The effect of percutaneous dilatation of renal arterial stenosis on captopril renography in hypertension

Abstract

Background. The clinical effects of percutaneous transluminal renal artery angioplasty (PTRA) in patients with renal vascular stenosis and hypertension is controversial. Methods. We consecutively recruited all 23 patients referred for evaluation of renovascular hypertension that eventually underwent unilateral PTRA, to be investigated with captopril MAG3 renography (CR), both before and after the endovascular procedure. Data were evaluated on an intention‐to‐treat basis. Results. We found that the relative MAG3 clearance of the stenotic kidney increased (from 29.9±14% to 35.1±14%, p = 0.01) and that the creatinine levels fell following the intervention (from 110±19 to 99±17 µmol/l, p = 0.0003). Blood pressure levels were also lowered (from 173±32/93±17 to 158±31/86±15 mmHg, p<0.006) while the mean number of anti‐hypertensive drugs was unchanged following PTRA (2.9±1.4 before and 2.8±1.3 drugs after the intervention, respectively, p = 0.6). Conclusion. This prospective trial showed statistically significant improvements of individual kidney function as measured by CR and blood pressure in subjects with suspected renovascular hypertension treated with PTRA. Although the endovascular procedure was found to be safe, the magnitude of the absolute improvements was rather modest.

• Captopril
• hypertension
• outcome
• renal artery stenosis
• renography

Introduction

The clinical effects of percutaneous transluminal renal artery angioplasty (PTRA) on kidney function in patients with renal vascular stenosis and hypertension is controversial [1–3]. Although the endovascular techniques have improved, in particular with the introduction of stenting, it is still not clear whether such intervention actually improves the overall renal function. Earlier studies have shown favourable effects on blood pressure (BP) control following PTRA, but very few of these were prospective and analysed on an intention‐to‐treat basis [2]. At the University Hospital of Linkoping, captopril renography has been the preferred technique for screening for renovascular hypertension for more than a decade [4]. In 1996, 99mTc‐mercapto‐acetyl‐triglycin (MAG3) was introduced as renographic agent in Linkoping. Since then, all patients referred for evaluation of renovascular hypertension were investigated with a standardized MAG3 renography. In order to evaluate the prognostic information using this technique, and in particular to evaluate the clinic effects of PTRA, all subjects undergoing PTRA were re‐evaluated with MAG3 renography after the PTRA procedure. Both the initial and the follow‐up renographic examinations were performed after the patient had ingested 25 mg captopril, on top of existing medication, to increase the sensitivity to detect signs of renal artery stenosis, a study design that to our knowledge has not been used previously. Herein we present the results of unilateral PTRA on renographic MAG3 clearance and BP in 23 subjects with renal artery stenosis analysed prospectively on an intention‐to‐treat basis.

Methods and patients

Patients

All subjects that were referred to the university hospital for evaluation of renovascular hypertension and who subsequently underwent unilateral PTRA were consecutively recruited and followed prospectively. The patients all had renographic findings indicative of renovascular hypertensions in addition to an elevated BP [4] and subsequently underwent renal arterial angiography. When the renal angiography confirmed the suspicion of significant (>50% diameter) stenosis, there was an immediate attempt to correct this by PTRA, with or without stenting.

Captopril renography

All patients were investigated with captopril stimulated renographic examinations (captopril renography, CR). Hydration was performed the hour before the CR by drinking 10 ml/kg body weight water. One hour before the CR test the patient received a crushed 25 mg captopril pill (Capoten®, Bristol‐Myers Squibb AB), which was swallowed with water. There were no particular restrictions regarding medication before the CR, i.e. patients taking angiotensin‐converting enzyme (ACE) inhibitors or angiotensin II receptor antagonists were allowed to continue with this medication. BP was measured in the supine position prior to the captopril intake after 15 min of rest (in three cases, however, the BP record was lost for technical reasons and the BP recorded at the doctors office was used in the analysis instead). In each patient, the urine flow‐rate during acquisition of the renographic examination was calculated from voiding immediately before and after the performance of the test. The patient was in the supine position over the gamma camera with the kidneys, heart and bladder in the field of view. Starting immediately after the intravenous injection of 70 MBq 99mTc‐MAG3 (Mallinckrodt Ltd.), serial 10‐second/frame images were obtained for 16–20 min. A single‐headed large field of view gamma camera (GE XRT, Starcam 2000) equipped with a general‐purpose parallel‐hole collimator was used and the images were collected in a 128×128 matrix. Subsequent analyses were made using in‐house software for renal scintigraphy. Regions of interest (ROI) were manually drawn around the whole kidney and automatically to get perirenal background ROI. The renographic curves of the right and left kidneys were analysed separately, and the relative function of each kidney (in per cent of total function) as well as the absolute function (camera‐based MAG3‐clearance in ml/min) were calculated using a Patlak plot according to the algorithm proposed by Granerus & Moonen [5]. Interpretation was made with strict adherence to the criteria agreed on at the consensus conference in Santa Fe, 1995 [6]. The patient was considered to have a positive CR if one of the kidneys had 45% or less of total kidney function. In two cases (patients 7 and 11 in Table I), there was a marked difference in the time to maximal uptake (10 min in both cases), which can also be indicative of renal artery stenosis, although this is an unspecific finding. In those two cases, another renographic examination was performed without pre‐treatment with captopril. This second renography did not show a difference in time to maximal MAG3 uptake, which increased the suspicion of renovascular hypertension enough to allow a renal angiography and confirmed presence of a renal artery stenosis that was immediately treated with PTRA.

Table I. Characteristics of patients before and after percutaneous dilatation of unilateral renal arterial stenosis. Kidney function is the individual function (expressed as % of total kidney function) of the kidney with stenotic artery subjected to PTRA.

Patient no.	Age at dilatation	Gender	Weight (kg)	Height (cm)	BMI (kg/m^2)	BP pre‐PTRA (mmHg)	BP 3 months post‐PTRA (mmHg)	Creathinine pre‐PTRA (µmol/l)	Creathinine 3 months post‐PTRA (µmol/l)	Kidney function pre‐PTRA (%)	Kidney function post‐PTRA (%)	Medication pre‐PTRA	Medication post‐PTRA
1	74	m	78	178	25	216/110	178/96	111	94	62	63	B, T, L, CCI, A2	B, T, CCI
2	51	m	90	176	29	146/100	116/76	101	106	29	29	B, L, CCI,	B, T, CCI
3	29	f	53	161	20	166/122	132/84	106	85	20	47	B, L, CCI, Alf	B, CCI,
4	48	m	70	184	21	132/92	120/84	154	112	13	13	L, CCI	CCI
5	72	f	71	168	25	210/108	176/102	108	94	24	27	Ami	Ami, L, CCI
6	49	f	60	160	23	190/104	166/98	112	102	43	43	L, CCI, ACE	L, CCI, ACE
7	77	f	54	157	22	180/70	168/66	123	110	53	53	S, L, CCI	S, L, CCI
8	65	m	95	174	31	190/90	210/90	110	95	29	33	B, L, Alf, ACE	B, L, Alf, ACE
9	66	f	69	163	26	250/116	220/102	90	94	6	25	B, T, Alf	B, L, Alf
10	53	m	79	175	26	150/86	172/108	87	84	32	35	B, CCI, Alf	B, L, CCI, Alf
11	56	m	94	180	29	200/120	186/106	117	80	50	51	B, CCI	B, CCI
12	54	f	60	165	22	146/80	150/96	81	90	36	36	B	B
13	76	f	71	163	27	148/76	146/82	96	87	15	33	L, ACE, Clon	L, CCI, ACE,Clon
14	74	f	63	167	23	176/102	140/80	118	112	20	20	B, L, CCI, Alf, Clon	B, L, CCI, Alf
15	59	f	74	164	28	160/72	160/68	91	91	14	14	S, B, CCI, A2, Alf	B, CCI, A2
16	58	f	51	158	20	240/100	224/94	81	73	18	35	B, CCI, Alf, ACE	B, CCI, Alf, ACE
17	65	k	72	164	27	160/90	126/80	122	92	29	29	B, L, CCI, Ami	B, CCI
18	59	m	75	175	24	150/80	150/76	92	76	29	31		S, CCI
19	70	f	68	162	26	174/70	146/52	117	119	31	26	B	B
20	65	m	87	176	28	172/100	160/96	139	133	26	31	S, B, CCI, Alf, ACE	S, B, CCI, Alf, ACE
21	60	m	70	180	22	130/90	130/80	111	92	30	37	B, CCI	B, CCI
22	38	m	65	165	24	156/112	122/96	124	103	33	29	A2
23	65	f	110	155	46	150/68	156/66	145	143	45	67	S, B, ACE	S, B, L, A2, ACE

S, spironolactone; B, beta receptor blocker; T, thiazide diuretic; L, loop‐diuretic; CCI, calcium‐channel inhibitor; A2, angiotensin II receptor blocker; Alf, alfa receptor inhibitor;ACE, angiotensin‐converting enzyme inhibitor; Clon, clonidine; Ami, amiloride.

PTRA

The patients fasted for at least 4 h prior to the PTRA. Thirty minutes before the angiography, the patient received a rectal application of 5–10 mg of diazepam and a subcutaneous injection of 0.25–0.5 mg of atropine. Renal aortography was performed with a pigtail catheter placed in the abdominal aorta, slightly above the level of the renal arteries. The finding of a >50% reduction of the arterial diameter was followed by an attempt to correct this: A 6 F or 7 F introducer sheath was placed reaching up from the femoral artery to the aorta; 5000 IE of Heparin was injected intra‐arterially through the sheath. The lesion was traversed with standard 4 F catheters and teflon‐coated or hydrophilic 0.032 guide‐wires. Next, the guide‐wire was switched to a Rosen, or similar, stiff wire with short tip. A balloon catheter approximately 20% bigger than true non‐stenotic arterial diameter was chosen for the dilatation. The result of the dilatation was evaluated visually by an immediate angiography. If the result was unsatisfactory, a matching stainless steel balloon expandable stent was placed at the site of the lesion. In case of an osteal lesion, the stent was placed protruding a couple of millimetres into the aortic lumen.

The patient was kept in bed for 10 h following the PTRA and 5000 IE of low molecular weight heparin (Fragmin®) was administered subcutaneously on the same and the 2 following days. The patients' antihypertensive drugs were kept unchanged during the stay in hospital unless very significant hypotension occurred. The great majority of patients (19/23) underwent renographic examination on the day after the PTRA as a safety precaution for assessment of potential complications or acute failure of the endovascular procedure. None of these 19 examinations displayed occurrence of any such problems. All patients underwent a follow‐up clinic examination after 3 months and a second CR after 3 months or more (there could be delay due to availability of the examination). In several cases, the patient underwent CR after 3 months as well as after 12 months. In these cases, the latest examination was that used in Table I and in calculations of outcome. The creatinine following the PTRA was that measured at the clinical follow‐up at 3 months.

Ethics

The study was approved by the Ethics Committee of Linköping University and performed in accordance with the Declaration of Helsinki.

Statistics and correction for blood pressure medication

Statistical calculations were made using StatView 4.5 (Abacus Concepts Inc., Berkeley, CA, USA) software. Comparisons within and between groups were made with Student's paired and unpaired two‐tailed t‐test and linear correlations with Pearson's test or as stated in the text. Mean±standard deviation is given unless otherwise stated. Statistical significance was considered at the 5% level (p≤0.05).

Many of the patients underwent considerable changes in the antihypertensive medication during the follow‐up. To allow a reasonably accurate appreciation of the effects of the PTRA on blood pressure, despite changes in concomitant medication, we assumed that the antihypertensive effect of any antihypertensive drug was −12/8 mmHg [7]. Thus, “corrected systolic BP” was the measured systolic blood pressure level plus the number of antihypertensive drugs×12. The diastolic corrected blood pressure was calculated in the corresponding fashion.

Results

We recruited 29 patients from 1996 to 2003. Five of these had bilateral renovascular disease and underwent bilateral dilatation. Since the aim of this trial was to investigate the potential improvement in the percentage of MAG3 clearance in the stenotic kidney, these patients were excluded from the study. In one patient, there was deterioration of blood pressure control shortly after dilatation of the left renal artery, which turned out to be caused by the occurrence of right‐sided renal arterial stenosis. This patient was also excluded from the investigation. Table I shows characteristics and outcome of the 23 patients that were investigated. Two of the included subjects had diabetes.

As can be seen in Table I and Figure 1 the relative MAG3 clearance of the stenotic kidney expressed as a percentage of total function increased at the subsequent CR (performed on average 8 months later) following PTRA of the stenosis (from 29.9±14% to 35.1±14%, p = 0.01). Furthermore, creatinine levels fell following the intervention (from 110±19 to 99±17 µmol/l, p = 0.0003, see also Table I). Blood pressure levels were also lowered by the intervention (uncorrected blood pressures fell from 173±32/93±17 to 158±31/86±15 mmHg, p = 0.0009 and 0.006 for systolic and diastolic BP, respectively). The mean number of anti‐hypertensive drugs was unchanged following intervention (2.9±1.4 prescribed antihypertensive drugs before and 2.8±1.3 prescribed drugs after the intervention, respectively, p = 0.6, see also Table I), although there were large individual variations (Table I). No major complications were seen following the PTRA procedure in any of the 23 patients. However, one patient, number 6 in the table, had an initial misplacement of the stent, which was subsequently removed and placed in the femoral artery. She recovered uneventfully.

Figure 1 Graphic illustration of the effect of percutaneous dilatation of the renal arterial stenosis on the individual kidney function(expressed as % of total kidney function).

Patients who responded with an improvement of the renographic function of the kidney with the stenotic artery (defined as an increase of 5% or more, n = 7), compared with those that had no or very modest improvement (eight patients had no change, one experienced 4% and another patient 5% deterioration, total number of n = 10), were similar with regard to age, body mass index (BMI), number of antihypertensive drugs, percentage of kidney function at stenotic side, creatinine levels and blood pressure at baseline. On the other hand, patients who had an improvement of systolic blood pressure corrected for the amount of antihypertensive medication (improvement ranging from 9 to 62 mmHg, n = 16) had higher initial corrected systolic and diastolic blood pressures (221±39/124±20 mmHg) than those that did not experience a reduction of corrected systolic blood pressure (183±31/98±15 mmHg, p = 0.04 and 0.01 for systolic and diastolic blood pressures, respectively) following intervention (an increase ranging from 5 to 34 mmHg, n = 6). In a linear multivariate analysis of age, BMI, creatinine, corrected systolic BP, amount of antihypertensive drugs and kidney function of the stenotic side at baseline as independent variables and the improvement of blood pressure or not as dependent variable, there was a positive relation between the blood pressure (p = 0.03) and creatinine (p = 0.01) levels to improvement. BMI, on the other hand, was negatively related to improvement (p = 0.03). Among the six patients that did not improve corrected systolic BP, five patients did improve the relative function of the kidney with the stenotic artery (range 2–22%). Similarly, there were no statistically significant relationships between the change in corrected blood pressures to changes in either improvement of renographic function of the kidney with the stenotic kidney or change in creatinine following the PTRA.

There was no difference with regard to effects of PTRA on blood pressure, change in renographic function of kidney with stenotic artery or creatinine when the four subjects that did not receive a stent was compared with the remaining 19 subjects. In one case (patient 1 in Table I), the kidney that was subjected to PTRA had better renographic function then the non‐treated kidney/artery. This was a consequence of advanced general arteriosclerosis and judgement by the interventional radiologist to primarily dilate the renal artery that potentially could result in the most benefit to the patient.

The result of the CR performed after both 3 and 12 months was available in 13 subjects. The renographic result was very similar in both these investigations, the r‐value for the correlation being 0.99 and r² = 0.98 (p<0.0001).

Discussion

In this prospective study of patients with suspected renovascular hypertension, PTRA was followed by an increase in captopril–renographic MAG3 clearance of the kidney with stenotic artery that was statistically significant. The absolute increase was rather modest, however, being only 5% on average. Similarly, we found statistically significant improvements of blood pressure and decrease of serum creatinine, with moderate absolute reductions of −15/7 mmHg and −11 µmol/l, respectively. On a more individual basis, 16 individuals had an improved corrected systolic blood pressure, and a minority, seven patients, experienced an unchanged or worsened blood pressure control. One single patient, 38 years old, was cured with regard to blood pressure levels, by the intervention, and judged not to need antihypertensive medication any more. In general, the PTRA was quite safe and no major complications were caused by the intervention. Since recruitment was performed on a consecutive basis, the results should apply to patients meeting the inclusion criteria.

We prospectively recruited all patients that were assumed to have renovascular hypertension during 8 years, and the final cohort consisted of 29 subjects. In a unique American epidemiological study, the prevalence of renal arterial stenosis was estimated to be 6.8% according to ultrasonography of renal arteries in elderly subjects [8]. Thus the rather modest number of patients referred for evaluation of renovascular hypertension reflects reluctance to search for this condition among physicians that care for the 1,000,000 subjects that live in the catchment area of the University Hospital of Linkoping. However, it is highly likely that many patients with renovascular hypertension are not easily identified due to treatment with ACE inhibitors or angiotensin II receptor antagonists, which particularly effective mask the symptoms related to the disease [9].

The blood pressure decrease following the intervention, 15/7 mmHg, was of a similar range as can be achieved by addition of a single antihypertensive drug [10]. Bearing this comparison in mind, the risks and in the particular the costs of the PTRA procedure might be hard to motivate. However, it is quite possible that the blood pressure might have increased further if the stenotic arteries had been left untreated, as arteriosclerosis and degree of stenosis increases with time [11]. On the other hand, it is also possible that dilatation of a stenotic renal artery prevents deterioration that might have clinical significance if the patient is followed up for a longer time [12], although the speed of deterioration of kidney function in renal artery stenosis is debated [9], [13].

Since restenosis after PTRA is rather uncommon, we did not find it clinically and ethically justified to perform a follow‐up angiography in the patients [14], [15]. Furthermore, to our knowledge, our study is unique with respect to use of captopril also in adjunction with the follow‐up renography, which significantly increases the chance of detection of renal artery stenosis and renovascular hypertension [6], [16–18]. We cannot completely exclude the possibility of poor correction of the stenosis as the explanation of the fact that seven patients did not achieve a better BP control following the PTRA. However, the majority of these patients did indeed improve renal function assessed by the renographic examination. In the multivariate analysis, BMI was negatively related to the outcome, which suggests that several patients had other reasons for the increased blood pressure than those related to a decrease in renal artery perfusion. Insulin resistance and the metabolic syndrome are highly related to BMI and are causes of high blood pressure that are not expected to be improved by PTRA [19]. The fact that improvement of blood pressure following PTRA was positively related to blood pressure and creatinine levels at baseline in the multivariate analysis in our study is a confirmation of earlier trials [20], [21]. The great majority of our patients received a stent (19/23 = 83%), which might relate to the relatively good results on creatinine levels compared with some other trials [22–24]. Our study only included four subjects below 50 years of age, which limits the statistical power of calculation with age as a variable for outcome. It is reasonable to assume that long‐standing hypertension, being renovascular or of other origin, and high age of the majority of patients in our study negatively effected the outcome.

In conclusion, this prospective trial showed statistically significant improvements of blood pressure control and measures of kidney function in subjects with suspected renovascular hypertension treated with PTRA when analysed on an intention‐to‐treat basis. The PTRA procedure was quite safe but the magnitude of the absolute improvement was rather modest.