant difference in outcome between the two approaches. Furthermore, renal artery stenting is becoming the standard of care in treating renal artery stenosis, and the model reflects this shifting clinical practice. Given the uncertainty of the probabilities of morbidity and mortality resulting from untreated renovascular hypertension, we tested a range of reasonable probabilities for all relevant clinical variables. The conclusion of the model is remarkably robust across all reasonable ranges of clinical probabilities.
In general, the conclusion of a cost-effective strategy does not necessarily imply institution of that strategy. For example, a health care system or third-party payer may be more interested in implementing an imaging strategy starting with MR angiography as the most cost-effective strategy from their perspective. Alternatively, a facility without access to MR angiography or conventional angiography may initiate immediate enhanced medical therapy, despite its increased cost compared with MR angiography. National policy makers may be more interested in implementing an imaging strategy with conventional angiography, given its dominance from a societal perspective. Or we, as a society, may choose to do nothing and spend the limited resources on other interventions that may save more lives at a more reasonable cost. The current cost-effectiveness analysis merely illustrates the trade-offs between increased expenditure and increased life expectancy.
In conclusion, the decision analysis model shows that strategies that include preliminary imaging saved more lives compared with immediate institution of enhanced medical therapy, at an extremely reasonable cost, in the evaluation and treatment of medication-resistant hypertension. Further, MR angiography is the preferred imaging alternative in the evaluation of renovascular hypertension when accounting for direct costs, of the strategies tested, although conventional angiography is a cost-effective alternative. Considering the total cost of illness, we found that conventional angiography dominated all other strategies.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. van Helvoort-Postulart, C. D. Dirksen, P. J. Nelemans, A. A. Kroon, A. G. H. Kessels, P. W. de Leeuw, G. B. C. Vasbinder, J. M. A. van Engelshoven, and M. G. M. Hunink Renal Artery Stenosis: Cost-effectiveness of Diagnosis and Treatment Radiology, August 1, 2007; 244(2): 505 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Vrtiska, J. G. Fletcher, and C. H. McCollough State-of-the-Art Imaging With 64-Channel Multidetector CT Angiography Perspectives in Vascular Surgery and Endovascular Therapy, March 1, 2005; 17(1): 3 - 8. [Abstract] [PDF]
|
|
|
|
 |
|
 J. P. Bolduc, V. L. Oliva, E. Therasse, M.-F. Giroux, L. Bouchard, P. Perreault, A. Cliche, and G. Soulez Diagnosis and Treatment of Renovascular Hypertension: A Cost-Benefit Analysis Am. J. Roentgenol., March 1, 2005; 184(3): 931 - 937. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
