Atheroembolism during percutaneous renal artery revascularization

Patient population 

This investigation was conducted with approval from the Wake Forest University Health Sciences Institutional Review Board. All RA-PTAS procedures were performed between July 2005 and December 2006 in patients with hemodynamically significant atherosclerotic RVD (≥60% diameter-reducing stenosis by renal duplex ultrasonography) and severe, difficult to control hypertension with or without associated renal insufficiency.

Materials reviewed 

All outpatient clinic, hospital, and endovascular operating suite records were reviewed. Collected data included patient demographics, anthropometrics, medical comorbidities, medications, laboratory results, procedure-specific data, and complications related to RA-PTAS. Percent renal artery stenosis was calculated using a method congruent with that described in the Asymptomatic Carotid Atherosclerosis Study.21 Distal renal artery aspirates were collected at the time of RA-PTAS following angioplasty ± stent placement, but prior to deflation of the distal occlusion balloon, and analyzed at the University of California at San Francisco as described below. All data were maintained in a de-identified electronic database.

Operative management 

All RA-PTAS were performed in an endovascular operating suite at Wake Forest University Baptist Medical Center. Patients were admitted on the day before RA-PTAS to a 23-hour day hospital for intravenous hydration and oral administration of 600 mg N-acetylcysteine every 12 hours. Patients with severe renal insufficiency (estimated glomerular filtration rate [eGFR] ≤30 ml/min/1.73 m2) were also treated with intravenous sodium bicarbonate. Only one renal artery lesion was treated at any given procedural setting. Bilateral lesions were treated in a staged fashion to minimize the volume administered and nephrotoxic effects of iodinated contrast and to avoid the potential for bilateral ischemic renal complications.

Femoral sheath access was employed for all procedures in this report and intravenous heparin administered (50 to 100 U/kg per surgeon preference) once access was secured. Iodixanol (Visipaque, Amersham Health, Princeton, NJ) was used as the intra-arterial contrast medium for all procedures. Selective renal artery cannulation was accomplished using a minimal contact technique of ostial engagement with an angled guiding catheter telescoped through a 10-cm 6 French sheath. Guidewire crossing of the lesion was performed using a commercially available temporary balloon occlusion and aspiration 0.014-inch DEP guidewire system (Guardwire, Medtronic, Minneapolis, Minn). The distal occlusion balloon was inflated in the distal main renal artery immediately after crossing the lesion and occlusion confirmed by hand injection of contrast. RA-PTAS was then performed using angioplasty balloons with or without balloon mounted stents (Genesis, Cordis, Miami FL; or Racer, Medtronic, Minneapolis Minn) of operator choice sized to match the distal, normal appearing renal artery. Predilation angioplasty was performed at the discretion of the operating surgeon with the balloon occlusion device in place and inflated. Stents were positioned to extend 1 to 2 mm into the aortic lumen while completely covering the lesion. Following angioplasty ± stent deployment, the static column of blood proximal to the temporary occlusion balloon was evacuated using a rapid exchange aspiration catheter (Export catheter, Medtronic, Minneapolis Minn). Sixty ml of blood was aspirated and the aspiration catheter was then used to inject at least 20 ml of heparinized 0.9% saline to flush any residual debris from the renal artery. The temporary occlusion balloon was then deflated and completion renal arteriography was performed. Routine follow-up included clinic visits at 4 weeks, 6 months, 1 year and, then, yearly with repeat measurement of blood pressure and serum creatinine.

Specimen collection and analysis 

Specimen analysis and processing was conducted at the University of California San Francisco. The initial 5 ml of aspirated blood was transferred to tubes containing 0.109 molar buffered sodium citrate (BD Vacutainer, Becton Dickinson and Company, Franklin Lakes, NJ) and stored at 4°C until particle analysis was performed.

For particle analysis, the samples were diluted in an equal volume of 0.9% saline and passed through a 20 μm filter. Fragments retained on the 20 μm filter were then resuspended in sterile saline and passed through a 60 μm filter. Particles 20 to 60 μm in diameter (contained in the effluent after passage through the 60 μm filter) were sized and counted using a Coulter Counter (Beckman Multisizer 3, Beckman Coulter Inc, Fullerton Calif). Particles >60 μm in diameter (retained on the 60 μm filter) were removed, resuspended, counted, and categorized in 100 μm increments under a light microscope at 100x magnification against a 100 μm background grid. In order to estimate the relative contribution of contaminant debris to specimen particle counts, “sham” particle counts were performed on sterile saline passed through filters and resuspended in a manner identical to the handling of specimens. Due to the size-specific methods of particle number and size determination, total counts were grouped into size categories of 20 to 60 μm and >60 μm, reflecting the respective thresholds for use of Coulter Counter vs light microscopy methods for counting and sizing.

Renal function was evaluated using the abbreviated modification of diet in renal disease formula for eGFR calculated from pre- and postoperative serum creatinines:22

eGFR/1.73 m2 = 186 × (Serum Creatinine)−1.154 × (Age)−0.203 × (0.742 if female) × (1.210 if African American).

Post revascularization renal function was assessed at 4- to 8-week follow-up to allow for the effects of perioperative volume expansion and contrast administration to resolve. This follow-up time point was chosen due to its coincidence with routinely scheduled follow-up employed by our group. Renal function response was categorized on a per-procedure basis as improved or worsened if a ≥20% increase or decrease in eGFR was observed, respectively. Those not meeting these criteria were defined as unchanged.

Blood pressure response was assessed using the highest measured brachial blood pressure pre- and postoperatively taken in a sitting position. Blood pressure response was defined according to previously published guidelines23 as follows: cured – diastolic blood pressure (DBP) less than 90 mm Hg and systolic blood pressure (SBP) less than 140 mm Hg off all antihypertensive medications; improved – DBP <90 mm Hg and/or SBP <140 mm Hg on the same or reduced number of medications or a reduction in DBP of at least 15 mm Hg on the same or a reduced number of medications; failed – no change or inability to meet the criteria for cure or improvement.

Statistical methods 

Descriptive statistics are reported as mean ± standard deviation for continuous factors, and counts and percents for categorical factors. Associations between clinical factors and captured debris were examined using t tests for dichotomous factors and univariate regression analyses for continuous factors. Log transformations of total embolic debris counts 20 to 60 μm and >60 μm were analyzed to stabilize variances of those outcomes. Renal function, blood pressure, and antihypertensive agent outcomes following RA-PTAS were examined using paired t tests. Statistical significance was assessed at the 0.05 alpha-level. Log transformations of captured debris counts (20 to 60 μm and >60 μm), and clinical or operative predictors of blood pressure response and eGFR response to RA-PTAS were examined using multiple linear regression models. For each outcome, a “best” model was selected using a forward stepwise selection procedure where the most significant candidate variables with P-value < .15 were entered one-by-one and retained if they remained significant at the .10 level after entry. Residuals for each model were examined after variable selection to evaluate fit and look for influential observations. All statistical analyses were performed using SAS software, version 9 (SAS Institute Inc, Cary, NC).

Study group 

Summary demographic and preoperative clinical data are displayed in Table I. Twenty-eight RA-PTAS procedures were performed with DEP in 17 female and 10 male patients with a mean age of 72.4 ± 8.6 years. Mean preoperative eGFR and creatinine were 48.9 ± 17.9 mL/min/1.73mL2 and 1.49 ± 0.55 mg/dL, respectively. No patient was dialysis-dependent. All patients were hypertensive (mean systolic and diastolic blood pressures were 164.1 ± 21.3 and 81.3 ± 9.0 mm Hg, respectively) with a mean antihypertensive agent usage of 3.4 ± 1.3.

Table I. Descriptive statistics for patients undergoing RA-PTAS (N = 27)

		n (%)	Mean ± SD	Minimum	Maximum
	Demographics
	Age (years)		72.4±8.6	58	88
	Caucasian race	21(82)
	Female gender	17(63)
	Preoperative medical data⁎
	SBP (mm Hg)		164±21	124	221
	DBP (mm Hg)		81±9	64	96
	Antihypertensive agents		3.4±1.3	1	6
	Serum creatinine (mg/dL)		1.5±0.6	0.7	3.0
	eGFR (mL min/1.73m2)		48.9±17.9	15.7	88.6
	Preoperative renal insufficiency	20(74)
	Weight (kg)		77.2±17.2	45.9	114.0
	Coronary artery disease	8(30)
	Diabetes mellitus	7(26)
	COPD	2(8)
	Tobacco use
	Former	13(48)
	Current	4(15)
	Hyperlipidemia	17(63)
	Stroke/TIA	11(42)
	Complicated hypertension	3(11)
	Statin use	9(33)
	Fibrate use	3(11)
	Aspirin use	17(63)
	Clopidogrel use	7(26)
	Aspirin and clopidogrel use	5(18)

RA-PTAS, renal artery angioplasty and stenting; SBP, systolic blood pressure; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; COPD, chronic obstructive pulmonary disease; TIA, transient ischemic attack.:

⁎ eGFR <60 mg/mL/1.73m2.

Procedural data 

The 28 RA-PTAS procedures performed during the study period included 26 stent procedures and two primary angioplasty procedures (both for restenosis). Procedural data are summarized in Table II. Treated were 25 primary atherosclerotic lesions and four re-stenotic lesions (three from prior RA-PTAS and one from prior renal artery endarterectomy). Twenty-six patients were treated for unilateral disease and one for bilateral disease in a staged fashion. All treated patients had two kidneys. Mean preoperative renal artery stenosis was 70 ± 12%. Mean balloon occlusion time was 15.7 ± 4.3 minutes, and balloon occlusion for DEP was angiographically complete in 26/28 procedures (93%). Lack of complete balloon occlusion was due to an early renal segmental arterial branch that remained perfused in one patient and a poststenotic dilatation in the second patient resulting in incomplete occlusion with full device inflation. DEP was attempted and abandoned in a single patient with poststenotic renal artery ectasia which did not allow meaningful balloon occlusion due to size mismatch; this patient therefore did not undergo specimen collection and was not included in analysis. No complications related to RA-PTAS with DEP were observed. Clinical follow-up occurred at a mean of 48 days postintervention.

Table II. Descriptive statistics for renal angioplasty or stenting procedures performed with distal embolic protection

		n (%)	Mean ± SD	Minimum	Maximum
	Preoperative stenosis (%)		70±12	60	95
	Renal artery diameter (mm)		5.1±1.1	3.6	8.2
	Stent diameter (mm) (N = 25)		5.3±0.6	4.0	6.0
	Renal stent/artery diameter ratio (%) (N = 25)		95±17	72	137
	Stent placed	26(93)
	Stent type (N = 26)
	Genesis	24(93)
	Racer	2(8)
	Predilation prior to stent placement (N = 25)	5(20)
	Complete balloon occlusion achieved	26(93)
	Balloon occlusion time (minutes) (N = 18)		15.7±4.3	8.0	22.0
	Contrast volume (mL)		60.3±31.3	20	130

N = 28 procedures unless otherwise noted.

Renal function response at 4 to 8 weeks 

Renal function responses are summarized in Table III. Both pre- and postoperative laboratory data for determination of renal functional response at 4 to 8 weeks were available for 22/27 patients. Mean change in creatinine and eGFR were −0.14 ± 0.24 mg/dL and 5.8 ± 8.9 mL/min/1.73mL2, respectively; both were statistically significant. Renal function response was defined as improved in eight patients (36%), unchanged in 14 patients (64%), and worsened in none. Pre- vs postoperative eGFRs in the two patients with incomplete renal artery balloon occlusion were 39.2 vs 31.3 and 27.2 vs 42.1 mL/min/1.73mL2, respectively.

Table III. Renal function and blood pressure responses following percutaneous renal revascularization with distal embolic protection

		Preoperative	Postoperative	Difference	P value
	Renal function
	Serum creatinine (mg/dL)	1.6±0.6	1.4±0.6	−0.1±0.2	.013
	eGFR (mL/min/1.73m2)	47.1±16.7	52.9±17.9	5.8±8.8	.006
	Blood pressure (mm Hg)
	Systolic	163±21	141±23	−22±30	.001
	Diastolic	81±9	72±11	−8±11	<.001
	Antihypertensive agents⁎	3.2±1.2	2.9±1.2	−0.3±0.4	.011

eGFR, Estimated glomerular filtration rate. Descriptive statistics are reported as mean ± standard deviation. N = 22 for all parameters except:

⁎ N = 24.

Blood pressure response at 4 to 8 weeks 

Pre- and postoperative blood pressure and antihypertensive medication use are summarized in Table III. Data on blood pressure response were available for 25/27 patients. Statistically significant decreases in SBP, DBP, and number of antihypertensive agents were observed. At 4- to 8-week follow-up, blood pressure response was defined as cured in zero patients, improved in 12 (48%), and failed in 13 (52%).

Embolic particle analysis 

Particle size and count data are summarized in the figure (Fig). Mean total number of embolic particles counted per procedure was 2033 ± 1553 for particles 20 to 60 μm and 265 ± 132 for particles >60 μm. Mean “sham” total particle count was 26.5 ± 10.4 (range 11 to 46). For captured particle counts 20 to 60 μm, significant multivariate associations were observed with African American race (P = .002), predilation (P = .005), and stent diameter (P < .0001). Preoperative aspirin use was associated with significantly lower captured particle counts (P = .016). For captured particles >60 μm, a significant multivariate association was observed between the ratio of renal artery stent diameter to renal artery diameter (P = .009) and increasing particle counts.

View Large Image Download to PowerPoint

Fig. Particle counts by size from renal artery aspirates following RA-PTAS with distal embolic protection. Lower and upper sides of the “boxes” are drawn at the 25th and 75th percentiles, respectively, with a horizontal line at the medians and “whiskers” extending the 10th and 90th percentiles.

Mean particle counts for patients treated for restenotic lesions (N = 4) were: 4121 ± 2682 (<60 microns) and 298 ± 132 (>60 microns). Additional multiple regression analysis excluding patients treated for restenotic lesions did not significantly affect predictors of embolic counts or renal functional outcomes.

Associations between captured particle counts and renal function/blood pressure responses: In multivariate models, a significant association was observed between increasing captured particle counts >60 μm and decreasing postoperative eGFR (P = .015). Other significant predictors of postoperative eGFR included preoperative eGFR (P < .001) and aspirin use (P = .006). No significant associations between blood pressure response (systolic or diastolic) and captured embolic particle counts were observed.