Impact of Baseline Renal Function on Outcomes of Renal Artery Stenting in Hypertensive Patients

Abstract

Renal artery stenting may improve blood pressure (BP) and renal function in resistant hypertension patients; however, benefit may differ depending on the degree of renal dysfunction. The authors analyzed 67 consecutive patients receiving stenting for obstructive renal artery disease between 2002 and 2005. Patients were categorized as normal or mildly impaired according to estimated glomerular filtration rate (eGFR) (≥60 mL/min/1.73 m²), moderately impaired (eGFR 30 to 59 mL/min/1.73 m²), and severely impaired (eGFR <30 mL/min/1.73 m²). In patients with eGFR ≥60, systolic BP did not significantly improve from baseline. However, in patients with an eGFR between 30 and 59 mL/min/1.73 m², systolic BP decreased by 12 mm Hg at 6 months (P=.02) and 14 mm Hg at 12 months (P=.01). Greater benefit was observed in patients with eGFR <30 mL/min/1.73 m², with a 16 mm Hg (P=.10) and 21 mm Hg (P=.02) decrease at 6 and 12 months, respectively. Renal function was stable across all groups. Renal artery stenting reduced BP and produced greatest benefit in patients with baseline impaired renal function.

Whether renal artery revascularization is superior to medical management alone for improvement in blood pressure (BP) and stabilization of renal function remains undefined. ¹ , ² , ³ , ⁴ Of the limited data characterizing patients who may derive benefit from renal intervention, determination of timing of intervention as renal dysfunction progresses remains unclear. Early intervention in patients with preserved renal function may avert progressive renal dysfunction and improve BP. ⁵ , ⁶ Conversely, intervention in patients with already developed renal disease may better select patients at risk for clinically significant renovascular obstructive disease. ⁷ Intervention in these patients could stabilize if not improve renal function as well as promote BP benefit. ⁸ Alternatively, if extensive renal dysfunction has already occurred, improvement of renal blood flow with intervention may not yield benefit for BP or renal dysfunction. ⁹ We hypothesized that in patients with resistant hypertension and renal dysfunction, renal artery intervention could stabilize renal dysfunction with BP improvement. We examined changes in BP and renal function in patients following renal artery stenting, comparing patients with and without baseline chronic kidney disease.

Methods

We investigated 67 consecutive patients with difficult‐to‐control hypertension who received stenting for obstructive renal artery disease (>70% stenosis) by interventional cardiologists at Yale‐New Haven Hospital and the West Haven Veterans Affairs Medical Center between 2002 and 2005. Patients were referred for renal artery stenting following either a diagnostic magnetic resonance angiography, renal Doppler ultrasound, or antecedent renal angiography. Postdischarge follow‐up was available for 60 patients (90%).

Baseline characteristics included demographics, BP, serum creatinine, and antihypertensive medications (Table I and Table II). Office BP, number of classes of medications, and doses were catalogued using available follow‐up data from outpatient records (primary care physician, cardiologist, or nephrologist) at 6 and 12 months after the procedure. Intervals were selected to allow adequate time for periprocedural fluctuations of renal function to equilibrate as well as to marginalize the effects of the long‐term natural progression of underlying renal disease.

Table I.

 Baseline Characteristics (N=67)

Age, y	71±9
Caucasian, No. (%)	57 (95)
Male, No. (%)	38 (63)
Comorbid conditions, No. (%)
Hypertension	60 (100)
Diabetes	19 (32)
Dyslipidemia	40 (67)
Coronary artery disease	39 (65)
Cerebrovascular disease	7 (12)
Peripheral artery disease	28 (47)
Congestive heart failure (EF <50%)	11 (18)
Serum creatinine, mg/dL	1.9±1.0
eGFR, mL/min/1.73 m2	44±21
Systolic BP, mm Hg	156±23
Diastolic BP, mm Hg	76±12
No. of medications	3.3±1.1
Class of medications, No. (%)
Diuretics	38 (63)
Calcium channel blockers	33 (55)
β‐Blockers	54 (90)
ACE inhibitors	21 (35)
Angiotensin receptor blockers	10 (17)
α‐Blockers	14 (23)
Aldosterone antagonists	4 (7)
Vasodilators	10 (17)
Central‐acting agents	2 (3)
Bilateral renal artery stenosis, No. (%)	16 (27)
Distal protection device, No. (%)	28 (47)

Abbreviations: ACE, angiotensin‐converting enzyme; BP, blood pressure; EF, ejection fraction; eGFR, estimated glomerular filtration rate.

Table II.

 Baseline Renal Function and BP Stratified by Renal Function

	eGFR ≥60 mL/min/ 1.73 m 2 (n=12)	eGFR 30–59 mL/min/ 1.73 m 2 (n=32)	eGFR <30 mL/min/ 1.73 m 2 (n=16)
Age, y	64±12	72±6a	75±8a
Serum creatinine, mg/dL	1.0±0.2	1.6±0.3a	3.2±1.1a
eGFR	77±14	45±10a	20±6a
Systolic BP, mm Hg	149±20	158±22	156±28
Diastolic BP, mm Hg	74±10	79±13	72±12
No. of medications	3.3±1.5	3.1±1	3.5±1.2

Abbreviations: BP, blood pressure; eGFR, estimated glomerular filtration rate. ^a P<.05.

We also obtained serum creatinine values at baseline and follow‐up and calculated estimated glomerular filtration rate (eGFR) derived from the abbreviated Modification of Diet in Renal Disease (MDRD) equation. ¹⁰ Calculated eGFR incorporates serum creatinine, age, sex, and ethnicity. ¹¹ Chronic kidney disease was defined as normal or mildly impaired (eGFR ≥60 mL/min/1.73 m²), moderately impaired (eGFR between 30 and 59 mL/min/1.73 m²), and severely impaired (eGFR <30 mL/min/1.73 m²). ¹¹ Prior trends in serum creatinine levels were unavailable. Most recent serum creatinine used to calculate eGFR obtained preprocedure was used for baseline. Follow‐up values were obtained from routine outpatient laboratory assessments within 3 months of the 6‐ and 12‐month appointments, respectively. BP measurement techniques were not standardized between the various practitioners; however, individual patient measurements during follow‐up were uniformly obtained. In addition, all practitioners were cognizant of the recommendations for BP measurement by the Seventh Report of the Joint National Committee on the Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7). ¹² With the patient in the seated position, the lowest recorded baseline and office BPs were recorded as well as current medications at time of encounter. Data from patients who required dialysis or died were censored during follow‐up and were considered negative outcomes.

Discrete variables were expressed as counts and analyzed by chi‐square testing. Probability values of P<.05 were considered significant. Statistical evaluation of continuous variables was determined by independent Student t test, and paired Student t test was used when comparing baseline and follow‐up values. The study protocol was reviewed and approved by the Yale University Human Investigational Committee and the Veterans Affairs Investigational Review Board.

Results

Of the 67 patients receiving renal artery interventions, the majority were Caucasian, older than 70 years, and had uncontrolled systolic hypertension, with a majority of patients (77%) prescribed ≥3 antihypertensive medications (Table I and Table II). Follow‐up information was not available for 7 patients, limiting capture of 6‐ and 12‐month outpatient data. Of the patients lost to follow‐up, 1 patient had a baseline serum creatinine of 5.0 mg/dL (eGFR 9 mL/min/1.73 m²) who received stenting for renal salvage prior to planned immediate initiation of hemodialysis. The remaining 6 patients were uniformly resistant hypertensive patients with coronary artery disease, with mean baseline serum creatinine levels of 1.6±0.5 mg/dL (eGFR 44±12 mL/min/1.73 m²) and BP of 162 ± 16/83 ± 12 mm Hg while taking a mean of 3.8±1.2 medications. Excluding the renal salvage patient, baseline characteristics did not differ from those with available follow‐up data. Of the 60 patients with available follow‐up values, 12 had preserved renal function (eGFR ≥60 mL/min/1.73 m²), 32 had moderately impaired renal function (eGFR between 30 and 59 mL/min/1.73 m²), and 16 had severely impaired renal function (eGFR <30 mL/min/1.73 m²). While patients with preserved renal function were younger (64±12 vs 73±7 years; P=.002), there were no significant differences in baseline characteristics, BP values, or medications between the groups.

In all cases, the procedure was successful, with <20% residual stenosis. Distal embolic protection was used in 47% of patients with no difference in outcomes between those who received distal protection and those who did not. ¹³ No patients demonstrated contrast‐induced nephropathy. There were no periprocedural deaths or major complications that required surgical intervention. Femoral artery hematomas occurred in 4 patients, and 2 patients required postprocedural transfusions. Limited renal artery dissection occurred in 2 patients, ameliorated by deployment of additional stents. Limited arterial perforation that resolved after delivery of a covered stent and medical management was observed in 1 patient.

Hemodialysis was required in 3 patients, 2 at month 8 postintervention and 1 at month 10 postintervention. Patients who underwent dialysis had lower baseline eGFR <21 mL/min/1.73 m² (mean serum creatinine 4.8 mg/dL [range 3.1–5.7]); however, 7 other patients with similar baseline eGFR did not require dialysis at 12 months (mean serum creatinine 3.3 mg/dL [range 2.7–4.5]). One patient died at month 7 with exacerbation of underlying pulmonary disease and pneumonia.

Following renal artery stenting, we observed sustained reduction of systolic BP at 12 months without change in number of medications (Table III). Not only was renal function maintained, mild improvement of eGFR from baseline was found overall (Table IV). However, improvement in systolic BP and eGFR was largely restricted to patients with baseline renal dysfunction. Patients with severe renal dysfunction did demonstrate BP reduction after intervention, as well as eGFR improvement at 1 year.

Table III.

 Change in Blood Pressure From Baseline After Renal Artery Stenting

	Δ SBP, mm Hg				Δ DBP, mm Hg				Medications
	6 mo	P Value	12 mo	P Value	6 mo	P Value	12 mo	P Value	6 mo	P Value	12 mo	P Value
All patients (N=60)	−12±28a	0.00	−15±28a	.00	0±15	.86	−4±14	.07	−0.2±1.3	.21	−0.2±1.3	.34
eGFR ≥60 mL/min/1.73 m2 (n=12)	−6±21	.38	−12±32	.24	−2±10	.41	−3±15	.56	+0.1±1.2	.82	−0.2±1.0	.58
eGFR 30–59 mL/min/1.73 m2 (n=32)	−12±27a	.02	−14±26a	.01	−1±5	.82	−3±14	.25	−0.1±1.2	.57	−0.1±1.4	.69
eGFR <30 mL/min/1.73 m2 (n=16)	−16±36	0.10	−21±30a	.02	−2±19	.71	−6±14	.18	−0.6±1.5	.13	−0.3±1.3	.39

Abbreviations: Δ, change; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; SBP, systolic blood pressure.

Table IV.

 Change in Renal Function From Baseline After Renal Artery Stenting

	Δ SCr, mg/dL				Δ eGFR, mL/min/1.73 m 2
	6 months	P Value	12 months	P Value	6 months	P Value	12 months	P Value
All patients (N=60)	−0.1±0.7	.51	−0.1±0.4a	.01	+5±11a	.00	+3±12	.08
eGFR ≥60 mL/min/1.73 m2 (n=12)	−0.1±0.1	.67	−0.1±0.2	.86	+1.2±6.0	.51	−1±15	.91
eGFR 30–59 mL/min/1.73 m2 (n=32)	−0.2±0.3a	.01	−0.1±0.3	.24	+7.3±13a	.00	+4±12	.10
eGFR <30 mL/min/1.73 m2 (n=16)	+0.1±1.2	.69	−0.5±0.6a	.01	+2.4±8	.28	+6±6a	.01

Abbreviations: Δ, change; eGFR, estimated glomerular filtration rate; SCr, serum creatinine.

Discussion

Increased awareness of vascular risk factors and heightened recognition of the clinical signs of renovascular hypertension have led to greater use of noninvasive diagnostic imaging (Doppler ultrasonography, magnetic resonance angiography, computed tomography angiography), as well as selective angiography to characterize renal artery lesions for possible intervention. ¹⁴ Once significant obstructive lesions are angiographically identified, the question remains as to which patients should receive mechanical intervention. There is conflicting evidence regarding patient characteristics that predict response to intervention. Our data suggest greater BP benefit in patients with underlying renal dysfunction and that eGFR <60 mL/min/1.73 m² may identify patients who would benefit from intervention.

The lack of improvement in renal function in patients with preserved renal function likely reflects difficulty in detection using current metrics of serum creatinine or eGFR. However, the reason for lack of significant BP improvement in the preserved eGFR group remains unclear. Using the simple model that renal artery obstruction with poststenotic hypoperfusion results in systemic renin‐angiotensin‐aldosterone system (RAAS) activation, peripheral vasoconstriction, and hypertension, the discrepancy of BP response is not clearly explained. It is possible that renal artery obstruction may play a greater part in BP modulation in the diseased kidney than a normal kidney. That under normal circumstances renal blood flow is in far excess of metabolic demand, ¹⁵ “ischemia” is unlikely the sole mediator of RAAS contribution to elevated BP. Patients with preserved renal function may represent an early phase of ischemic nephropathy in which secondary pathologic changes have not yet developed. In patients who already have manifested sequelae of renovascular hypertension with decreasing renal function, the milieu of the kidney may be altered. Local effects of RAAS have been demonstrated and can be activated by renal injury. In a rat model of angiotensin II (Ang II)–mediated hypertension, renal cortical cell endosomes containing Ang II have been observed to undergo AT₁ receptor–mediated endocytosis and induce transcription and gene expression. ¹⁶ Similarly in a chronic renal obstruction rat model, along with the shear stress associated with hypertension, autocrine and paracrine effects may demonstrate altered sensitivity compared with a normal‐functioning RAAS regulation. ¹⁷ Another experimental study that used a similar model of renovascular obstruction–induced hypertension (2‐kidney‐1‐clip hypertension in a rat) showed attenuation of BP increase by administration of an angiotensin‐converting enzyme inhibitor or angiotensin receptor blocker and identified the role of Ang II inhibition in the early phases of renovascular hypertension and less of an effect in the late phases of renovascular hypertension. ¹⁸ Translating this concept to a clinical model, patients who are treated with agents that block the RAAS with early renovascular lesions may mitigate potential local effects before manifesting clinically significant dysfunction. However, as atherosclerosis progresses both on a microvascular and macrovascular level, some patients may develop chronic, late‐stage effects of renovascular disease that may increase the consequence of macrovascular obstruction. These patients may warrant consideration for revascularization. As such, renal artery revascularization may have multiple manifestations at the local and systemic level in late‐stage disease compared with patients with only early‐stage disease and normal kidney function. The precise pathways by which RAAS mediators manifest in atherosclerotic renovascular disease are not well understood in humans and further investigation is necessary.

In patients with multiple risk factors of age, diabetes, dyslipidemia, smoking, and essential hypertension, atherosclerotic renovascular disease manifest microvascular and parenchymal disease in other vascular territories than the kidneys. Addressing renal artery stenosis in patients with chronic kidney disease has significant theoretic implications for reducing cardiovascular disease risk if renal artery stenosis contributes to cause and is not the sole consequence of vascular disease. Renal vascular insufficiency frequently accompanies progressive kidney dysfunction and resistant hypertension, contributing to an estimated 25% of chronic or progressive renal failure cases. ¹⁹ Furthermore, the risk of major adverse cardiac events is magnified as renal dysfunction progresses to an eGFR <60 mL/min/1.73 m². ²⁰ , ²¹ Although ischemic nephropathy can cause end‐stage renal disease, ²² many individuals with chronic kidney disease may succumb to cardiovascular disease before the onset of end‐stage renal disease. ²³ Mortality benefit of renal artery intervention has yet to be demonstrated.

Investigation of effectiveness of this treatment has been complicated by evolving technology and pharmaceuticals as well as identification of appropriate candidates for treatment. Prospective trials comparing renal artery balloon angioplasty with medical management neither demonstrated BP benefit, nor distinguished the relative contributions of simultaneous changes in medical therapy vs the effects of revascularization. ²⁴ , ²⁵ , ²⁶ However, these trials were complicated by significant crossover and represent outcomes from potentially inferior balloon angioplasty techniques and infrequent or no use of endovascular stenting. ¹ , ² Four‐year follow‐up of a registry of 1058 patients receiving stents demonstrated BP improvement with reduction in number of medications. Patients who had baseline impaired renal function (defined as serum creatinine >2.0 mg/dL) had a relative increased mortality risk compared with those with normal renal function. ⁶ However, this study did not describe specific factors associated with magnitude of BP response to intervention among the groups. Since the only prospective trials examining percutaneous interventions do not reflect currently employed stent and delivery technology, further prospective studies are required to adjudicate conflicting results from registries and observational series.

Overall, renal artery stenting is a safe procedure without evidence of iatrogenic progression of renal dysfunction. Distal protection devices have been suggested to offer additional periprocedural renal protection. No significant overall benefit has been demonstrated, ¹³ with the exception of mild improvement in renal function with additional use of a glycoprotein IIb/IIIa inhibitor with distal protection. ²⁷ Results from randomized trials examining renal artery stenting that require optimal medical management in both arms (antihypertensive agents including inhibitors of the RAAS, aspirin, statins) will characterize more completely the effect of renal artery stenting on BP reduction and address the predictive value of baseline renal function on mortality. ²⁸

Limitations

This investigation is limited as it is a small retrospective study, confined to 2 centers, with 10% of patients lost to follow‐up. In addition, renal function assessments prior to the baseline serum creatinine were unavailable to determine a trend of renal dysfunction prior to intervention. This limits commentary on the effect of intervention on the progression of renal dysfunction. Conclusions from the measurements of renal function that were followed postprocedure suggest that the stenting procedure is not harmful, with slight improvement of eGFR observed over the 1‐year follow‐up period. However, the study was not designed to analyze incremental clinical benefit of stenting on progressive renal dysfunction, and a larger prospective study could better evaluate this question.

Conclusions

This analysis represents a unique investigation of renal artery stenosis interventions using contemporary stenting technology in the context of the updated metric of eGFR. Our data suggest that patients with moderately or even severely impaired baseline renal function may receive BP and renal function stabilization with renal artery stenting compared with those with preserved renal function. Although stenting is generally a safe procedure, appropriate patient selection is required to justify the modest risks.