Journal of Human Hypertension (1999) 13, 217–220  1999 Stockton Press. All rights reserved 0950-9240/99 $12.00 http://www.stockton-press.co.uk/jhh COMMENTARY Atherosclerotic renal artery stenosis: is it worth diagnosing?

KS Eardley and GW Lipkin Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH, UK

Atherosclerotic renal artery stenosis (ARAS) is the com- ostial. Studies of revascularisation with angioplasty monest cause of secondary hypertension and is the show poor short and long term patency rates. Renal cause of end stage renal failure in up to 20% of patients artery stenting leads to high initial technical success starting dialysis. Associated with it is a high morbidity and long term patency. Recent randomised controlled and appalling mortality. The aetiology of ischaemic trials in patients with renovascular hypertension dem- nephropathy is complex and is not simply related to onstrate no clear benefit of adequate revascularisation renal artery narrowing. Captopril renography is sensi- over medical therapy. Renal artery stenting for renal tive and specific for diagnosing ARAS in patients with protection in ARAS appears more encouraging and cur- normal renal function. In those with renal impairment rent randomised controlled trials are in progress to gadolinium-enhanced MRA or spiral CT angiography answer the question definitively. clearly define renal anatomy. Over 80% of ARAS is

Keywords: renal artery stenosis; diagnosis; revascularisation; hypertension; renal failure

Introduction sponding to a stenosis of 70–80%.7 Renal oxygen delivery is far in excess of demand and therefore Atherosclerotic renal artery stenosis (ARAS) is the true ischaemia is likely only with near occlusive commonest cause of secondary hypertension. The stenosis. In support of this, ipsilateral kidney func- prevalence of atheromatous renovascular hyperten- tion does not differ from the contralateral side in 1 sion (RVH) is reported to be 1–5%, rising to 30% 8 2 patients with unilateral ARAS. Examination of in those presenting with malignant phase. Where renal histology in patients with ARAS gives clues to critical stenoses subtend the whole renal mass other mechanisms involved.7 Small vessel plugging (bilateral disease or stenosis of a solitary functioning from atheremboli is common. Marked intrarenal kidney) ARAS may lead to progressive ischaemic 3 arterial narrowing and pre-existing hypertensive renal failure. Prospective renal angiographic stud- nephrosclerosis frequently coexist. Nevertheless, by ies in patients starting dialysis over the age of 50 4 analogy with fibromuscular disease, stenosis can indicate ARAS to be the cause in 10–20% of cases. cause hypertension and renal artery occlusion in Furthermore, it is implicated in a similar proportion patients with ARAS almost always leads to loss of of patients presenting with acute renal failure, usu- renal function.8 ally in association with the introduction of angioten- sin-converting enzyme inhibitors.4 In our aging population with falling coronary mortality, the Is ARAS progressive? prevalence of ARAS is rising.4 Is ARAS worth diagnosing? The answers to the ARAS is progressive and associated with loss of following questions should help us decide. renal function. A large retrospective angiographic study reported that 44% of the renal artery stenoses progressed, 16% to occlusion, over a mean period of Is stenosis the main problem? 52 months. High-grade stenosis (greater than 75%) represented an imminent risk to renal function with A ‘significant’ stenosis is defined as luminal nar- 9 rowing of greater than 50% in most studies.5,6 This 39% progressing to occlusion over 13 months. The rate of progression is surprisingly linear at 4– definition probably greatly overestimates the preva- 9,10 lence of renovascular hypertension and ischaemic 12%/year. Recently the use of duplex ultrasound has provided prospective data on the natural history renal failure due to ARAS. Renal autoregulation Ͻ maintains constant renal blood flow down to a mean of ARAS. Progression from 60% stenosis to 60% perfusion pressure of 70–80 mm Hg, usually corre- or greater was 23% over 1 year and 42% at 2 years. Of those with Ͼ60% stenosis initially, 5 and 11% subsequently occluded the renal artery by 1 and 2 years respectively.11 Renal atrophy (Ͼ1 cm decrease Correspondence: Dr KS Eardley, Specialist Registrar in Nephrol- ogy, Queen Elizabeth Hospital, Edgbaston, Birmingham B15 of renal length) occurs in 21% of kidneys with 2TH, UK greater than 60% renal artery stenosis at baseline Received 30 October 1998; accepted 2 December 1998 and is associated with loss of renal function.12 Atherosclerotic renal artery stenosis KS Eardley and GW Lipkin 218 Decline in renal function in patients with ARAS is benefit by aggressive treatment of known cardio- dependent on the initial angiographic findings. At 2 vascular risk factors such as hypertension and elev- years, 3% of patients with unilateral stenosis will ated plasma lipids.20,21 have reached end stage renal failure (ESRF) as com- pared with 18% of those with bilateral ARAS and Does revascularisation beneficially affect 55% of those with renal artery occlusion and contra- lateral stenosis.13 However, the rate of decline in hypertension or renal failure? renal function is variable and not linear. Many ARAS is a disease of the aorta. In over 80% of cases patients with severe bilateral ARAS maintain stable the stenosis is produced by encroachment of aortic renal function over 5 years although the median atheromatous plaques at the renal artery ostium annual loss of GFR is 4 (1–8) ml/min with a cumu- (within 1 cm of aortic lumen).22 Thirty percent are lative requirement for dialysis of 30%.14 bilateral.3 Not surprisingly this impacts on the suc- cess and complication rates of attempted revascu- Can we screen for it? larisation. In major centres surgical revascularis- ation is associated with a 95% technical success. In order to identify patients with significant ARAS Perioperative mortality in this elderly atheroscler- who may benefit from intervention one must have otic population is significant at 2.5–6%.3 In view of a screening test that is non-invasive, sensitive and this, percutaneous transluminal renal artery angio- specific. The gold standard test is intra-arterial digi- plasty (PTRA) was introduced. PTRA successfully tal subtraction angiography. However, there are risks dilates non-ostial ARAS in 75–94% of cases with a associated with the use of contrast and of arterial 1 year restenosis rate of 10–30%.23 For the more fre- puncture. Captopril renography is non-invasive and quent ostial lesions initial success is achieved in may define functional significance. The sensitivity only half (due to elastic recoil) with half of these and specificity is high for detecting unilateral ARAS restenosing within 6 months.24 These technical (Ͼ90%) but its usefulness is markedly reduced in problems have been overcome by the use of endolu- the presence of renal impairment and bilateral dis- minal stents. Several series report initial technical ease.15 Spiral CT and gadolinium-enhanced mag- success of over 90% with less than 20% restenosis netic resonance angiography (MRA) are both non- over 2 years.5,19,24–26 invasive and provide excellent images of the renal A clear benefit of revascularisation over medical arteries. The sensitivity and specificity of both tests management of atheromatous renovascular hyper- for the detection of ARAS is in excess of 90%. Spiral tension has not been demonstrated. Ramsay CT angiography requires large volumes of intra- reviewed 10 uncontrolled series of PTRA for athero- venous contrast with the risk of nephropathy in matous RVH. Hypertension was cured in only 19% patients with impaired renal function. Gadolinium and improved (reduced number of anti- is not nephrotoxic and MRA may provide infor- hypertensives) in 52%.27 It has been argued that low mation on renal function and trans-stenotic press- technical success of PTRA explains the poor results. ure gradients.16 However, the results of three recent trials of renal artery stenting in 231 patients are no better. Stenting Does ARAS independently predict cured hypertension in 0–16%, with improvement (mean reduction of half of one hypotensive drug) in patient mortality? around half of patients.19,25,26 Even the modest Patients with ARAS frequently have evidence of improvement in blood pressure may be overesti- widespread atherosclerotic disease. Prospective mated in these uncontrolled studies as recorded studies of renal angiography have been performed at blood pressure in patients with RVH spontaneously the same time as angiography for symptomatic falls with follow up.28 atheromatous vascular disease elsewhere. The Three prospective randomised controlled trials prevalence of significant ARAS is 45–59%, 28–38%, comparing PTRA +/− stenting have recently and 15–23% in patients investigated for peripheral reported. The Scottish and Newcastle study fol- vascular, aorto-occlusive, and coronary artery dis- lowed 55 patients randomised to PTRA (36% ease respectively.4 restenosis) or medical therapy. No patient was Not surprisingly, cardiovascular mortality in these cured. Angioplasty resulted in a modest reduction patients is very high. ARAS is an independent risk in systolic blood pressure in those patients with factor for mortality which correlates with the num- bilateral atheromatous stenosis. Major procedural ber and severity of renal arteries stenosed.17 Hyper- complications occurred in 20%.28 The EMMA study tensive patients with ARAS have a 5 and 10 year group randomised 49 patients with proven haemo- survival rate, 10% less than those with essential dynamically significant ARAS to PTRA +/− stent or hypertension.18 For those with renal impairment medical treatment. At 6 months blood pressure did due to ARAS the survival is appalling. The prob- not differ between the groups although median ability of survival at 3 years being 92% with normal number of antihypertensive drugs fell by 0.8 per renal function, 74% with mild and 52% in those patient in the intervention group.29 Similarly, the with more severe renal failure.19 Survival on dialysis Dutch Renal Artery Stenosis Intervention Cooperat- is worse than for any other patient group. At 2 years ive Study found no benefit of intervention on blood nearly half will have died and at 5 years only 12% pressure at 3 months.6 The reasons for lack of blood will be alive.4 By extrapolation from other high risk pressure response are unclear but probably include patient groups one might anticipate a significant pre-existing essential hypertension, inclusion of Atherosclerotic renal artery stenosis KS Eardley and GW Lipkin 219 patients with incidental ARAS, renal impairment renovascular hypertension in patients with grade III or and failure of reversal of long standing adaptive IV hypertensive retinopathy. N Engl J Med 1979; 301: arterial changes. 1273–1276. In recent years the focus of attention has moved 3 Greco BA, Breyer JA. Atherosclerotic ischaemic renal to revascularisation as a means of halting the pro- disease. Am J Kidney Dis 1997; 29: 167–187. 4 Scoble JE. The epidemiology and clinical manifes- gression of ischaemic renal failure due to severe tations of atherosclerotic renal disease. In: Novick A, bilateral ARAS. Prevention of dialysis has major Scoble J, Hamilton G (eds). Renal Vascular Disease. resource and quality of life implications. In selected WB Saunders: London, 1996, pp 303–314. cases renal revascularisation undoubtedly improves 5 Harden PN et al. Effect of renal artery stenting on pro- renal function. Hansen et al30 reports 20 patients gression of renovascular renal failure. Lancet 1997; with bilateral renal artery occlusion who recovered 349: 1133–1136. dialysis independent renal function following surgi- 6 Van Jaarsveld BC et al. The Dutch renal artery stenosis cal treatment. Renal parenchymal viability was intervention cooperative (DRASTIC) study. Abstract maintained by collateral blood supply.30 A recent presented at the International Society of Hypertension review of eight surgical series indicated postoper- meeting, Amsterdam 1998. 3 7 Textor SC, Smith-Powell L. Pathophysiology of renal ative improvement in renal function in around half. failure in ischaemic renal disease. In: Novick A, Scoble Despite the high rate of restenosis PTRA may also J, Hamilton G (eds). Renal Vascular Disease. WB Saun- be of benefit. In a review of six series improvement ders: London, 1996, pp 289–304. and stabilisation of renal function occurred in 40 8 Farmer CKT, Reidy J, Cook G, Scoble JE. Individual and 30% of patients respectively.3 Major compli- kidney function is not dictated by the presence of cations occurred in 3–10% and procedural mortality stenosis in atherosclerotic nephropathy. Abstract was 1–3%. In patients with renal failure renal artery presented at the Renal Asociation, Autumn meeting. stenting results in durable technical success. Two London 1998. recent studies of renal artery stenting demonstrate 9 Schreiber MJ, Pohl MA, Novick AL. The natural his- significant slowing of the rate of progression of CRF tory of atherosclerotic and fibrous renal artery disease. 5,19 Urol Clin North Am 1984; 11: 383–392. up to 4 years. 10 Tollefson DFJ, Ernst CB. Natural history of atheroscler- Some factors appear to be predictive of stabilis- otic renal artery stenosis associated with aortic dis- ation or improvement in renal function following ease. J Vasc Surg 1991; 14: 327–333. revascularisation. These include a rapid decline in 11 Zierler RE, Bergelin RO, Isaacson JA, Strandness DE. renal function and preoperative serum creatinine Natural history of atherosclerotic renal artery stenosis: less than 300 ␮mol/l. Preintervention renal biopsy A prospective study with duplex ultrasonagraphy. has been used by some to assess the degree of irre- J Vasc Surg 1994; 19: 250–258. versible parenchymal damage. Assessment of indi- 12 Caps MT et al. Risk of atrophy in kidneys with atheros- vidual renal function may guide the appropriate kid- clerotic renal artery stenosis. Kidney Int 1998; 53: ney to revascularise. This may be important as there 735–742. seems to be a discrepancy between renal size and 13 Connolly JO et al. Presentation, clinical features, and function.31 outcome in different patterns of atherosclerotic renov- ascular disease. Quart J Med 1994; 87: 413–421. Interventional treatment for ARAS may have 14 Baboolal K, Evans L, Moore RH. Incidence of endstage effects beyond the kidney. ‘Flash’ pulmonary oed- renal disease in medically treated patients with severe ema may occur in patients with severe bilateral bilateral atherosclerotic renovascular disease. Am J ARAS despite normal left ventricular function and Kidney Dis 1998; 31: 971–977. is cured by adequate revascularisation.32 A recent 15 Nally JV. Captopril renography. In: Novick A, Scoble report of improved angina and heart failure status J, Hamilton G (eds). Renal Vascular Disease. WB Saun- in patients undergoing renal artery stenting for ders: London, 1996, pp 195–204. ARAS is intriguing.33 16 Olbricht CJ, Arlart IP. Magnetic resonance angiogra- Morbidity related to the stenosis in patients with phy – the procedure of choice to diagnose renal artery atheromatous renovascular disease has probably stenosis? NDT 1998; 13: 1620–1622. been overestimated. Identification should lead to 17 Conlon PJ et al. Survival in renal vascular disease. J Am Soc Nephrol 1998; 9: 252–256. aggressive cardiovascular risk factor intervention in 18 Isles C et al. Survival associated with renovascular dis- an attempt to reduce the appalling associated mor- ease in Glasgow and Newcastle: A collaborative study. tality. ‘Flash’ pulmonary oedema merits revascu- Scot Med J 1990; 35: 70–73. larisation. However, it appears that intervention for 19 Dorros G et al. Four year follow up of Palmaz-Schatz atheromatous RVH is a drug sparing exercise with a stent renovascularisation as treatment for atheroscler- significant complication rate and should be reserved otic renal artery stenosis. Circulation 1998; 98: 642– for those patients with truly drug resistant hyperten- 647. sion. Percutaneous stenting prevents occlusion and 20 The systolic hypertension in the elderly program will hopefully reduce the need for dialysis in selec- (SHEP) co-operative research group prevention of ted patients although the results of randomised con- stroke by antihypertensive drug treatment: Final trolled trials are awaited. results of SHEP. JAMA 1991; 265: 3255–3264. 21 Scandanavian Simvastatin Survival Study Group. Ran- domised trial of cholesterol lowering in 4444 patients References with coronary heart disease: the Scandanavian Simva- statin Survival Study (4S). Lancet 1994; 333: 1308– 1 Derkx FH, Schalekamp MA. Renal artery stenosis and 1312. hypertension. Lancet 1994; 344: 237–239. 22 Kaatee R et al. Atherosclerotic renal artery stenosis: 2 Davis BA, Crook JE, Vestal RE, Oates JA. Prevalence of ostial or truncal? Radiology 1996; 199: 637–640. Atherosclerotic renal artery stenosis KS Eardley and GW Lipkin 220 23 Sos TA. Angioplasty for the treatment of azotemia and 28 Webster J et al. Randomised comparison of percu- renovascular hypertension in atherosclerotic renal taneous angioplasty vs continued medical therapy for artery disease. Circulation 1991; 83 (Suppl I): I-162– hypertensive patients with atheromatous renal artery I-166. stenosis. J Hum Hypertens 1998; 31: 823–829. 24 Beutler JJ et al. Randomised clinical trial between stent 29 Plouin PF, Chatellier G, Darne B, Raynaud A. Blood placement and balloon angioplasty in atherosclerotic pressure outcome of angioplasty in atherosclerotic renovascular hypertension. Abstract presented at the renal artery stenosis. Hypertension 1998; 31: 823–829. International Society of Hypertension meeting, 30 Hansen JF et al. Contemporary surgical management of Amsterdam 1998. renovascular disease. J Vasc Surg 1992; 16: 319–331. 25 Blum U et al. Treatment of ostial renal artery stenoses 31 Scoble JE, Look GJR. Individual kidney function in with vascular endoprotheses after unsuccessful bal- atherosclerotic nephropathy. NDT 1998; 13: 842–844. loon angioplasty. N Eng J Med 1997; 336: 459–465. 32 Pickering TG et al. Recurrent pulmonary oedema in 26 White CJ et al. Renal artery stent placement: Utility in hypertension due to bilateral renal artery stenosis: lesions difficult to treat with balloon angioplasty. JAm Treatment by angioplasty or surgical revascularisation. Coll Cardiol 1997; 30: 1445–1450. Lancet 1988; ii: 551–552. 27 Ramsay LE, Waller PC. Blood pressure response to per- 33 Khosla S. Effects of renal artery stent implantation in cutaneous transluminal angioplasty for renovascular patients with unstable angina or congestive heart fail- hypertension: An overview of published series. BMJ ure. Am J Cardiol 1997; 80: 363–366. 1990; 300: 569–572.