Renal angioplasty for treatment of hypertensive patients with fibromuscular dysplasia. No country for old men

The first percutaneous transluminal renal angioplasty (PTRA) was performed by Andreas Gruntzig and co-workers in 1978 in a 61 years old woman and yielded a rapid and marked fall in blood pressure associated with some increase in renal blood flow [1]. However, quite soon after that report, it became clear that the expectations of patients undergoing PTRA were not fulfilled always and particularly so for those with atherosclerotic renal artery stenosis (AS RAS). Indeed in one of the first meta-analyses published on this topic [2] reporting the experience of 10 centres, only 88 of 464 (19%) hypertensive patients with AS RAS were cured by renal angioplasty whereas this occurred in 50% of 193 patients with renal artery stenosis due to fibromuscular dysplasia (FMD RAS). In spite of the uncertainties on the benefit of PTRA in controlling blood pressure, the propagation of this procedure continued unrelentingly [3] without undergoing the rigorous scientific scrutiny that is usually applied to assess the benefit of pharmacological treatment. At present PTRA, because of its lower rate of morbidity and mortality, the lower cost and shorter hospitalization has, almost entirely, replaced surgery for treatment of renovascular hypertension (RVH) [4]. Actually, nowadays, the surgical approach is confined to the few cases in which PTRA technically fails or to patients with major atheromatous lesions that require extensive vascular intervention [5]. Until recently only few controlled, relatively underpowered studies have addressed the antihypertensive effects of PTRA in patients with AS RAS [6–8]. In 2009, the results of the Angioplasty and Stenting for Renal Artery Lesions (ASTRAL) study, the first large randomized controlled trial that compared the blood pressure and renal outcome of hypertensive patients with atherosclerotic renovascular disease treated with PTRA and medical therapy vs. medical therapy alone were reported [9]. This trial found no evidence of significant clinical benefit from revascularization casting further doubts on the value of a generalized application of PTRA in these patients. So far similar controlled studies are not available for hypertensive patients with FMD RAS. Yet in the recent and exhaustive systematic review by Trinquart et al.[10], encompassing 47 angioplasty studies carried out in 1616 patients with this disease, the overall rate of hypertension cure after revascularization was 46% with a combined rate for cure and improvement of 86.4%. However, this rate was markedly lower (36%) when cure was defined as a blood pressure below 140/90 mmHg without medical therapy and even lower (27%) when the analysis was restricted to the few studies in which 20 or more patients were enrolled. In spite of the declining rate of benefit when stringent criteria of cure are used, among the experts in the field there is consensus that PTRA remains the treatment of choice for hypertensive patients with FMD RAS [11]. In this issue of the Journal of Hypertension, Smit et al.[12] report the results of a prospective study in which they examined blood pressure and renal function in 51 patients with FMD RAS who underwent PTRA for difficult to treat hypertension. As controls they elected to use an equal number of patients, matched for age and blood pressure values, in whom angiography revealed normal renal arteries and were treated with medications only. The authors themselves admit that this somehow unorthodox control group represents a limitation of the study, justified by the difficulties of enrolling patients with FMD RAS who reject the option of revascularization. The findings of their study can be summarized as follows: after 12 months of follow-up the two groups achieved similar reductions in SBP and DBP (17/7 mmHg for patients treated with PTRA vs. 18/7 mmHg for those on medical therapy alone); in the latter group, a better control of pressure was obtained at the cost of a significant increase in daily need of antihypertensive drugs; when blood pressure changes were categorized the rate of hypertensive cure, defined as values below 140/90 mmHg without medical treatment, was 5% among patients treated with PTRA whereas 43% of these patients were classified as improved; no benefit in glomerular filtration rate (GFR), estimated from serum creatinine according to MDRD formula, were found in both groups; in revascularized patients a weak but significant correlation was found between baseline immunoreactive renin and the fall in SBP at 12 months. Are there plausible explanations to reconcile the few cases of blood pressure normalization observed by Smit et al.[12] with the much higher rate of cure reported in Trinquart's meta-analysis? That proposed by authors, that is, the more stringent criteria used in their study to define hypertension cure, seems insufficient because in Trinquart's review when the analysis was restricted to studies using the same criteria and with a comparable number of patients, blood pressure control was achieved in a substantially greater percentage of cases (27%). The age of patients examined in Smit's article offers a much more likely explanation. Indeed in their series, the average age was 52 years (including an 88-year-old patient) whereas in the 47 angioplasty studies quoted in Trinquart's article the average age was 42 years (range 28–59). Moreover meta-regression analysis of all these studies clearly showed an inverse relationship between age of patients and the rate of hypertension cure after revascularization. The observation that statistically the likelihood of cure is halved for the increase in mean age of 10 years implies that if at age 30 years normotension is expected to occur after PTRA in 60% of the cases, at age 50 years the rate of cure falls to 15% simply because of aging. The same reasoning applies to known duration of hypertension, the probability of cure being halved for an increase in time of exposure to high blood pressure of 5 years [10], a threshold that was very likely surpassed in many patients of Smit's cohort. In support of the view that age and duration of hypertension are crucial factors in determining the antihypertensive effect of PTRA, is the finding that in a series of 36 pediatric patients with FMD RAS, 94% of cases were cured after revascularization [13]. Several additional considerations can explain the low rate of cure found in Smit's cohort. First, the authors included in their analysis only patients with medial FMD RAS, that is those appearing at angiography as the typical string of beads. They did so because the inclusion of other forms of FMD may cause diagnostic errors. However, excluding the less frequent unifocal FMD may be a source of underestimation of the effects of PTRA, because it is known that dilatation of this type of renal artery stenosis (RAS) is associated with a greater fall in blood pressure [10]. Moreover, no attempt was made to quantify the degree of RAS and the decision of performing PTRA was taken only because blood pressure was difficult to treat. Estimating the severity of FMD and its hemodynamic consequences is admittedly difficult [14]. However, ultrasound evaluation of some distal velocimetric indices (acceleration and acceleration time in particular) are useful in this context in that they are significantly correlated with the degree of the stenosis and even more so in renal arteries with FMD [15]. Moreover quantification of arterial narrowing may help in detecting patients in whom a hemodynamically irrelevant RAS coexists with essential hypertension. This is far from being a rare situation because FMD are not infrequently found also in renal arteries of normotensive patients [16]. The possibility that in many patients of Smit's cohort the FMD RAS was of little relevance in causing the elevation of blood pressure, is indirectly supported by the finding that the average baseline immunoreactive active renin (20 mU/l) was well within the low-normal range of values for this kind of assay [17] and even more so considering that prior to PTRA 67% of these patients were on treatment with antagonists of the renin–angiotensin system which, presumably, should have markedly increased their naive renin levels. It is of interest that in spite of the limited number of patients in whom renin was measured (35/51) and of the alterations due to multiple pharmacological interferences, in Smit's article [12] a weak but significant relationship was found between plasma renin concentration prior to PTRA and the long-term reduction of SBP. The revival of interest for renin as a predictor of blood pressure outcome in response to revascularization is welcome. Indeed the evaluation of renin as an indicator of the hemodynamic relevance of RAS has been regrettably forgotten in recent years despite the evidence provided by elegant ‘in vivo’ studies showing the close relationship between the progressive reduction of renal blood flow and the increase in renin secretion [18]. The limited use of renin profiling in hypertensive patients with RAS is even more surprising if one considers the scarce consensus achieved by other predictors like the measurement of transtenotic pressure gradient (obviously hampered by technical drawbacks) or the evaluation of more esoteric biomarkers like BNP [19]. Unfortunately in Smit's report [12] renin measurements were not pursued further during follow-up. Indeed determination of renin pre and post-PTRA, in combination with ultrasound interrogation of the dilated artery, could raise the suspicion of restenosis, an event that in dilated FMD RAS is reported in 28% of cases [20] and, in some patients, may have reversed the blood pressure lowering initially induced by PTRA. A final comment deserves the observation that after PTRA, there were no changes in renal function. This finding goes along with that of six other studies in which the effect of revascularization with PTRA was examined in patients with FMD RAS and normal renal function at baseline [10]. The only study in which a benefit was found is that of Tegtmeyer et al.[21] in which mean serum creatinine fell from a baseline of 2.4–1.7 mg/dl in 14 patients followed for a mean of 33 months. When changes in renal function were categorized 86% of patients were classified as improved and the remaining 14% as stabilized. However, the absence of clearcut reductions in serum creatinine after PTRA does not necessarily imply the lack of benefit. Indeed small but relevant improvements of GFR can be missed simply because serum creatinine is quite a rough marker of overall function of both kidneys. We and others [22,23] have shown that when the changes in GFR following PTRA are estimated with techniques, which allow the evaluation of single kidney function like renal scintigraphy, the amelioration in GFR in the dilated kidney is partially counterbalanced by the reduction in the hyperfiltering contralateral kidney. The ongoing randomized Medical and Endovascular Treatment of Atherosclerotic and Artery Stenosis (METRAS) study [24] which among other goals, addresses the effects of PTRA on GFR using renal scintigraphy will hopefully clarify this issue. In summary, the study of Smit et al.[12], in spite of some limitations, has the merit of showing clearly that even in patients with FMD RAS when PTRA is applied merely on the basis of resistance to blood pressure treatment according to the so-called ‘oculostenotic reflex’ and without taking into due consideration critical factors like the age of patients and the duration of hypertension, the rate of cure is very low, just like in patients with AS RAS. However, rather than dismissing PTRA right away as a useless treatment, efforts should be directed to select patients with RAS who may still benefit from PTRA. It has been known for a long time that simple clinical signs like an abdominal bruit, and mild hypokalemia in a patient with moderate to severe diastolic hypertension may greatly increase the likelihood of true RVH [25]. In the experience of the Italian Group for the Study of Renovascular Hypertension, the prevalence of FMD RAS, angiographically proven, in a cohort of 459 hypertensive patients clinically selected was 14% and the rate of hypertension cure in those who underwent PTRA (mean age 37 years) was 31.3% with 37.4% of improvement [26]. The prediction of blood pressure outcome could be further enhanced if, prior to PTRA, patients were examined with ultrasound evaluation of acceleration and acceleration time and of renin profiling, possibly carried out when patients are off treatment with medications, which alter renin secretion. Using this approach, we achieved a high degree of accuracy in predicting the favourable or neutral effect of PTRA on blood pressure [27]. Treatment withdrawal may be cumbersome, but a similar recommendation has been put forward for assessing the aldosterone/renin ratio as screening test of primary hyperaldosteronism [28]. Moreover, the recent development of direct renin assays which are sufficiently accurate, much simpler and faster than conventional plasma renin activity [17], may provide clinicians with a reliable tool to pinpoint patients with FMD RAS in whom a timely PTRA should be carried out with a high likelihood of cure before the renal and vascular alterations induced by a long-standing hypertension prevent the beneficial effect of revascularization. ACKNOWLEDGEMENTS Conflicts of interest There are no conflicts of interest.