Key words
renal angiography - angioplasty - renal arteries - renovascular hypertension - atherosclerosis
- renal insufficiency
Introduction
Renal artery stenosis (RAS) is believed to be present in 1 % to 5 % of patients with
hypertension [1]. Furthermore, the prevalence in elderly patients has been reported to be as high
as 7 % [2]. The great majority of cases are associated with atherosclerotic disease [3]. Fibromuscular dysplasia accounts for 10 % to 30 % of cases of renovascular hypertension
and is referred to as its most common cause in young adults and children [4]. Other causes for secondary renovascular hypertension include vasculitis and embolic
disease.
RAS is associated with progressive ischemic nephropathy, hypertension, left ventricular
hypertrophy, congestive heart failure, and pulmonary edema, also known as Pickering
syndrome [5]. The diagnosis is made by duplex ultrasound, renal arteriography, magnetic resonance
angiography, or computed tomography angiography [1]. Therapeutic options include surgical vascular repair, medical therapy by means
of blood pressure control and statins, as well as percutaneous angioplasty (PTA).
In the 1990 s, with the development of percutaneous angioplasty, percutaneous renal
artery revascularization became the established standard in the treatment of relevant
stenosis and the number of procedures increased significantly [6]. The largest randomized trials so far, Stenting and Medical Therapy for Atherosclerotic
Renal-Artery Stenosis (CORAL) [7] and Angioplasty and Stenting for Renal Artery Lesions (ASTRAL) [8], failed to show any benefit of revascularization over medical therapy. However,
both trials presented fundamental flaws in their study design: patients enrolled in
ASTRAL were minimally symptomatic or presented subclinical, likely non-obstructive
renal artery lesions. Ca. 50 % of the patients enrolled in CORAL had no clinically
significant renal failure and one third were diabetic [9]
[10]. Saad et al. reported that the revascularization, despite restoring blood flow and
reducing tissue hypoxia, failed to reduce the markers of chronic inflammation responsible
for tissue injury [11]. The results from previous retrospective as well as randomized trials have been
heterogeneous. This has led to a sustained ambiguity and uncertainty in the management
of patients with RAS.
In our university medical center we performed a total of 128 revascularization procedures
during a period from 2000 to 2015.
The aim of this study was to analyze the clinical outcome of patients who underwent
renal stent angioplasty and to look for further insights into patient selection and
the role of PTA in reducing renal injury and hypertension. The presence of concomitant
diseases and their effect on the outcome should be assessed.
Materials and Methods
This retrospective study was HIPAA compliant and was approved by our institutional
review board, which waived the need for informed consent. This retrospective case-control
study aimed to analyze the data of all percutaneous renal artery angioplasty interventions
performed at our university medical center between 2000–2015. The inclusion criteria
were: a) presence of an atherosclerotic stenosis > 50 %; b) angioplasty with placement
of a stent. The exclusion criteria were: a) ineligibility for PTA; b) insufficient
pre-procedural clinical information.
The clinical data were extracted from the internal patient medical records and, when
possible, primary care clinics were contacted in order to gain follow-up information.
Age, gender, concomitant diseases, technical aspects of the procedure, creatinine,
and blood pressure values pre- and post-procedural (time interval: 6 months to 1 year),
complications as well as medical therapy before and after the procedure were extracted
and statistically analyzed. The GFR was calculated for each patient using the CKD-EPI
(Chronic Kidney Disease Epidemiology Collaboration) equation [12].
A total of 128 percutaneous renal artery angiographies were performed in 107 patients
([Fig. 1]). Children (n = 3) and patients with a transplanted kidney were excluded (n = 3).
10 procedures were repeated due to in-stent-stenosis (8.2 %), in two patients PTA
was performed in two different accessory arteries (1.7 %). In eight patients the procedure
was repeated on the contralateral side (6.6 %). All procedures were performed by two
experienced interventional radiologists with more than 15 years of experience in interventional
radiology.
Fig. 1 STARD diagram of the study population with available follow-up data. Children and
patients with transplanted kidneys (NTX) were excluded from the study.
Abb. 1 STARD-Diagramm der Studienpopulation mit verfügbaren Follow-up-Daten. Kinder und
Patienten mit transplantierten Nieren (NTX) wurden aus der Studie ausgeschlossen.
Patients with clinical suspicion of RAS were evaluated by an expert team of nephrologists
and interventional radiologists. The preprocedural workup included renal Doppler ultrasound,
blood tests, and in some cases MRI renal angiography. All patients signed an informed
consent form. All procedures were performed under conscious sedation and local anesthesia.
After puncture of the femoral artery, an Arrowflex 5F-introducer sheath (Teleflex,
Wayne, PA, USA) was placed in the infrarenal abdominal aorta. A 5F pigtail catheter
(Cordis, Fremont, CA, USA) was introduced over a 0.035-inch guide-wire (Terumo, Tokyo,
Japan) under fluoroscopic guidance into the suprarenal abdominal aorta and an abdominal
angiography using iodinated contrast medium (Imeron, Bracco, Milan, Italy) was performed
in all patients to confirm stenosis. Removal of the pigtail catheter and placement
of a 5- or 4-F Cobra 2 or Sidewinder-1 main catheter (Terumo, Tokyo, Japan) into the
ostium of the stenotic renal artery was performed. After intraarterial injection of
2500 units of heparin, the stenosis was passed and the main catheter was conducted
into the post-stenotic part of the main renal artery. After that, a 0.014-inch Spartacore
guide-wire (Abbott, Chicago, IL, USA) was advanced through the main catheter, followed
by angioplasty using a balloon mounted Herkulink-Stent (Abbott, Chicago, IL, USA)
under fluoroscopic guidance. Postprocedural 2500 units of heparin were injected intraarterially.
Renal angiography was repeated at the end of the procedure to evaluate therapeutic
success ([Fig. 2]).
Fig. 2 DSA image showing severe proximal stenosis of the right renal artery before (arrow)
and after angioplasty (star).
Abb. 2 DSA-Bild mit Nachweis schwerer proximaler Stenose der rechten Nierenarterie vor (Pfeil)
und nach Angioplastie (Stern).
The majority of patients were male (65.7 %) and the mean age was 64 years (range:
18–84) ([Table 1]). Femoral artery access was used in most cases (n = 119, 93 %). In 3 instances a
transbrachial approach was used (2.5 %). The right renal artery had the highest number
of procedures (52.3 %). The degree of stenosis was determined visually by the operating
interventionalist, with moderate corresponding to stenosis of 50–70 % (14.9 % of cases),
severe 70–90 % (63.4 %) and extremely severe above 90 % (21.8 %). 9 patients had a
solitary kidney (8.9 %) and in 17 patients kidney atrophy could be seen (16.8 %).
In 7 patients the revascularization procedure was performed in an atrophic kidney
(6.9 %). Concomitant diseases included ischemic heart disease (n = 34, 33.7 %), cerebrovascular
disease (n = 26, 25.7 %), atherosclerosis (n = 27, 26.7 %) and type II diabetes (n = 19,
18.8 %). Chronic hypertension and renal insufficiency were often associated with RAS
(99 % and 42.8 %, respectively). In 5 patients (4.9 %) a dissection of the kidney
artery could be observed immediately after dilation, 3 of which required stent insertion.
There were no major complications such as bleeding, kidney loss, or death. Following
the intervention, 300 mg as an oral bolus and 75 mg of clopidogrel for 1 month, and
a lifelong 100 mg daily dose of acetylsalicylic acid were prescribed to all patients.
Table 1
Baseline characteristics of patients who underwent renal stent angioplasty between
2000 and 2015.
Tab. 1 Ausgangsdaten der renalen Stentangioplastien zwischen 2000 und 2015.
|
male gender (n/%)
|
69 (65.7)
|
|
age range
|
|
|
|
14/13.8
|
|
|
21/20.8
|
|
|
34/33.7
|
|
|
32/31.7
|
|
PTA (n/%)
|
|
|
|
59/48.8
|
|
|
54/44.6
|
|
|
8/6.6
|
|
degree of stenosis (n/%)
|
|
|
|
15/14.8
|
|
|
64/63.4
|
|
|
22/21.8
|
|
solitary kidney (n/%)
|
9/8.9
|
|
renal atrophy (n/%)
|
|
|
|
7/6.9
|
|
|
9/8.9
|
|
|
1/0.99
|
Statistical analysis
Concomitant diseases were divided into 5 categories for statistical purposes: vascular
(arteritis and fibrous dysplasia); cardiovascular (atherosclerosis and coronary artery
disease); lifestyle (obesity, smoking, diabetes mellitus type II); cardiac (heart
insufficiency und left ventricular hypertrophy) and renal (renal failure). Furthermore,
patient age was categorized into four age groups ([Table 1]). Sample characteristics are given as absolute and relative frequencies or mean
+/- standard deviation, whichever is appropriate.
All three outcome parameters (BP, Cr and GFR) were analyzed separately, while the
same modeling approach was performed. If necessary, the parameters were transformed
by calculating the logarithmic values to meet the required model assumptions.
To analyze the course over post-intervention time for the outcome parameter, the change
from baseline was modeled with a baseline-adjusted mixed effect model repeat measurement.
The cluster structure was given by repeated measures within one patient due to the
potential for several interventions and repeated measures within each intervention
at different time points. The variables degree of renal dysfunction, gender, age,
and degree of stenosis were included as predictors. Additionally, the respective interaction
with the time interval of observation was included. Moreover, in all models Re-PTA
and the five concomitant diseases, as described above, were added to control for potential
confounding. A backward elimination for the predictors, their interaction term, and
the confounder was performed using the likelihood ratio test for model comparison.
For the final model we performed a multilevel model with patients as random effects
and measurement variables as fixed effects. We reported regression coefficients, confidence
intervals for regression coefficients, and the corresponding p-values.
All of the models present available case analyses. A two-tailed p < 0.05 was considered
to be statistically significant. Nominal p-values are reported without correction
for multiplicity. All of the analyses were performed using StataCorp Stata 15 (Texas,
USA) and the statistical package R version 3.4.4 (The R Foundation for Statistical
Computing, Vienna, Austria, 2018).
Results
The mid-term follow-up of creatinine levels was possible in 34 patients (33.7 %).
The presence of comorbidities, age, gender, and degree of stenosis failed to show
a significant correlation to renal outcome. No significant improvement in renal function
could be observed in the follow-up (mean Cr drop –0.015 mg/dL; [Fig. 3]). However, higher baseline Cr levels were related to a steeper drop in Cr levels
(p 0.002; difference to baseline –0.25 mg/dL, 95 %CI: –0.36, –0.07). Patients with
advanced kidney failure and ipsilateral parenchymal atrophy showed, in comparison
with milder degrees of renal insufficiency, a significant improvement in renal function
in the follow-up (p 0.021, difference to baseline –0.33 mg/dL, 95 %CI: 0.05, –0.60).
The follow-up GFR values show congruent results to Cr. The GFR showed a slight improvement
in the follow-up (mean 0.019 ml/min). Lower baseline GFRs were related to higher GFR
values in the follow-up (p 0.004; difference to baseline –0.20 ml/min, 95 %CI:-0.31,
–0.04), and the patients with ipsilateral renal atrophy showed a significant therapeutic
benefit (p 0.080; difference to baseline –0.31 ml/min, 95 %CI: –0.59, –0.01).
Fig. 3 Boxplot showing an insignificant drop in creatinine values 6–12 months after stent
angioplasty (mean –0.015 mg/dL).
Abb. 3 Der Boxplot zeigt einen nicht signifikanten Abfall der Kreatininwerte 6–12 Monate
nach der Stentangioplastie (Mittelwert –0,015 mg/dL).
Mid-term BP values were available for analysis in 28 patients (27.7 %). A significant
drop in MAP could be observed (mean –5.27 mmHg; [Fig. 4]). Higher baseline values showed a more pronounced drop in the follow-up BP values
(p < 0.001; diff. to baseline –0.72 mmHg; 95 % CI: –1.4, –0.40). Patients with ages
ranging from 51 to 60 years old showed a significant improvement in MAP (p 0.030;
diff. to baseline –19.2 mmHg; 95 % CI: –34, –4.3; [Fig. 5]). The presence of comorbidities, gender, degree of stenosis, and degree of parenchymal
atrophy failed to show a significant mid-term effect on the BP outcome.
Fig. 4 A relevant drop in the mean arterial pressure (MAP) values 6–12 months after revascularization
(mean 5.27 mmHg) could be observed.
Abb. 4 Ein relevanter Abfall des mittleren arteriellen Drucks (MAP) konnte 6–12 Monate nach
der Revaskularisation (durchschnittlich 5,27 mmHg) nachgewiesen werden.
Fig. 5 Patients with ages ranging from 51–60 years showed a significantly better BP response
to revascularization.
Abb. 5 Die Patienten der Altersgruppe zwischen 51–60 Jahren zeigten eine signifikant bessere
Ansprechbarkeit des Blutdrucks auf eine Revaskularisation.
Sufficient information about oral therapy was available in 95 patients (94 %). The
average number of antihypertensives was 3.2 pre-PTA vs. 3.0 post-PTA (–0.2; p = ns).
In 5 patients antihypertensive medication was discontinued after revascularization
(4.9 %).
Discussion
The aim of this study was to analyze the effectiveness of stent angioplasty in patients
with atherosclerotic renal artery stenosis in terms of renal function and blood pressure
control. Possible interacting factors such as age, gender, comorbidities, degree of
stenosis, and renal parenchyma atrophy were taken into consideration. In our patient
collective we observed a statistically significant improvement in BP control after
therapy, whereas no significant improvement in renal function could be seen. Nevertheless,
we observed that higher baseline Cr and BP values were associated with significantly
steeper drops in Cr and BP, respectively, in the follow-up. No interaction could be
observed with gender, degree of stenosis, or presence of comorbidities.
RAS represents a complex entity with increasingly divergent results in the literature
as to whether the revascularization is or is not associated with an improvement in
renal function and BP control. CORAL, a recent large randomized clinical trial for
RAS, similarly showed a statistically significant improvement in BP as well as a reduction
in medical therapy in patients undergoing revascularization [13]. From the 47 studies reviewed by Mousa et al., 37 (78.7 %) showed an improvement
in BP control after revascularization [14]. Arthurs et al. described a decrease in the rate of renal injury and an improvement
in blood pressure control, although the latter was limited to 6 months [15]. Another retrospective study showed that PTA was successful in reversing resistant
hypertension in patients with atherosclerotic RAS and that the decline in GFR is associated
with chronic kidney damage and is therefore irreversible [16].
To understand the reported divergence in BP results, it may be important to distinguish
renovascular hypertension, which results from renal ischemia, from renal artery disease,
which may or not be responsible for hypertension. While in renovascular hypertension
revascularization is expected to improve blood pressure, in renal artery disease there
may be no causal relationship between the two entities [1]. In a previous study, RAS was identified in only 14 % of patients with clinically
suspected renovascular hypertension [17]. This may explain the relatively worse BP outcome in patients under 50 years old,
who may present with other underlining pathologies. The studies on hypertension made
by Goldblatt et al. in the 1930 s were essential to the understanding of the pathophysiological
mechanism underlying renovascular hypertension and kidney injury. He stated that renal
ischemia is responsible for an increase in the systemic arterial blood pressure, which
is accompanied by a severe disturbance of renal function [18]. The activation of the renin-angiotensin-aldosterone system (RAAS) and the subsequent
recruitment of additional pressor pathways as oxidative stress, sympathoadrenergic
activation, and impaired vasodilatory responses are involved in the process of hypertensive
parenchymal renal injury [19]. In ischemic renal damage, markers of inflammation may fail to decline after the
restoring of tissue perfusion and oxygenation [11]. This may explain the absence of renal function improvement in our patient collective,
as well as in larger clinical trials, such as ASTRAL and CORAL.
Previous studies have assessed the importance of appropriate patient selection [20]
[21]. The Hercules trial failed to show a predictive value of per-procedural BNP in blood
pressure change but stated that the PTRA may play an important role in appropriately
selected patients [20]
[22]. Establishment of reliable predictive factors of chronic kidney disease progression
may be useful to predict long-term benefit of revascularization [21]. New noninvasive methods for the assessment of renal ischemia and irreversible tissue
injury such as ratio of magnetic resonance parenchymal volume to isotopic single kidney
glomerular filtration rate ratio [23] and blood oxygen level-dependent (BOLD) magnetic resonance imaging [11]
[24] may be considered in the future for appropriate patient selection.
We retrospectively analyzed a large collective of patients for both RF and BP control
taking into consideration different possible cofounding factors. Limitations of our
study included the retrospective study design, the heterogeneous patient collective,
the limited availability of follow-up data, and the absence of a control group of
patients treated conservatively.
Renal artery stent angioplasty is a safe procedure [25]. In concordance with previous studies, there were no major adverse events and the
rate of minor events was very low [26]. In contrast, both large prospective trials had a very high rate of major complications,
ASTRAL with 9 % and CORAL with 13 % major stent-related complications. This might
be due to the limited experience in some centers, which is only speculation but is
supported by the low number of patients included in the study by those centers.
In conclusion, the highly divergent study results, and the absence of rigorously selected
patients in the largest randomized trials indicate the need for the development of
new study designs focused on appropriate patient selection. The cooperation between
clinicians and interventional radiologists should be reinforced in order to develop
multidisciplinary standardized protocols taking into consideration reliable predictive
factors and to prevent obsolete interventions in unselected patient groups.
