Renal Artery Plaque Composition is Associated with Changes in Renal Frame Count Following Renal Artery Stenting

ABSTRACT: Background. Benefit of percutaneous revascularization for atherosclerotic renal artery stenosis (RAS) may be attenuated by distal embolization of atheroemboli. The purpose of this study was to characterize RAS plaque composition with intravascular ultrasound virtual histology (IVUS-VH) and to explore the relationship between plaque components and renal frame count (RFC) after renal revascularization. Methods. Seventeen patients (75 ± 7.5 years; 18 lesions) undergoing RAS revascularization were included. Before stenting, automated IVUS-VH pullback (0.5 mm/sec) with analysis of the minimal luminal diameter (MLD) frame and entire atherosclerotic segment was performed. RFC was also determined before and after stenting. Results. The VH component analysis of the segment demonstrated predominantly fibrous tissue (56.3 ± 11.4%), followed by necrotic core (21.8 ± 8.6%), dense calcification (13.2 ± 6.6%) and fibrofatty tissue (8.7 ± 4.0%). Analysis of the MLD frame also demonstrated mostly fibrous tissue (62.1 ± 11.1%), with smaller amounts of necrotic core (15.6 ± 7.3%), fibrofatty (13.9 ± 9.6%), and dense calcification (8.4 ± 6.0%). A trend toward more fibrous tissue (p = 0.074), less necrotic core (p = 0.095) and less dense calcification (p = 0.075) at the MLD compared to the segment was observed. Analysis of the entire atherosclerotic segment revealed a positive correlation between % necrotic core and change in RFC (r = 0.582; p = 0.029), with increasing necrotic core associated with an increase in RFC after revascularization. Conclusion. Both the MLD frame and segmental analysis of atherosclerotic RAS lesions demonstrate predominantly fibrous tissue with smaller amounts of necrotic core, fibrofatty tissue, and dense calcification. Increased necrotic core correlates with a lack of improvement in RFC after stenting.  

J INVASIVE CARDIOL 2011;23:227–231

Key words: renal stent, hypertension, virtual histology, intravascular ultrasound

__________________________________

Percutaneous revascularization of atherosclerotic renal artery stenosis (RAS) improves management of hypertension and helps preserve renal function.^1–3 However, the magnitude of this benefit in clinical practice is variable, in part due to the lack of predictive markers for favorable patient and lesion selection.⁴ The potential for liberation of lipid-rich atheroembolic debris from plaque disruption during balloon angioplasty and stenting has been postulated as an important variable associated with post-procedural glomerular filtration rate (GFR) decline and potentially attenuating long-term clinical benefit.⁵ In this regard, the ability to precisely characterize RAS plaque composition may help in identifying lesions more likely to result in embolization during stenting and therefore benefit from preventive pharmacologic and device strategies. The primary purpose of this study was to characterize RAS plaque by intravascular ultrasound virtual histology (IVUS-VH) and to explore the relationship between plaque composition and change in renal frame count (RFC) after renal artery stenting.

Methods

Institutional review board approval was obtained from the University of California, San Diego for this study. Seventeen consecutive patients (with a total of 18 RAS lesions) undergoing clinically indicated renal artery stenting and IVUS evaluation at the time of their revascularization procedure were included. Patients underwent renal artery revascularization for uncontrolled hypertension (blood pressure > 140/90 mmHg despite treatment with ≥ 2 antihypertensive medications) in the presence of a ≥ 70% renal artery stenosis by visual estimate. Data were obtained from a prospectively collected database and confirmed by a chart review. Clinic blood pressure data were obtained from clinic visits prior to renal artery revascularization. Intra-procedural blood pressures were determined from the intra-abdominal aortic pressure prior to catheter engagement of the renal ostium. The presence of coronary artery disease was defined as a history of myocardial infarction, prior coronary revascularization, positive stress test, typical angina, or ≥ 50% stenosis on coronary angiography. The presence of non-renal artery peripheral arterial disease was defined as a history of claudication, abnormal ankle-brachial index, prior peripheral revascularization, or known carotid or lower extremity stenoses of ≥ 50%. GFR was estimated using the Cockcroft-Gault equation.⁶

Procedural details. All procedures were performed by board-certified interventional cardiologists using standard techniques at the discretion of the operator. The renal artery was engaged with a modified “no touch” technique and a 6 French (Fr) guide catheter was used in all procedures.⁷ Anticoagulation and antiplatelet therapy included aspirin, clopidogrel, and bivalirudin in all patients. IVUS and intervention were both performed using a 0.014˝ wire for support. Seventeen of the lesions were predilated after IVUS interrogation with Viatrac balloons (Abbott Vascular, Abbott Park, Illinois) or a Maverick balloon (Boston Scientific, Natick, Massachusetts). One severely stenotic lesion was pre-dilated prior to IVUS passage. Seventeen lesions were subsequently stented with either Herculink (Abbott Vascular) or Palmaz Genesis stents (Cordis Corporation, Bridgewater, New Jersey). One lesion was treated with balloon angioplasty alone.

Angiography. Cineangiography at 30 frames/sec was performed at baseline and after intervention and stored digitally. All measurements were made offline by an observer blinded to the clinical and IVUS data. RFC was defined as the number of cineangiographic frames for contrast to reach the distal renal parenchyma after initial renal artery opacification and measured as previously described by our laboratory.⁸ RFC was determined at baseline and after intervention in 17/18 lesions (with one lesion excluded due to poor image quality). Lesion stenosis severity was determined by quantitative angiography using Heartlab imaging software (version 2.13.08.00.00, AGFA, Mortsel, Belgium).

Intravascular ultrasound. Automated IVUS pullback was performed with a 3.5 Fr, 20 MHz Eagle Eye Gold catheter (Volcano Corporation, Rancho Cordova, California) at 0.5 mm/sec using a dedicated sled device for 16 of 18 lesions. In two lesions, manual pullback was performed. For these two lesions, only MLD frame data are reported. The pullback was initiated at least 5 mm distal to the target lesion. All pullbacks were performed back through the ostium of the vessel with the guiding catheter disengaged. The entire pullback was defined as the lesion segment. VH data were obtained during the pullback, stored on DVD and analyzed offline. Vessel and lumen dimensions and plaque burden were determined for the segmental analysis and the MLD. IVUS-VH measurements were by an experienced observer blinded to the clinical and angiographic data.

Vessel and lumen dimensions, plaque burden, and VH composition were determined for the entire pullback segment and MLD frame using previously published methods.9 VH plaque composition was analyzed in an automated fashion via Volcano S5 software version 2.2.3.2236. The plaque components were coded by the analysis software as follows: dark green (fibrous tissue), light green (fibrofatty tissue), red (necrotic core), and white (dense calcification) (Figure 1).

Statistical analysis. Data are presented as means ± standard deviations, except in figures where the standard error of mean is used. Paired t-test was used for pre- and post-procedure comparisons of individual patient data. Unpaired t-test was used to compare data between the MLD and segmental pullback analysis. Linear regression analysis was used to examine the relationship between plaque composition and change in RFC before and after stenting. A p-value of < 0.05 was considered statistically significant.

Results

Baseline subject characteristics. The study cohort consisted of 17 subjects (75 ± 7.5 years; 9 male) (18 lesions) with multiple risk factors for atherosclerosis (Table 1). Baseline blood pressure obtained in the clinic was recorded as 141 ± 22/73 ± 9.1 mmHg, while baseline intra-procedural blood pressure was 161 ± 15/71 ± 14 mmHg, with the use of 3.0 ± 1.1 antihypertensive medications. Ten of the 17 subjects (59%) were being treated with an HMG-CoA reductase inhibitor at the time of the renal revascularization.

Angiographic characteristics. All patients underwent successful revascularization of the renal artery lesions without procedural complications and all 18 lesions were classified as ostial.10 After pre-dilation (pre-dilation balloon diameter: 4.9 ± 0.8 mm), 17/18 lesions were stented (stent diameter and length: 5.7 ± 0.7 mm and 16.2 ± 2.1 mm, respectively). Reference vessel diameter was 5.7 ± 1.1 mm. Renal artery stenosis was reduced from 66.0 ± 13% to 8.6± 17%, respectively (p < 0.001). RFC was reduced from 34.5 ± 10.2 frames to 30.5 ± 11.1 frames (p = 0.191) after stenting, resulting in a mean change in RFC of -2.5 ± 13 frames. A reduction in RFC (-11.9 ± 7.6 frames) was observed in 10 of the 17 renal arteries, while an increase in RFC (7.1 ± 8.0 frames) was observed in the other 7.

IVUS dimensions and VH pullback analysis. An average of 37 ± 15 frames were analyzed for each lesion with a mean VH segment analysis length of 16.2 ± 9 mm. Average plaque volume was 318 ± 178 mm³. The VH component analysis of the segment demonstrated predominantly fibrous tissue (56.3 ± 11.4%), followed by necrotic core (21.8 ± 8.6%), dense calcification (13.2 ± 6.6%) and fibrofatty tissue (8.7 ± 4.0%) (Figure 2).

Analysis of the MLD frames demonstrated a plaque burden of 73.2 ± 16.5%, a minimal lumen diameter of 2.5 ± 0.8 mm, minimal lumen area of 9.1 ± 6.1 mm², and reference vessel diameter of 5.8 ± 1.2 mm. The VH analysis of the MLD frame demonstrated mostly fibrous tissue (62.1 ± 11.1%) with smaller amounts of necrotic core (15.6 ± 7.3%), fibrofatty (13.9 ± 9.6%), and dense calcification (8.4 ± 6.0%). The comparison between the VH composition of the MLD and the segment analysis revealed a trend toward more fibrous tissue (p = 0.074), less necrotic core (p = 0.095) and less dense calcification (p = 0.075) at the MLD (Figure 2). Furthermore, the MLD frame had significantly more fibrofatty tissue versus the lesion segment (p = 0.041).

Relationship between plaque composition and change in renal frame count. Analysis of the entire lesion segment in the 15 RAS lesions with compete data revealed a positive correlation between % necrotic core and change in RFC (r = 0.582; p = 0.029), with increasing necrotic core associated with an increase in RFC after revascularization (Figure 3). There was a negative correlation with % fibrous plaque and change in RFC, which did not reach statistical significance (r = -0.449; p = 0.093; Figure 3). There was no correlation between the change in RFC and the % lesion segment composition of fibrofatty (r = 0.01; p = 0.969) or dense calcification (r = 0.233; p = 0.404). No correlations were observed between the change in RFC and individual plaque composition at the MLD (r = 0.218 and p = 0.401 for fibrous tissue; r = 0.052 and p = 0.842 for necrotic core; r = 0.02 and p = 0.938 for fibrofatty; and r = 0.325 and p = 0.203 for dense calcification) or plaque burden at the MLD (r = 0.083; p = 0.749). There was no relationship between IVUS or angiographic vessel or lumen dimensions and change in RFC (data not shown).

Blood pressure and renal function at follow-up. Follow-up clinic blood pressure measurements and serum creatinine levels were assessed at a median follow-up duration of 10.6 weeks in all 17 patients. At follow-up, there was a significant decrease in systolic blood pressure (141 ± 22 mmHg versus 124 ± 8.3 mmHg at baseline; p = 0.002) and mean blood pressure (96 ± 11.5 mmHg versus 88 ± 5.9 mmHg at baseline; p = 0.009). There was no change in diastolic blood pressures at follow-up compared to baseline (73 ± 9.1 mmHg versus 70 ± 6.1 mmHg; p = 0.152). The mean number of antihypertensive medications was the same at follow-up as compared with baseline (3.0 ± 1.1 versus 3.1 ± 1.0; p = 1.00). Serum creatinine (1.2 ± 0.5 mg/dl versus 1.1 ± 0.4 mg/dl at baseline; p = 0.70) and estimated GFR (60.6 ± 21.1 ml/min versus 57.9 ± 17.9 ml/min at baseline; p = 0.429) were also unchanged at follow-up. There were no significant correlations between plaque composition and change in GFR (data not shown).

Discussion

There are two novel findings of this study. First, the IVUS-VH evaluation of RAS plaque across the entire lesion segment demonstrates a predominantly fibrous composition. Second, and more clinically relevant, an association was present between increasing % necrotic core and lack of improvement in renal frame count following percutaneous RAS revascularization. These data suggest that plaque composition may effect renal perfusion (as angiographically estimated with the measurement of RFC), possibly due to distal embolization of material in the necrotic core.

Characterization of renal artery plaque by IVUS-VH. In contrast to the coronary circulation, there have been few studies examining the use of IVUS in the renal arteries. The use of IVUS is safe and helpful in providing vessel measurements for stent size selection and delineation of the renal ostium.¹¹ While use of traditional gray-scale IVUS in the renal arteries may identify the presence of atherosclerosis, this technique does not reliably identify specific plaque components.^12,13 In this regard, the use of IVUS backscatter signal analysis for virtual plaque histology determination may provide valuable additional clinical information. This technology has been validated in the coronary circulation, where it demonstrates predictive accuracy for the four plaque types (fibrous, fibrofatty, necrotic core, and dense calcification).¹⁴ However, there has been limited application of this technology in the peripheral circulation, with isolated reports detailing its use in carotid, renal atherosclerotic disease, and renal fibromuscular dysplasia.^15–19

A recent study by Kataoka et al examined the IVUS-VH characteristics of atherosclerotic renal artery lesions and demonstrated a predominance of fibrous plaque at the MLD, a finding corroborated by the present study.¹⁵ In contrast to the methods employed by Kataoka et al, the present study used an automated pullback, allowing for segmental volumetric analysis of plaque composition. Volumetric analysis demonstrated complex atheroma present throughout the entire segment of pullback composed mostly of fibrous and necrotic core elements. These findings are similar to the reported histology of autopsy-derived renal atheroma samples. McCormack et al examined renal arterial segments taken from 94 patients with diastolic hypertension and renal artery atherosclerosis.20 The authors examined the histological properties of the atheroma lesions with multiple staining techniques and found that in the majority of patients with renal atherosclerosis there was eccentric fibrous plaque accumulation, predominantly composed of collagen. In one-third of the patients, circumferential atherosclerosis with areas of lipid and focal calcium were also noted.

Distal embolization. The specific composition of renal artery plaque may have important clinical implications. The release of embolic material, particularly after the mechanical trauma of stent deployment, could theoretically contribute to adverse outcomes, including deterioration of renal function.²¹ Distal embolization of embolic particles during renal artery stenting, particularly during post-dilation, has been observed by both microembolic signal detection and direct capture of debris from distal protection filters.²² The vast majority of released particles are < 60 µm in size, although larger particles up to 500 µm in size have been reported from ex vivo studies.^5,21 Kawarada et al demonstrated that even wire passage through renal atherosclerotic lesions may result in distal embolization.²² Therefore, assessment of the atherosclerotic burden and atheroma composition throughout the renal vasculature and not only at the MLD site may be clinically relevant. In this regard, segmental volumetric assessment of plaque composition by VH could be a viable method to evaluate the culprit lesion and adjacent plaque. In the present study, evidence of differential plaque composition characteristics of the focal MLD frame and the adjacent lesion segment — a finding which could be the result of plaque remodeling similar to that described in the coronary circulation^23,24 — is shown. Additionally, this study demonstrates an association between increasing necrotic core and decreasing fibrous plaque in the lesion segment, with an increase or lack of improvement in RFC after stenting. These findings suggest an interaction between plaque composition and change in renal perfusion, as assessed by the angiographic parameter of RFC. Physiologically, we hypothesize that atheroma architecture (specifically, lipid-rich lesions with less surrounding fibrous tissue) may represent lesions more prone to liberation of debris, while well-developed fibrotic lesions are less likely to lead to distal embolization. In addition, it is possible that after renal stenting, release of vasoreactive substances could contribute to the lack of improvement in RFC.

The study of distal embolization in renal revascularization has been hampered, in part, by a lack of surrogate markers of renal perfusion. RFC has been evaluated as an objective assessment of macro- and microvascular perfusion. Mulmudi et al evaluated RFC in patients with angiographically normal renal arteries and in patients with FMD.²⁵ Patients with FMD had markedly prolonged RFC measurements in comparison with normal controls. Our laboratory has previously evaluated RFC before and after renal stenting in patients with atherosclerotic renal artery stenosis.⁸ In patients with hypertension and RAS, renal stenting resulted in a reduction in RFC, and a decline in RFC after stenting was associated with improvement in systolic blood pressure at 6-month follow-up. While lack of decline in RFC after percutaneous renal artery revascularization suggests impaired perfusion, discrimination of a macro- versus microvascular etiology is uncertain. However, given an angiographically successful stent result without intraluminal filling defects or residual stenoses, macrovascular obstruction is absent and any increase (or lack of decrease) in RFC is likely a function of small vessel/microvascular obstruction or dysfunction.

The degree of pre-existing renal dysfunction, anatomic variations in vascular anatomy, time course of clinical follow-up, and lack of sensitive markers of parenchymal damage all contribute to the variability in the manifestation of complications due to embolization after renal stenting.²⁶ In a randomized, controlled trial exploring the benefit of embolic protection devices (EPD) and glycoprotein IIb/IIIa inhibitor use during renal artery stenting, little benefit of using either treatment alone was seen, but an additive advantage from concomitant use of both therapies²⁷ in preventing reduction of GFR after stenting was observed. The findings of the present study help shed light on the lack of benefit seen with the use of EPD to prevent atheroembolism during renal artery stenting. Our results impute one hypothesis that the use of EPD may only benefit subjects with relatively larger amounts of necrotic core and less fibrous tissue.

Study limitations. This study is limited by the sample size, which was not sufficient to allow for robust correlations based on clinical endpoints. Though this study shows that plaque composition throughout the atherosclerotic involvement of the renal artery is mostly fibrotic and consistent with previous autopsy studies, no direct validation between ex vivo histological renal artery plaque analysis and IVUS-VH derived data are available. Another potential confounder is that advancing the IVUS catheter itself could lead to distal embolization of atherosclerotic material. Measurement of RFC, although previously documented to correlate with improvement in blood pressure following renal artery stenting,⁸ has also not been validated against other measures of renal flow, including fractional flow reserve, renal resistive index or direct Doppler flow wire measurements. In this context, the relationship between plaque composition and change in RFC suggests a potentially important pathophysiological association and is hypothesis generating.

Conclusion

Both the MLD frame analysis and the segmental volumetric analysis of atherosclerotic renal artery stenosis lesions demonstrate predominantly fibrous tissue with smaller amounts of necrotic core, fibrofatty tissue, and dense calcification. There is an association with increasing necrotic core and impairment in renal perfusion following percutaneous revascularization. These findings warrant further evaluation in a larger study with physiological validation and clinical endpoints.

References

1. Beutler JJ, Van Ampting JM, Van De Ven PJ, et al. Long-term effects of arterial stenting on kidney function for patients with ostial atherosclerotic renal artery stenosis and renal insufficiency. J Am Soc Nephrol 2001;12:1475–1481.
2. Rocha-Singh KJ, Ahuja RK, Sung CH, Rutherford J. Long-term renal function preservation after renal artery stenting in patients with progressive ischemic nephropathy. Catheter Cardiovasc Interv 2002;57:135–141.
3. Zeller T, Frank U, Muller C, et al. Predictors of improved renal function after percutaneous stent-supported angioplasty of severe atherosclerotic ostial renal artery stenosis. Circulation 2003;108:2244–2249.
4. Bax L, Woittiez AJ, Kouwenberg HJ, et al. Stent placement in patients with atherosclerotic renal artery stenosis and impaired renal function: A randomized trial. Ann Intern Med 2009;150:840–848.
5. Edwards MS, Corriere MA, Craven TE, et al. Atheroembolism during percutaneous renal artery revascularization. J Vasc Surg 2007;46:55–61.
6. Hoste EA, Kellum JA. Incidence, classification, and outcomes of acute kidney injury. Contrib Nephrol 2007;156:32–38.
7. Feldman RL, Wargovich TJ, Bittl JA. No-touch technique for reducing aortic wall trauma during renal artery stenting. Catheter Cardiovasc Interv 1999;46:245–248.
8. Mahmud E, Smith W, Palakodeti V, et al. Renal frame count and renal blush grade: Quantitative measures that predict the success of renal stenting in hypertensive patients with renal artery stenosis. J Am Coll Cardiol Intv 2008;1:286–292.
9. Prasad A, Cipher DJ, Prasad A, et al. Reproducibility of intravascular ultrasound virtual histology analysis. Cardiovasc Revasc Med 2008;9:71–77.
10. Kaatee R, Beek FJ, Verschuyl EJ, et al. Atherosclerotic renal artery stenosis: Ostial or truncal? Radiology 1996;199:637–640.
11. Dangas G, Laird JR Jr, Mehran R, et al. Intravascular ultrasound-guided renal artery stenting. J Endovasc Ther 2001;8:238–247.
12. Leertouwer TC, Gussenhoven EJ, van Jaarsveld BC, et al. In vitro validation, with histology, of intravascular ultrasound in renal arteries. J Hypertens 1999;17:271–277.
13. Yasuda G, Takizawa T, Takasaki I, et al. Intravascular ultrasound imaging of atherosclerotic renal arteries: Comparison between in vitro and in vivo studies. Nephrol Dial Transplant 1998;13:1690–1695.
14. Nair A, Kuban BD, Tuzcu EM, et al. Coronary plaque classification with intravascular ultrasound radiofrequency data analysis. Circulation 2002;106:2200–06.
15. Kataoka T, Mathew V, Rubinshtein R, et al. Association of plaque composition and vessel remodeling in atherosclerotic renal artery stenosis: A comparison with coronary artery disease. JACC Cardiovasc Imaging 2009;2:327–338.
16. Prasad A, Zafar N, Mahmud E. Assessment of renal artery fibromuscular dysplasia: Angiography, intravascular ultrasound (with virtual histology), and pressure wire measurements. Catheter Cardiovasc Interv 2009;74:260–264.
17. Schiro BJ, Wholey MH. The expanding indications for virtual histology intravascular ultrasound for plaque analysis prior to carotid stenting. J Cardiovasc Surg (Torino) 2008;49:729–736.
18. Irshad K, Millar S, Velu R, et al. Virtual histology intravascular ultrasound in carotid interventions. J Endovasc Ther 2007;14:198–207.
19. Vince DG, Davies SC. Peripheral application of intravascular ultrasound virtual histology. Semin Vasc Surg 2004;17:119–125.

20. McCormack LJ, Poutasse EF, Meaney TF, et al. A pathologic-arteriographic correlation of renal arterial disease. Am Heart J 1966;72:188–198.
21. Holden A, Hill A, Jaff MR, Pilmore H. Renal artery stent revascularization with embolic protection in patients with ischemic nephropathy. Kidney Int 2006;70:948–955.
22. Kawarada O, Yokoi Y, Takemoto K. The characteristics of dissemination of embolic materials during renal artery stenting. Catheter Cardiovasc Interv 2007;70:784–788.
23. Virmani R, Burke AP, Farb A, Kolodgie FD. Pathology of the vulnerable plaque. J Am Coll Cardiol 2006;47(8 Suppl):C13–C18.
24. Hong MK, Mintz GS, Lee CW, et al. Comparison of virtual histology to intravascular ultrasound of culprit coronary lesions in acute coronary syndrome and target coronary lesions in stable angina pectoris. Am J Cardiol 2007;100:953–959.
25. Mulumudi MS, White CJ. Renal frame count: A quantitative angiographic assessment of renal perfusion. Catheter Cardiovasc Interv 2005;65:183–186.
26. Dubel GJ, Murphy TP. Distal embolic protection for renal arterial interventions. Cardiovasc Intervent Radiol 2008;31:14–22.
27. Cooper CJ, Haller ST, Colyer W, et al. Embolic protection and platelet inhibition during renal artery stenting. Circulation 2008;117:2752–2760.

_______________________________________

From the Division of Cardiology, UCSD Medical Center, University of California, San Diego, California.
The authors report no conflicts of interest regarding the content herein.
Manuscript submitted February 28, 2011, provisional acceptance given March 14, 2011 final version accepted March 30, 2011.
Address for correspondence: Dr. Ehtisham Mahmud, MD, University of California, San Diego, Division of Cardiology, UCSD Medical Center, 200 West Arbor Drive, San Diego, CA 92103-8784. Email: emahmud@ucsd.edu