Ultrasound velocity criteria for renal in-stent restenosis

Ultrasound velocity criteria for renal in-stent restenosis Yung-Wei Chi, DO, Christopher J. White, MD, Stanley Thornton, MD, and Richard V. Milani, MD, New Orleans, La Background: Renal artery stent placement is a recognized treatment for patients with hemodynamically significant renal artery stenosis when medical therapy fails. Duplex ultrasound (DUS) is the primary method used for noninvasive assessment of renal artery patency. Arterial stents alter the compliance of the artery, which could make the standard reference values, based on native renal artery velocities, inaccurate. This study attempted to determine DUS criteria for renal artery in-stent restenosis (ISR). Methods: We studied 67 consecutive patients with suspected renal artery ISR based on abnormal renal DUS results, defined as peak systolic velocity (PSV) >200 cm/s and renal/aortic velocity ratio (RAR) >3.5. The ISR patients were compared with 55 consecutive nonstented patients who underwent renal DUS evaluation and renal angiography. Those with >50% angiographic narrowing in each group were analyzed, and renal PSV and RAR were compared. Results: In the 67 patients with renal stents and 55 patients without renal stents, a statistically significant correlation was found for both PSV and RAR in detecting renal ISR and renal artery stenosis as defined by quantitative angiography (P.02). For any level of angiographic stenosis >50%, the ISR group had relatively higher PSV and RAR compared with the nonstented group. Receiver operating characteristic curves indicated that PSV >395 cm/s or RAR >5.1 were the most predictive of angiographically significant ISR >70%. Conclusion: The current DUS criteria for native renal arteries may overestimate the degree of angiographic ISR due to changes in compliance. We recommend that DUS laboratories make adjustments in PSV and RAR obtained by DUS when monitoring the patency of renal stents for ISR. (J Vasc Surg 2009;50:119-23.) Renal artery stent placement is a recognized treatment for patients with hemodynamically significant renal artery stenosis when medical therapy fails. 1-3 Surgical reconstruction or bypass is effective, but its use is limited by procedural morbidity and mortality. 4,5 In-stent restenosis (ISR) rates after renal artery stenting range up to 20%. 6-14 Despite the need for continuing surveillance for renal artery stent patency, guidelines for follow-up and noninvasive criteria for ISR have not been established. Assessment of ISR after renal stent placement is important for the clinical care of patients. Duplex ultrasound (DUS) imaging of the renal arteries is the primary noninvasive diagnostic method used to detect renal artery stenosis 15-17 and also has been used to monitor renal artery stent patency. The current DUS evaluation relies on blood velocity information to predict stenosis in the native (nonstented) renal arteries; these criteria have been validated by angiography. 16,17 The deployment of a metal stent results in changes in vessel compliance that affect the velocity of blood flow, which in turn affects DUS measurements. 18 This theory had been studied extensively and confirmed in the carotid stents. 18-20 Because the carotid and renal arteries are similar, both being 6-mm vessels, this concept was extrapolated to the renal artery stents, which may render From the Ochsner Clinic Foundation. Competition of interest: none. Reprint requests: Yung-Wei Chi, DO, Department of Cardiology, Section of Vascular Medicine, Ochsner Clinic Foundation, 2005 Veterans Blvd, Metairie, LA 70002 (e-mail: ychi@ochsner.org). 0741-5214/$36.00 Copyright 2009 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2008.12.066 the native (nonstented) renal artery velocity criteria inaccurate. 21,22 The purpose of this study was to provide DUS criteria for severe renal ISR 70%. METHODS Patients. Group I consisted of 67 consecutive patients with an abnormal renal DUS result, which was defined as DUS-predicted stenosis 60% after renal stent placement. These patients were referred for follow-up selective renal angiography 3 months of the abnormal renal DUS result. Group II consisted of 55 consecutive patients during the same period without renal stent placement. They underwent a renal DUS evaluation 3 months of a diagnostic renal angiogram as a part of coronary or peripheral angiography, or both. Ultrasound analysis. Group I patients (renal stents) had DUS examinations at follow-up intervals of 3, 6, and 12 months and then yearly after their procedures. De novo baseline DUS results for this group were not available. For group II patients (native renal arteries without stents), the index DUS was performed 3 months of the angiogram. All DUS examinations were performed by registered vascular technologists in the same vascular laboratory accredited by the Intersocietal Commission for the Accreditation of Vascular Laboratories using a Phillips HDI 5000 US machine (Phillips Medical Systems, Bothell, Wash) with a 3- to 5-MHz probe according to 2008 vascular professional performance guidelines established by the Society for Vascular Ultrasound. Velocities in group I were measured at three locations (proximal, middle, and distal) within the renal stent as well as distal to the stent. In group II, velocity measurements of the proximal, middle, and distal artery 119

120 Chi et al JOURNAL OF VASCULAR SURGERY July 2009 Table I. Mean peak systolic velocity and renal/aortic velocity ratio for 50% angiographic stenosis in both groups Mean of those with 50% stenosis Group % Stenosis PSV (cm/s) RAR Group I a 72 21 452 213 6.00 3.5 Group II b 70 20 360 153 4.91 2.7 P.66.002.020 PSV, peak systolic velocity; RAR, renal/aortic systolic velocity ratio. Fig 1. Scatterplots of the (A) peak systolic velocity (PSV) and (B) renal/aortic velocity ratio (RAR) correlated to angiographic stenosis. ISR, In-stent restenosis. were obtained. The highest measurements of peak systolic velocity (PSV) and end diastolic velocity (EDV) were recorded in both groups. Aortic velocity was taken at the abdominal aorta at the level of the renal arteries. The renal/aortic systolic velocity ratio (RAR) for each group was calculated by dividing the PSV for the respective group by the aortic PSV for that group. The standard DUS velocity criteria for 60% stenosis in native renal arteries (PSV 200 cm/s and RAR 3.5) were used to screen for renal ISR. 16,17 Two grades of stenosis, 0% to 59% and 60% to 99%, were determined from these criteria. Group I patients with DUS results showing stenosis 60% were referred for diagnostic angiography. Meanwhile, group II patients underwent a baseline DUS 3 months of a renal angiogram that was performed as a part of a coronary or peripheral angiographic study, or both. This DUS information was gathered and analyzed, and two experienced and certified readers interpreted the results. Interobserver agreement for the US results was assessed using statistics. Angiographic analysis. By convention, ISR is defined as a 50% angiographic narrowing within the stent. Selective renal angiography using anteroposterior and left anterior oblique views was performed. The view that demonstrated the most severe stenosis was used for quantitative measurements using electronic calipers and an automated edgedetection algorithm. 14 The percentage of stenosis was determined by using the minimal luminal diameter (MLD) divided by the reference vessel diameter (RVD) the diameter of the nearest normal appearing segment distal to the stenosis and expressed as a percentage. 14 The same protocols were used to measure ISR: The percentage of ISR was determined by using the narrowest stent diameter (MLD) divided by the normal appearing stent diameter distal to the ISR (RVD) or the distal normal appearing native renal artery (RVD). In both groups, only those with 50% angiographic stenosis were included in the final analysis. Three experienced angiographers performed and interpreted the angiograms. Interobserver agreement for the angiographic results was assessed using statistics. Statistical analysis. The DUS velocities in groups I and II were compared. Continuous variables were expressed as mean SD. Those with 50% angiographic narrowing in each group were analyzed, and the DUS values of renal PSV and RAR of the two groups were compared using a general linear model to account for the percentage of stenosis (Minitab Inc, State College, Pa). A value of P.05 was accepted as representing a significant difference. Receiver operating characteristic (ROC) curves were constructed using PSV and RAR as continuous variables to determine the DUS parameters with the highest sensitivity and specificity for detecting angiographic ISR 70%. RESULTS In group I, which included 67 patients with abnormal renal DUS results who underwent angiography, 31 (46%) had an angiographic narrowing of 50%. In group II, 30 of 55 patients (55%) had an angiographic narrowing of 50%. For all patients with 50% stenosis by angiography, the DUS findings in group I (n 31) were compared with group II (n 30). A statistically significant correlation was noted for both PSV (P.02) and RAR (P.02) in detecting angiographic renal ISR and renal artery stenosis (Fig 1). The mean angiographic stenosis was 72% 21% in group I compared with 70% 20% in group II (P.66; Table I). The mean PSV in group I was 452 cm/s com-

JOURNAL OF VASCULAR SURGERY Volume 50, Number 1 Chi et al 121 yielded the highest accuracy rate (88%), with a sensitivity of 94% and a specificity of 86% in detecting 70% ISR (Table III). Fig 2. The receiver operating curve of in-stent restenosis peak systolic velocity for detecting 70% in-stent restenosis in group I. Fig 3. Receiver operating curve of in-stent restenosis renal/aortic velocity ratio for detecting 70% in-stent restenosis in group I. pared with 360 cm/s in group II (P.002). The mean RAR in group I was 6.0 compared with 4.9 in group II (P.02). Both PSV and RAR increased to a greater degree in the ISR group per percentage of stenosis (Fig 1). In other words, for any level of angiographic stenosis 50%, the ISR group had higher PSV and RAR values than the nonstented group (Fig 1). A good level of interobserver agreement for the angiographic results between the three angiographers was noted (.82; P.05). For the DUS interpretations, there was good interobserver agreement between the two readers (.80; P.05). The ROC curves associated with the analysis of group I s PSV and RAR levels are shown in Figs 2 and 3. These findings were used to analyze multiple thresholds for PSV and RAR for accuracy in detecting the degrees of renal ISR. Independently, a PSV of 395 cm/s yielded a sensitivity of 83%, a specificity of 88%, and an overall accuracy of 87% to predict 70% ISR (Table II). Separately, an RAR of 5.1 DISCUSSION Renal artery DUS imaging is widely accepted as the noninvasive test of choice to estimate the severity of renal artery stenosis. Velocity information such as the PSV and RAR correlates with angiographic stenosis. The practice of using DUS criteria developed for native renal arteries to determine stenosis within renal artery stents has not been studied extensively. Metal stents result in alterations in the compliance of the vessel wall that reduce vascular compliance and increase the velocity of blood flow. 18,21,22 Appropriate DUS criteria for ISR are essential to noninvasively assess the restenosis rate after renal stenting to guide management and to reduce unnecessary invasive angiography and its associated costs and morbidity. Published studies for DUS criteria for renal ISR have used PSV, RAR, resistive index, or a combination of these to assess ISR, 21-23 whereas others used differences in resistive index, 23-26 acceleration time, acceleration index, and velocity waveform 26 to determine the degree of ISR. Bakker et al 21 and Napoli et al 22 established the usefulness of DUS for detecting renal ISR when laboratory-specific velocity threshold values were used. Bakker et al 21 recommended an increase in PSV from 180 to 226 cm/s and a lowering of RAR from 3.5 to 2.7 to increase sensitivity and specificity for ISR. Napoli et al, 22 on the other hand, showed a drop in accuracy in the detection of renal ISR if the velocity criteria for nonstented renal artery was applied, and further suggested lowering PSV (from 180 to 144 cm/s) and RAR (from 3.5 to 2.53 ) to increase sensitivity and specificity to detect ISR. In two recently published prospective studies, Nolan et al 27 found the current established velocity criteria for nonstented renal artery (PSV 200 cm/s or RAR 3.5) were highly predictive of renal ISR. In addition, Rocha- Singh et al 28 in the Renal Artery Stenting with Noninvasive Duplex Ultrasound Follow-up (RENAISSANCE) trial demonstrated a high concordance rate between renal artery DUS and angiographic renal ISR using velocity criteria for nonstented renal arteries (RAR 3.5 or an absolute PSV 225 cm/s). In our study, renal PSV and RAR were helpful in determining renal ISR, but both increased to a greater extent in patients with ISR (group I) than those without stents (group II). This contrasts with the conclusions of Bakker et al, 21 Nolan et al, 27 and Rocha-Singh et al. 28 However, the main distinguishing factor in our study compared with the others was the simultaneous comparison of the renal ISR group (group I) with the nonstented renal artery stenosis group (group II), as shown in Fig 1. In addition, the focus of this study was to determine velocity criteria for severe ISR 70%, which was not the case in other studies. Although all of the studies agreed on the usefulness of DUS imaging in detecting renal ISR, the disagreement lay in the velocity thresholds for PSV and

122 Chi et al JOURNAL OF VASCULAR SURGERY July 2009 Table II. Optimal ultrasound-determined peak systolic velocity thresholds to differentiate angiographic 50% to 69% and 70% stenosis Angiographic stenosis Threshold PSV (cm/s) Sensitivity, % Specificity, % PPV, % NPV, % Accuracy, % 50%-69% Group I a 350 80 87 83 85 84 Group II b 285 90 91 86 94 91 70% Group I a 395 83 88 71 93 87 Group II b 300 89 80 47 97 82 NPV, Negative-predictive value; PPV, positive-predictive value; PSV, peak systolic velocity. Table III. I. Optimal ultrasound-determined renal/aortic velocity ratio thresholds to differentiate angiographic 50% to 69% and 70% stenosis Angiographic stenosis Threshold RAR Sensitivity, % Specificity, % PPV, % NPV, % Accuracy, % 50%-69% Group I a 4.1 80 82 77 84 81 Group II b 3.6 76 85 76 85 82 70% Group I a 5.1 94 86 71 98 88 Group II b 3.7 89 74 40 97 76 NPV, Negative predictive value; PPV, positive predictive value; RAR, renal/aortic systolic velocity ratio. RAR to determine the degree of renal ISR. This disparity should be resolved as more robust prospective trials take place. This study demonstrated that for any degree of angiographic stenosis 50%, the ISR group had relatively higher PSV and RAR than those without stents (Fig 1). These findings are consistent with DUS findings in native and stented carotid arteries. 19,20 A ROC curve for both PSV and RAR in determining 70% angiographic narrowing determined a PSV of 395 cm/s yielded a sensitivity of 83%, a specificity of 88%, and an overall accuracy of 87% to predict 70% ISR, as noted earlier. Separately, a RAR of 5.1 yielded the highest accuracy rate (88%), with a sensitivity of 94% and a specificity of 86% in detecting 70% ISR. Combining PSV and RAR did not provide additional improvement in diagnostic value than either parameter alone. Before reintervention of renal ISR is attempted, one should consider the clinical parameters such as worsening blood pressure control or declining renal function, or both in conjunction with the abnormal DUS result and not act on just the abnormal DUS result alone. This study has some limitations. Despite being one of the largest series with renal ISR and DUS velocities reported to date, 20,21 this is a single-center, retrospective cohort study with a relatively small sample size. Additional flaws include selection bias and the lack of DUS information immediately before and after stenting. Therefore, there is an uncertainty about the ability to generalize our results. CONCLUSIONS The current criteria for DUS-determined renal artery stenosis, which is based on native (nonstented) renal arteries, may overestimate the degree of angiographic ISR. Surveillance monitoring for renal stent patency should take into account that PSV and RAR obtained by DUS are likely to be higher for any given degree of arterial narrowing within the stent. In light of the disparities among various published studies to date, further DUS analysis before and after stenting in a consecutive series of patients needs to be performed. We acknowledge T. Cooper Woods, PhD, for his statistical support. AUTHOR CONTRIBUTIONS Conception and design: YC, CW, ST Analysis and interpretation: YC, CW, ST, RM Data collection: YC, ST Writing the article: YC Critical revision of the article: CW, RM Final approval of the article: YC, CW, RM Statistical analysis: YC, CW, ST Obtained funding: CW Overall responsibility: YC REFERENCES 1. ACC/AHA 2005 guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic): executive summary a collaborative report from the

JOURNAL OF VASCULAR SURGERY Volume 50, Number 1 Chi et al 123 American Association for Vascular Surgery/Society for Vascular Surgery, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, Society of Interventional Radiology, and the ACC/AHA Task Force on Practice Guidelines (Writing Committee to Develop Guidelines for the Management of Patients With Peripheral Arterial Disease) endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation; National Heart, Lung, and Blood Institute; Society for Vascular Nursing; TransAtlantic Inter-Society Consensus; and Vascular Disease Foundation. J Am Coll Cardiol 2006;47:1239-312. 2. Messina LM, Zelenock GB, Yao KA, Stanley JC. Renal revascularization for recurrent pulmonary edema in patients with poorly controlled hypertension and renal insufficiency: a distinct subgroup of patients with arteriosclerotic renal artery occlusive disease. J Vasc Surg 1992;15: 73-80; discussion 80-2. 3. Pickering TG, Herman L, Devereux RB, Sotelo JE, James GD, Sos TA, et al. Recurrent pulmonary oedema in hypertension due to bilateral renal artery stenosis: treatment by angioplasty or surgical revascularisation. Lancet 1988;2:551-2. 4. Cambria RP, Brewster DC, L Italien GJ, Moncure A, Darling RC Jr, Gertler JP, et al. The durability of different reconstructive techniques for atherosclerotic renal artery disease. J Vasc Surg 1994;20:76-85; discussion 86-7. 5. Stanley JC. The evolution of surgery for renovascular occlusive disease. Cardiovasc Surg 1994;2:195-202. 6. Harden PN, MacLeod MJ, Rodger RS, Baxter GM, Connell JM, Dominiczak AF, et al. Effect of renal-artery stenting on progression of renovascular renal failure. Lancet 1997;349:1133-6. 7. van de Ven PJ, Beutler JJ, Kaatee R, Beek FJ, Mali WP, Geyskes GG, et al. Transluminal vascular stent for ostial atherosclerotic renal artery stenosis. Lancet 1995;346:672-4. 8. Iannone LA, Underwood PL, Nath A, Tannenbaum MA, Ghali MG, Clevenger LD. Effect of primary balloon expandable renal artery stents on long-term patency, renal function, and blood pressure in hypertensive and renal insufficient patients with renal artery stenosis. Cathet Cardiovasc Diagn 1996;37:243-50. 9. Henry M, Amor M, Henry I, Ethevenot G, Allaoui M, Tricoche O, et al. Stent placement in the renal artery: three-year experience with the Palmaz stent. J Vasc Interv Radiol 1996;7:343-50. 10. Blum U, Krumme B, Flügel P, Gabelmann A, Lehnert T, Buitrago- Tellez C, et al. Treatment of ostial renal-artery stenoses with vascular endoprostheses after unsuccessful balloon angioplasty. N Engl J Med 1997;336:459-65. 11. White CJ, Ramee SR, Collins TJ, Jenkins JS, Escobar A, Shaw D. Renal artery stent placement: utility in lesions difficult to treat with balloon angioplasty. J Am Coll Cardiol 1997;30:1445-50. 12. Dorros G, Jaff M, Mathiak L, Dorros II, Lowe A, Murphy K, et al. Four-year follow-up of Palmaz-Schatz stent revascularization as treatment for atherosclerotic renal artery stenosis. Circulation 1998;98: 642-7. 13. Dorros G, Jaff M, Mathiak L, He T; Multicenter Registry Participants. Multicenter Palmaz stent renal artery stenosis revascularization registry report: four-year follow-up of 1,058 successful patients. Catheter Cardiovasc Interv 2002;55:182-8. 14. Lederman RJ, Mendelsohn FO, Santos R, Phillips HR, Stack RS, Crowley JJ. Primary renal artery stenting: characteristics and outcomes after 363 procedures. Am Heart J 2001;142:314-23. 15. Carman TL, Olin JW, Czum J. Noninvasive imaging of the renal arteries. Urol Clin North Am 2001;28:815-26. 16. Olin JW. Role of duplex ultrasonography in screening for significant renal artery disease. Urol Clin North Am 1994;21:215-26. 17. Olin JW, Piedmonte MR, Young JR, DeAnna S, Grubb M, Childs MB. The utility of duplex ultrasound scanning of the renal arteries for diagnosing significant renal artery stenosis. Ann Intern Med 1995;122: 833-8. 18. Lal BK, Hobson RW, Goldstein J, Chakhtoura EY, Duran WN. Carotid artery stenting: is there a need to revise ultrasound velocity criteria? J Vasc Surg 2004;39:58-66. 19. Chi YW, White CJ, Woods TC, Goldman CK. Ultrasound velocity criteria for carotid in-stent restenosis. Catheter Cardiovasc Interv 2007; 69:349-54. 20. Stanziale SF, Wholey MH, Boules TN, Selzer F, Makaroun MS. Determining in-stent stenosis of carotid arteries by duplex ultrasound criteria. J Endovasc Ther 2005;12:346-53. 21. Bakker J, Beutler JJ, Elgersma OE, de Lange EE, de Kort GA, Beek FJ. Duplex ultrasonography in assessing restenosis of renal artery stents. Cardiovasc Intervent Radiol 1999;22:475-80. 22. Napoli V, Pinto S, Bargellini I, Vignali C, Cioni R, Petruzzi P, et al. Duplex ultrasonographic study of the renal arteries before and after renal artery stenting. Eur Radiol 2002;12:796-803. 23. Zeller T, Rastan A, Rothenpieler U, Müller C. Restenosis after stenting of atherosclerotic renal artery stenosis: is there a rationale for the use of drug-eluting stents? Catheter Cardiovasc Interv 2006;68:125-30. 24. Zeller T, Sixt S, Rastan A, Schwarzwälder U, Müller C, Frank U, et al. Treatment of reoccurring instent restenosis following reintervention after stent-supported renal artery angioplasty. Catheter Cardiovasc Interv 2007;70:296-300. 25. Zeller T, Rastan A, Schwarzwälder U, Mueller C, Schwarz T, Frank U, et al. Treatment of instent restenosis following stent-supported renal artery angioplasty. Catheter Cardiovasc Interv 2007;70:454-9. 26. Sharafuddin MJ, Raboi CA, Abu-Yousef M, Lawton WJ, Gordon JA. Renal artery stenosis: duplex US after angioplasty and stent placement. Radiology 2001;220:168-73. 27. Nolan BW, Schermerhorn ML, Powell RJ, Rowell E, et al. Restenosis in gold-coated renal artery stents. J Vasc Surg 2005;42:40-6. 28. Rocha-Singh K, Jaff MR, Lynne Kelley E; RENAISSANCE Trial Investigators. renal artery stenting with noninvasive duplex ultrasound follow-up: 3-year results from the RENAISSANCE renal stent trial. Catheter Cardiovasc Interv 2008;72:853-62. Submitted Nov 11, 2008; accepted Dec 23, 2008.