Relationship between hemodynamic parameters of renal artery stenosis and the changes of kidney function after renal artery stenting in patients with hypertension and preserved renal function

Abstract

Introduction: The role of physiological assessment of renal artery stenosis (RAS) using renal fractional flow reserve (rFFR) and resting translesional pressures ratio (Pd/Pa ratio) in the prediction of benefit from revascularization is still unknown. Objectives: The aim of the study was to assess the relationship between hemodynamic data and the change in kidney function after renal artery stenting in secondary hypertension. Patients and methods: 34 hypertensive patients (50% males, median age 65 years) with at least 60% RAS, underwent stenting and were followed up for 6 months. Pd/Pa ratio (ratio of mean distal to lesion to proximal pressure) and hyperemic rFFR (after papaverine) were measured before the procedure. At baseline and after 6 months, the glomerular filtration rate (eGFR), serum cystatin C and albuminuria were determined. In receiver operating characteristic curves, two previously established cut-off values with the highest accuracy of identifying severe RAS were used: 0.93 for the Pd/Pa ratio and 0.8 for the rFFR. Results: No significant difference in eGFR was found between patients with decreased and normal Pd/Pa ratio (1.4 vs 7.9 ml/min, p = ns). Similarly, minor changes in eGFR were observed in patients with decreased vs normal rFFR (2.4 vs 4.1 ml/min, p = ns). In patients with decreased Pd/Pa ratio, albuminuria remained stable (change 1.4 mg/24 h) compared with an increase of 12.6 mg/24 h in the subgroup with Pd/Pa ≥ 0.93(p < 0.05). However, after exclusion of two outliers with significant baseline proteinuria (425 and 1095 mg/24 h, respectively), the difference in albuminuria change according to the baseline Pd/Pa ratio was no longer maintained. Conclusions: Hemodynamic parameters of RAS do not distinguish the patients who may benefit from renal artery stenting in terms of kidney function improvement in short-term follow-up.

Trial registration: ClinicalTrials.gov identifier: NCT01128933.

Key Words::

• Arterial hypertension
• fractional flow reserve
• Pd/Pa ratio
• renal artery stenosis
• renal artery stenting

Introduction

Atherosclerotic renal artery stenosis (RAS) is one of the causes of both secondary hypertension and impaired kidney function (1–4). In 15–20% of patients, RAS leads to end-stage renal failure and dialysis initiation (5,6). Despite excellent procedural outcomes of renal artery stenting, the clinical benefit of revascularization is still a matter of debate (7). The recently published CORAL study – the largest randomized controlled trial comparing percutaneous angioplasty and optimal medical therapy – did not prove revascularization advantages in terms of a reduction of major cardiovascular and renal events (8). The identification of potential predictors of kidney function after renal stenting would probably improve the appropriate selection of candidates for such an invasive treatment. Recent studies demonstrated that physiological assessment of RAS with intrarenal pressure measurements may be a useful diagnostic tool in diagnosis of severe lesions (9–12). First studies also suggested that resting translesional pressures ratio (Pd/Pa ratio) and renal fractional flow reserve (rFFR) may also predict the blood pressure response after renal revascularization (13,14). The aim of our study was to determine the potential relationship between hemodynamic variables evaluating RAS significance (Pd/Pa ratio and rFFR) and kidney function after renal artery stenting. The renal function was assessed using not only the estimated glomerular filtration rate (eGFR) but also the serum cystatin C concentration and albuminuria excretion that are independent risk factors for cardiovascular events (15).

Materials and methods

Study group

Of 44 consecutive hypertensive patients with at least moderate unilateral RAS in non-invasive studies referred to renal angiography, 34 (50% of males, mean age 65 years) with at least 60% stenosis in angiography underwent renal artery stenting. A clinical examination was performed to determine demographics, cardiovascular risk factors and related comorbidities. Hypertension was diagnosed if the blood pressure was over 140/90 mmHg or the patient was taking at least two antihypertensive drugs. Blood and urine samples were assessed before renal stenting and 6 months after the procedure. Baseline serum creatinine was measured and estimated glomerular fitration rate (eGFR) was calculated using the Modification of Diet in Renal Disease Study Equation (16). Urinary albumin excretion (UAE) was assessed using 24-h urine collection. Serum cystatin C was measured with an enzyme-linked immunosorbent assay (ELISA) method (Quantinkine; R&S Systems, Minneapolis, USA). Patients with severe valvular disease, NYHA III-IV heart failure and eGFR below 30 ml/min, history of contrast nephropathy or refusal to provide informed consent were not included. The study was approved by the ethics committee and signed inform consent was obtained from every patient.

Pressure measurements

Renal distal pressure (Pd) was obtained using the 0.014” Pressure Wire 5 (St Jude Radi Medical Systems, Sweden) placed distally to the lesion; proximal pressure (Pa) was measured from the guiding catheter tip. During pressure measurements, the tip was disengaged from the artery ostium to avoid pressure damping. Resting Pd/Pa ratio was calculated as the ratio of mean distal to mean proximal pressures. Hyperemic rFFR was calculated in the same way after administration of 30 mg of papaverine. Papaverine was given selectively into the renal artery via a 3F multifunctional catheter placed distally to the lesion (if possible) and quickly removed after drug administration. The procedure of renal stenting was successful if angiographic residual stenosis < 20% was obtained.

Follow-up

Before the procedure and after 6 months, renal duplex Doppler and computed tomography (for potential in-stent restenosis assessment) were performed. Angio-CT examinations were performed with a 64-detector CT scanner (Somatom Sensation Cardiac 64; Siemens, Erlangen, Germany). For renal ultrasound, an HD11 (Philips, Eindhoven, the Netherlands) with a multiphase 2–4-MHz convex array transducer was used. No significant restenosis (> 50% of the vessel diameter) was detected after 6 months. Patients were encouraged not to change their medications during the follow-up period. Follow-up visits were performed in all but one patient, who did not agree to have a follow-up visit.

Statistical analysis

The continuous variables are presented as means and standard deviations, as well as medians and interquartile ranges given in the brackets. The paired changes, depicted as medians with interquartile range, were assessed in a Signed-Rank test and compared between subgroups using a non-parametric Wilcoxon test with a two-tailed p-value below 0.05 considered significant. The predictive value of the hemodynamic variables for 70% diameter RAS in quantitative coronary angiography (QCA) (considered severe RAS) was calculated using the areas under the receiver operating characteristics curves (ROC). The best accuracy in identifying severe RAS had a Pd/Pa ratio of 0.93 (with 95% sensitivity and 55% specificity) and an rFFR of 0.80 (with 78% sensitivity and 78% specificity) as presented elsewhere (17). The R-Pearson correlations coefficients were estimated for categorical variables. Statistical analysis was performed using SAS V System ver. 9.2.

Results

Baseline characteristics of the studied group were presented in Table I.

Table I. Characteristics of the studied group.

Age, years	65.0 (53–71)
Males, %	50.0
Diabetes mellitus, %	23.5
Smoking: present/former, %	5.9/50.0
Hypercholesterolemia, %	82.4
Glomerular filtration rate, ml/min	69. 0 (60.9–85.9)
Diameter stenosis, %	74 (67–80)
Drug class taken at the study entry, %
ACEI/ARB	82
Beta-blocker	88
Calcium channel blocker	77
Diuretic	68
Alpha-1 adrenergic blocker	26
Other	9

Data presented as median with interquartile range or percentage. ACEI, angiotensin converting enzyme inhibitor; ARB, angiotensin II receptor blocker.

Baseline hemodynamic measurements and kidney function parameters

Mean Pd/Pa ratio was 0.83 ± 0.12 (median 0.84, interquartile range [IQR] 0.79–0.91) and rFFR was 0.75 ± 0.11 (median 0.77, IQR 0.71–0.82). Mean creatinine level was 99.5 ± 36.7 (median 88.5, IQR 76.0–110.8 μmol/l) with an eGFR of 70.9 ± 22.6 ml/min (median 69.0, IQR 60.9–85.9). Mean cystatin C concentration was 1.13 ± 0.56 mg/l (median 1.0, IQR 0.72–1.24) and strongly correlated with serum creatinine concentration (r = 0.68, p < 0.0001), eGFR (r = − 0.67, p < 0.0001) and albuminuria (r = 0.72, p < 0.0001). Patients with decreased Pd/Pa ratio (< 0.93) had a higher serum creatinine concentration (96.8 vs 71.0 mg/dl, p < 0.05) and were characterized by tendency towards a higher serum cystatin C concentration (1.07 vs 0.72 mg/l, p = 0.052), as presented in Table II.

Table II. The changes in kidney function parameters after renal artery stenting according to baseline hemodynamic assessment of renal artery stenosis.

	Pd/Pa < 0.93	Pd/Pa ≥ 0.93	p	rFFR < 0.8	rFFR ≥ 0.8	p
S-Creatinine before stenting	96.8 (80.4–115.0)	71 (63.5–80.5)	< 0.05	96.8 (82.0–110.8)	80.5 (63.5–106.0)	Ns
S-Creatinine after 6 months	93.8 (73.0–131.0)	76.3 (69.2–85.5)	Ns	94.0 (73.0–114.5)	81.2 (69.2–111.8)	Ns
S-Creatinine change, μmol/l	− 0.17 (− 9.7 to 5.1)	6.6 (5.0–10.8)	Ns	− 0.5 (− 9.7 to 4.5)	5.1 (− 6.3 to 11.5)	Ns
GFR before stenting	66.4 (54.0–80.5)	95.4 (82.5–96.7)	Ns (0.06)	66.4 (60.9–78.6)	81.4 (63.0–96.7)	Ns
GFR after 6 months	76.0 (47.5–80.9)	82.1 (78.3–89.0)	Ns	76.1 (47.5–80.3)	79.6 (54.8–89.6)	Ns
GFR change, ml/min	0.14 (− 2.3 to 8.0)	− 6.4 (− 17.7 to − 5.12)	Ns	0.5 (− 2.1 to 8.0)	− 3.7 (− 18.1 to 5.2)	Ns
Albuminuria before stenting	10.9 (3.7–32.3)	10.2 (3.6–10.4)	Ns	11.1 (3.8–49.8)	10.3 (3.6–17.6)	Ns
Albuminuria after 6 months	13.6 (8.2–78.3)	22.8 (7.9–80.8)	Ns	16.3 (8.2–82.2)	10.7 (7.5–51.8)	Ns
Albuminuria change, mg/24 h	1.4 (− 8.2 to 4.0)	12.6 (4.3–28.8)	< 0.05	1.4 (− 5.8 to 3.0)	4.1 (0.4–12.6)	Ns
S-Cystatin C before stenting	1.07 (0.88–1.32)	0.72 (0.68–0.76)	Ns (0.052)	1.0 (0.84–1.24)	1.0 (0.72–1.28)	Ns
S-Cystatin C after 6 months	1.08 (0.82–1.44)	1.08 (0.84–1.1)	Ns	1.08 (0.84–1.44)	1.06 (0.82–1.21)	Ns
S-Cystatin C change, mg/l	0.02 (− 0.24 to 0.20)	0.12 (0.05–0.36)	Ns	0.06 (− 0.12 to 0.20)	− 0.01 (− 0.28 to 0.26)	Ns

Data presented as medians and interquartile ranges (in brackets). S, serum; GFR, glomerular filtration rate; Ns, not significant.

The angiographic procedural success of stenting was obtained in all patients. Mean Pd/Pa ratio and rFFR after stent implantation were 0.97 ± 0.05 (median 0.98, IQR 0.95–1.0) and 0.97 ± 0.07 (median 0.98, IQR 0.94–1.0), respectively.

Follow-up

No significant difference in eGFR was found between patients with decreased and normal Pd/Pa ratio (1.4 vs − 7.9 ml/min, p = ns). Similarly, minor changes in eGFR were observed in patients with decreased vs normal rFFR (2.4 vs − 4.1 ml/min, p = ns). In patients with abnormal Pd/Pa ratio, albuminuria remained stable (change 1.4 mg/24 h) compared with significant increase of 12.6 mg/24 h in the subgroup with Pd/Pa ≥ 0.93 (p < 0.05; Table II). However, this difference resulted from substantial albuminuria reduction (by − 342 and − 470 mg/24 h) in two subjects with significant baseline proteinuria (425 and 1095 mg/24 h, respectively) in the abnormal Pd/Pa ratio subgroup. After exclusion of these outliers from the analysis, the difference in albuminuria reduction according to the baseline Pd/Pa ratio was no longer maintained.

During the follow-up period, cystatin C concentration remained stable in all studied subgroups of patients (data in Table II).

Discussion

The potential usefulness of Pd/Pa ratio and rFFR as a prognostic tool in renal artery stenting has been previously demonstrated only in few trials (13,14,18,19). Those trials evaluated blood pressure response but not the kidney function parameters after revascularization as a primary endpoint. In our study, we focused on renal parameters including eGFR, albuminuria and cystatin C serum concentration. We demonstrated that patients with decreased Pd/Pa ratio had a tendency (statistically non-significant) to have lower eGFR and increased serum cystatin C concentration, reflecting the presence of basal kidneys abnormalities. However, during the follow-up period, both Pd/Pa ratio and rFFR were not related to kidney function changes after renal artery stenting.

Only minor albuminuria changes were observed in relation to baseline hemodynamic measurements. It should be noted that majority of subjects had microalbuminuria (albuminuria below 300 mg/24 h). Only two subjects had baseline albuminuria more than 300 mg/day and they just achieved significant improvement as measured by the reduction of albuminuria. It suggests that subjects with more severe proteinuria may probably benefit from renal artery revascularization much more than other patients.

The recently published CORAL trial, including 947 patients with at least 60% RAS, demonstrated no difference between renal artery stenting and optimal medical therapy in terms of a combined endpoint – major cardiovascular events (8). Moreover, the prevalence of the renal secondary endpoints, like progressive renal insufficiency or permanent renal-replacement therapy, was also similar in both study arms. It is of note, in the case of 60–80% stenosis, that the additional measurement of systolic pressure gradient was required by the protocol. The stent was implanted only if at least a 20-mmHg systolic gradient was found. This subgroup of patients with hemodynamic stenosis assessment is similar to the study cohort presented in our report. Subjects with hemodynamically proven stenosis did not achieve the reduction of the combined end-point; however, the prevalence of the renal events is not depicted in the main results presentation.

There are a few limitations of our study. Our group is relatively small – the results represent the experience of one center. For comparison, previously published trials with pressure measurements, except the CORAL study, included up to 61 patients. As all consecutive patients referred to our site for renal angioplasty were recruited to the trial, the preprocedural GFR was only mildly declined. Thus, the results should be interpreted with caution. The follow-up period was quite short, so rather modest GFR changes were expected.

In conclusion, our study indicates that hemodynamic parameters of RAS do not distinguish the patients that may benefit from renal artery stenting in terms of kidney function improvement in short-term follow-up.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

ClinicalTrials.gov identifier: NCT01128933