- Email: [email protected]
Identification of predictors for lower extremity vein graft stenosis
Identification of Predictors for Lower Extremity Vein Graft Stenosis Andrew T. Gentile, MD, Joseph L. Mills, MD, Michael A. Gooden, MD, Alex Westerband, MD, Haiyan Cui, PhD, Scott S. Berman, MD, Glenn C. Hunter, MD, John D. Hughes, MD, Tucson, Arizona
BACKGROUND: The cause of intrinsic vein graft stenosis, which develops in at least 20% of infrainguinal autogenous bypass grafts during the intermediate follow-up Interval, is unknown. We performed standard duplex surveillance of all lower extremity bypass grafts and evaluated the potential of comorbid patient risk factors that might predict development of vein graft flow disturbance or high-grade graft stenosis. METHODS: Patients with at least 6 months of postoperative duplex surveillance were identified through our vascular registry. The association of clinical and hemodynamic profiles of graft performance were compared with specific patient risk factors, including demographics, cigarette smokIng, antihypertensive medical therapy, type and quality of conduit, degree of Ischemia, bypass run-off, and presence of infection, using stepwise logistic regression analysis. RESULTS: Ninety-three patients (55 male, 38 female; mean age 69) underwent 100 Infrainguinal bypasses. Twenty-six high-grade graft stenoses (>70%) were identified in 26 patients during tollow-up (mean 21 months) by graft-flow peak systolic velocity (PSV) >300 cm/sec on more than one duplex examination, and were electively revised. Graft flow disturbances (180 cm/sec >PSV <300 cm/sec) were identified in an additional 13 grafts (6 regressed, 7 observed). The need for graft revision was associated with an early graft flow disturbance (P = 0.02), or drop In ankle-brachial index >0.15 (P = 0.03), and the use of an alternative conduit In 13 of 100 grafts (P = 0.04). Only smoking was associated with the development of a duplex detected graft flow disturbance during follow up (P = 0.03). CONCLUSION: Grafts with early flow disturbances warrant close duplex surveillance to identify graft-threatening stenosis. Risk factors that may predict future lower extremity bypass graft steno-
From the Division of Vascular Surgery (ATG, JLM, MAG, AW, SS8, GCH, JDH), Department of Surgery, and the Biometry Department (HC), Arizona Cancer Center, University of Arizona Health Sciences Center, Tucson. Arizona. Requests for reprints should be addressed to Joseph L. Mills, MD. Chief, Division of VaSCUlar Surgery, University of Arizona Health Sciences Center, PO 80x 245072. 1501 N. Campbell Ave., Tucson, Arizona 85724-5072. Presented at the 25th Annual Meeting of The Society for Clinical Vascular Surgery, Naples, Florida, March 12-16, 1997.
218
@ 1997 by Excerpta Medica. Inc. All rights reserved.
sis are smoking and the use of alternative bypass conduits. Am J Surg. 1997;174:218-221. © 1997 by Excerpta Medica, Inc. emodynamically significant intrinsic vein bypass graft stenosis has been consistently demonstrated to develop in 20% to 30% of infrainguinal grafts,l.z.3.4 usually within the intermediate (l to 18 month) follow-up interval. 5,6 Duplex surveillance of autogenous infrainguinal arterial reconstructions has proven accuracy in identifying these blood-flow-limiting vein graft lesions and directing bypass graft revision prior to thrombosis. Autogenous vein grafts revised while they remain patent have reported cumulative 5-year assisted primary patency rates ranging from 76% to 95%. These rates approach those of nonstenotic, never revised grafts.' Mattos et alB note a 90% 2-year graft patency rate for a subset of patients undergoing elective graft revision for stenosis versus a 33% 2-year patency rate for a similarly matched cohort of patients with nonrevised grafts. In addition to mechanical factors such as vein graft defects, retained fibrotic valves, sclerotic segments or areas of graft injury from clamps, handling or endarterectomy that can cause graft stenosis, patient comorbid risk factors that predispose to the development of intrinsic graft abnormalities have not been clearly identified. As the natural history of graft-threatening lesions is almost certain progression to thrombosis if not repaired," identification of factors that may predispose to the generation of these lesions is paramount. We undertook this study to identify which patient risk factors, if any, may predict early infrainguinal bypassgraft blood flow abnormalities or the need for graft revision for high-grade stenosis to prevent graft thrombosis.
H
METHODS All patients who underwent initial placement of an infrainguinal bypass graft for chronic lower extremity ischemia due to atherosclerosis and had at least 6 months of postoperative noninvasive bypass graft surveillance were identified through the computerized vascular registry of the University of Arizona Health Sciences Center. The last 100 consecutive bypass grafts with adequate follow-up were identified from the registry database and extensive review of patient comorbid risk factors and bypass graft performance was performed. Noninvasive Graft Surveillance Bypass graft stenoses were identified through a program of vascular laboratory surveillance that included serial history and physical examinations, ankle-brachial Doppler-derived systolic blood pressure indexes (ABIs), and 0002·9610/97/$17.00 PII S0002·961 0(97)00087-1
I PREDICTORS OF VEIN color flow duplex graft surveillance at 3-month intervals for the first year and 6-month intervals thereafter. Color duplex interrogation of the adjacent inflow and outflow segments, both anastomoses and the entire venous bypass conduit was performed with either a 5.0 or 7.5 MHz linear array probe. Representative center-stream velocity spectra were recorded at multiple graft sites, with attention to the anastomotic regions of the vein graft . Velocity spectral measurements were made at a Doppler angle of 60 degrees. If a focal color-flow disturbance (mosaic flow pattern) or "flow jet" was noted, careful interrogation with measurement of peak systolic velocity (PSV) and velocity ratio (Vr = PSV at stenosis, V 2. divided by normal adjacent PSV. VI) were noted ." The bypass graft was considered to have a graft flow disturbance if the PSV exceeded 180 cm/sec with a Vr > 1.5. These sites of flow disturbance were graded and monitored for progression, regression, or stabilization with more frequent graft surveillance (generally at 4 to 6 week intervals). Similarly, graft interrogation was performed for a clinically significant (>0.15) drop in ABI noted during follow-up. Preocclusive (high-grade, >70%) bypass graft stenoses were identified if a significant increase in graft PSV was noted (PSV >300 cm/sec, Vr > 3.5) or if a low flow state in the distal graft was detected (PSV <45 cm/sec).6 Patients suspected of having a high-grade vein graft stenosis or blood -flow-limiting inflow or outflow lesions by clinical findings, ABls, or duplex scan underwent arteriography prior to elective graft revision. Vein conduit quality was judged clinically as good, adequate, or poor based on size (>4mm), distensibility, and presence of gross defects, and analyzed using chi-square analysis with outcomes assigned as normal graft, graft flow disturbance, and graft-threatening stenosis. Similarly, extremity run-off was determined by preoperative arteriography, and a mod ified run-off index was applied for run-off versus outcomes with run-off grades based on the accepted criteria of the Ad Hoc Committee for the Society for Vascular Surgerv.'? good (run off index 1-3), adequate (4-6), or poor (7-10). Data Analysis Hospital records of all pat ients undergoing infrainguinal bypass were reviewed for operative factors including surgical indication, conduit type/source, and anastomotic sites. The usual patient comorbid risk factors assessed included patient age, gender, the presence of hypertension (systolic blood pressure> 160 mm Hg, diastolic blood pressure >90 mm Hg) or the current use of beta blockers or calcium channel blockers. coronary artery disease (defined by the presence of Q waves on electrocardiography or history of myocardial infarction. previous coronary artery bypass surgery, history of angina or congestive heart failure), peripheral vascular history, and the use of warfarin anticoagulation. Data for tobacco abuse were based on accepted risk factor index criteria of the Ad Hoc Committee. with positive tobacco use as a smoke index of 1 or greater (range 0 to 3).10 In addition, patient clinical symptoms such as claudication, rest pain or tissue loss. gangrene or ischemic ulcers, and wound/foot infection were noted for multivariate analysis. The clinical and hemodynamic profiles of graft performance were compared with patient comorbid risk factors
GRAFT STENOSIS/GENTILE ET AL
I
TABLE I Operative Results of 100 Infrainguinal Bypasses
Characteristic
Number
Above-knee femoral-popliteal Below-knee femoral-popliteal
13 35
~~
~
Pedal Reversed saphenous vein In situ saphenous vein Alternative conduit'
9
63 24 13
"P = 0.04 in predicting graft revision.
among patients with an entirely normal graft surveillance history (n = 61) , those with moderate early graft flow disturbance (I80 cm/sec >PSV <300 cm/sec, n = 13), and those undergoing elective revision for graft-threatening stenoses (PSV >300 cm/sec, n = 26) using stepwise multivariate logistic regression analysis. Data points consisting of the presence or absence of the above risk factors and objective measurements of noninvasive vascular testing were stored on a relati onal computer database (Microsoft Access). Patient demographic and continuous variables were expressed as descriptive statistics with normally distributed data and were expressed as the mean :t standard error of the mean. Baseline characteristics were compared using the chi-square test for categoric variables with continuous variables compared using Student's t test. All statistics and data analysis were performed by the Biometry Department of the Arizona Cancer Center at the University of Arizona. Data were entered into the SAS database for generation of logistic regression analysis results. A P value of less than 0.05 was considered significant.
RESULTS The study cohort consisted of 93 patients (55 male. 38 female; mean age 69 years, range 49 to 88) who underwent 100 infrainguinal revascularizations at the University of Arizona Health Sciences Center from 1991 through June 1996. All patients had at least 6 months postoperative follow-up with a mean follow-up ofll months (range 6 to 65) . These 93 patients had a mean of 3.6 duplex scans per patient during the follow-up period. The indication for placement of the original vein graft was critical ischemia in 64 patients, short distance claudication in 30 patients, and popliteal aneurysm in 6 patients. Neither the indication for lower extremity revascularizat ion nor the preoperative ABI correlated with early graft flow disturbance or need for subsequent graft revision for high grade stenosis by chi-square analysis. The distribution of distal anastomotic sites and graft conduits are presented in Table I. Twenty-six high-grade graft stenoses (> 70%) were identified in 26 patients during follow-up by graft-flow peak systolic velocity >300 em/sec, Vr >3.5. on more than one duplex examination with confirmatory arteriography. These 26 patients underwent elective graft revision at a mean of 7.1 months (range 1 to 17) following initial graft placement. Moderate bypass graft flow disturbances were identified in 13 grafts (6 regressed, 7 stable and under observation). The common coexistent medical risk factors present in this pa-
THE AMERICAN JOURNAL OF SURGERY· VOLUME 174 AUGUST 1997
219
PREDICTORS OF VEIN GRAFT STENOSIS/GENTILE ET AL
I
tient cohort were consistent with most other demographic reports describing patients requiring lower extremity revascularization (Table II). Of the risk factors analyzed, only smoking was associated with an early graft flow disturbance (P = 0.03) in the multivariate regression model. Similarly, the presence of a graftthreatening, high-grade stenosis was associated with an early graft flow disturbance (P = 0.02), drop in ABI >0.15 (P = 0.03), and the use of an alternative bypass conduit in 13 of 100 grafts (P = 0.04). An alternative conduit was defined as any graft not composed of a single segment of greater saphenous vein, and most often consisted of spliced segments of greater saphenous and arm vein. Revision to prevent graft thrombosis was required in 8 of 13 alternative bypass conduits, and 4 of 5 graft thromboses noted in the entire study cohort occurred in patients with alternative conduits. Other traditional atherosclerotic risk factors such as hypertension, diabetes, renal failure, and age were not associated with either graft flow disturbance or high-grade stenosis by multivariate modeling. Similarly, the current use of beta blockers (25%), calcium channel blockers (30%), or presence of wound/foot infection (12%) were not associated with hemodynamic graft abnormalities.
COMMENTS As biologic conduits, lower extremity vein grafts are ultimately subject to developing atherosclerotic degenerative changes when placed in the arterial circulation. However, these early developing graft abnormalities are not typically atherosclerotic in nature, but most often are focal, fibrotic intimal lesions that have been thought to be more consistent with myointimal hyperplasia. Due to the bypass grafts' accessability and readily studied hemodynamic characteristics, the factors that promote the development of these unique vein graft lesions are ideally suited for investigation. We have been interested in the proliferative cell types identified in these vein graft stenoses 11 as well as the composition of the extracellular matrix constituents of these lesions (unpublished data). Available data demonstrate that approximately 60% of vein graft failures result from intrinsic graft stenosis, with a lower proportion due to the progression of occlusive disease in the inflow and/or outflow vasculature.P:':' These progressive myointimal lesions are thought to be the most important cause of intermediate (1 to 18 months) infrainguinal vein graft stenosis and may represent a pathologic or maladaptive response of the vein graft to hemodynamic forces, which could be affected by individual patient risk factors. From these data, we conclude that early graft flow disturbances noted during graft surveillance and the quality of the bypass conduit are associated with the development of lower extremity vein graft stenosis. We consistently noted that in patients with entirely normal follow-up examinations virtually none developed significant graft abnormalities or spontaneous graft thrombosis. Of the 13 patients with low to moderate graft flow disturbances (180 cm/sec > PSV <300 cm/sed, none of these observed grafts with stable PSVs (7) or grafts with velocities noted to regress (6) progressed to graft thrombosis. Of the 13 grafts composed from alternative (not greater saphenous vein) sources of autogenous vein or composite 220
TABLE II Demographic Patient Risk Factors (n = 93)
Risk Factor Hypertension Diabetes Smoking Coronary disease Beta blocker Calcium channel blocker Quality vein (good) Quality run off (good) Wound/foot infection
Percent 65% 33% 67% 42% 25% 30% 49% 28% 12%
segments of vein, 8 of 13 grafts eventually required graft revision, and 4 of the 5 graft thromboses noted in this entire series occurred in patients with alternative vein conduits. The use of alternative venous conduits for lower extremity revascularization has been noted by others to double the risk of graft stenosis requiring revision. The Oregon group noted 26% of -altemarive infrainguinal bypass grafts required revision to maintain assisted patency compared with 12% of single-segment greater saphenous vein bypass conduits." Among the comorbid medical conditions considered in the multivariate model for predicting graft flow abnormalities, only smoking was independently associated with flow disturbances noted during follow-up (P = 0.03). Tobacco addiction remains a significant health problem in most industrialized countries, and is intimately linked to atherosclerotic disease of the entire cardiovascular system. Cigarette smoking is unquestionably the most preventable cause of cardiovascular morbidity and mortality and is known to be a potent risk factor for the development and progression of peripheral arterial disease. IS The association between smoking and cardiovascular disease is widely accepted," with smokers accounting for approximately 70% to 90% of patients with atherosclerosis obliterans. I? The relationship between risk factors and the presence of peripheral arterial disease has been well documented in several European studies.18 The effect of smoking on the extent of arterial disease has been reported in the Edinburgh Study of 617 patients with intermittent claudication who were compared with 722 control patients. This report identified several hemostatic and rheologic factors (fibrinogen) that were associated with intermittent claudication, and these factors were independent of smoking history. 19 Another large European study of more than 5,000 patients with peripheral arterial disease noted that the risk factors of smoking, diabetes, and increased age were associated with a higher prevalence of peripheral arterial dlsease.i? These population-based studies are mostly derived from risk factor assessment in patients with primary arteriosclerosis. Smoking has been assumed to have unfavorable effects on the patency of coronary and lower extremity saphenous bypass grafts and on the restenosis rate after carotid endarterectomy." Likewise, studies have revealed that smoking, among other comorbid risk factors, is implicated in vein bypass graft thrombosis.P:" In the British multicenter trial of antiplatelet drugs in patients undergoing lower extremity revascularization, multivariate analysis of risk factors re-
THE AMERICAN JOURNAL OF SURGERY. VOLUME 174 AUGUST 1997
PREDICTORS OF VEIN GRAFT STENOSIS/GENTILE ET AL
vealed that only ongoing cigarette use (as measured by serum thiocyanate levels) and an elevated fibrinogen level were associated with a higher rate of graft thrombosis.P In this study, patients were treated with either an antiplateler medication or a placebo. For vein grafts, 84% of the patients who were nonsmokers had patent grafts at 1 year compared with 63% of the smokers (P <0.02). Interestingly, these investigators also found ultrastructural changes of the vein walls, with thicker subendothelial basement membranes, occurred in smokers.P The mechanism by which tobacco accelerates atherosclerosis remain poorly understood but may include enhanced cellular proliferation in the vessel wall, loss of vasoactive or cytoprotecnve chemicals, and/or endothelial injury with increased permeability to lipoproteins or increased thrornbogenicity.2l.Z6 Smoking has been shown experimentally to inhibit the activity of endothelial nitric oxide synthase, a 26 potent vasodilator, in saphenous grafts. It seems logical that any vasoconstrictor may act synergistically with other medical or mechanical factors to promote exaggerated intimal remodeling in response to hemodynamic forces within arterialized bypass conduits. As most extremity vein graft stenoses occur during intermediate (l to 18 month) followup, recommendations for more compulsive early graft surveillance during this interval seem justified . There is little or no direct ev idence of the association of specific risk factors and early myointimal vein graft lesions . The forces initiating this process require investigation. In addition, identified risk factors (such as smoking) that may promote growth of these lesions independently or synergistically with other potential risk factors should be modified in the hopes of optimizing graft flow hemodynamics. We are unaware of any current pro spective follow -up studies documenting an association of specific risk factors on the hemodynamic performance of infrainguinal bypass grafts. In the present study, only smoking was associated with the development of early graft flow disturbances. We continue to recommend close graft surveillance during the first 1 to 2 years after infra inguinal revascularization when intrinsic vein graft abnormalities are most likely to appear and progress. Conduit quality and early developing graft flow abnormalities noted during graft surveillance continue to pro vide the most reliable information on the graft's current and expected performance.
REFERENCES
1. Szilagyi DE, Elliot JP, Hagemann JH, et al. Biologic fate of autogenous vein implants as arterial substitutes: clinical, angiographic, and histopathologic observations in femoro-popliteal operations for atherosclerosis. Ann Surg. 1973;178:232-246. 2. Mills [L, Bandyk OF, Gahtan V, Esses GE. The origin of intrainguinal vein graft stenosis : a prospective study based on duplex surveillance. ] Vase Surg. 1995;21 :16-25. 3. Taylor PR, Wolfe JHN, Tyrrell MR, et al. Graft stenosis-justification for l-year surveillance. BrJ Surg. 1990;77:1125-1128. 4. Erickson CA , Towne JB, Seabrook GR, et al. Ongoing vascular laboratory surveill ance is essential to maximize long-term in situ saphenous vein bypass patency. ] VaseSurg. 1996;23:18-27.
5. Nehler MR, Moneta Gl., Yeager RA, et al, Surgical treatment of threatened reversed infraingulnal vein grafts. ] Vase Surg.
1994;20:558-565. 6. Mills JL. The impact of duplex surveillance on infrainguinal bypass surgery. Adllanc Vasc Surg. 1996;4:201-210. 7. Sumner OS, Mattos MA, Hodgson KJ. Surveillance program for vascular reconstructive procedures. In: Yao JST , Pearce WH, eds. Long-term Results in Vascular Surgery. Norwalk, Conn: Appleton & Lange; 1993:33-59. 8. Mattos MA, Van Bemmelin PS, Hodgson KJ, et al. Does correction of stenoses identified with color duplex scanning improve infrainguinal graft patency? J Vase Surg. 1992;15:1059. Abstract. 9. Idu MM, Blankenstein JD, de Gier P, et al. Impact of a colorflow duplex surveillance program on infrainguinal vein graft patency: a five-year experience. ] Vase Surg. 1993;17:42-53. 10. Ad Hoc Committee on Reporting Standards, Society for Vascular Surgery. J Vase Surg. 1986;4:80-94. 11. Westerband A, Mills JL, Marek JM, et al. Immunocytochemical determination of cell type and proliferation rate in human vein graft stenoses. ] Vase Surg. 1997;25:64-67. 12. Whittemore AD, Clowes AW . Couch NP, Mannick JA . Secondary femoropopliteal reconstruction. Ann Surg. 1981 ;193:35-42. 13. Mills n, Fujitani RM. Taylor SM . The characteristics and anatomic distribution of lesions that cause reversed vein graft failure: a five-year prospective study. ] Vasc Surg . 1993;17:195-206. 14. Gentile AT. Lee RW , Moneta Gl., et al. Results of bypass to the popliteal and tibial arteries with alternative sources of autogenous vein. ] Vasc Surg. 1996;23:272-280. 15. Fowkes FGR . The aetiology of peripheral atherosclerosis. 8M] . 1989;298 :405-406. 16. Balkau B, Vray M, Eschwege E. Epidemiology of peripheral arterial disease. ] Cardiollase Pharm . 1994;23(suppI3):8S-16S. 17. lakier JB. Smoking and cardiovascular disease. Am ] Med . 1992;930A):8S-12S. 18. Novo S, Avellone G . DiGarbo V, et al. Prevalence of risk factors in patients with peripheral arterial disease. A clinical and epidemiological evaluation. Inl Angio. 1992;11:218-229. 19. Lee AJ, Fowkes FG, Rattray A, et al. Hemostatic and rheological factors in intermittent claud ication: the influence of smoking and extent of arterial disease . Brit] Haemat . 1996;92( 0 :226-230. 20. Newman AB, Siscovick OS, Manolio T A, er at. Ankle-arm index as a marker of atherosclerosis in the Cardiovascular Health Study. Cardiovascular Heart Study (CHS) Collaborative Research Group. Circulation . 1993;88:837-845. 21. Cooke JP, Ma AO. Medical therapy of peripheral arterial occlusive disease. In : Gerwertz Bl., ed. Surgical Clinics of North America. Philadelphia: WB Saunders; 1995;75:569-579. 22. Powell JT, Higman 0, Greenhalgh RM. Smoking and its influence on graft patency. In: Yao J, Pearce W, eds. Long-term Results ira Vascular Surgery . Norwalk, Conn: Appleton & Lange; 1993:917. 23. Cheshire NJ, Wolfe JH , Barradas MA, et al. Smoking and plasma fibrinogen, lipoprotein a and serotonin are markers for postoperative infrainguinal graft stenosis. Euro J Vasc Endooasc Surg . 1996;11:479-486. 24. Wiseman S, Kenchington G , Dain R, et al. Influence of smoking and plasma factors on femoropopliteal vein graft patency. 8M] . 1989;299:643-646. 25. Meade TW.lverson J, St irling Y. Effects of changes in smoking and other characteristics on clotting factors and the risks of ischemic heart disease. Lancet. 1987;ii:986-988. 26. Higman OJ. Srrachan AMJ. Buttery L, et al. Sm oking impairs the activity of endothelial nitric ox ide synthase in saphenous vein. Arterioscler Thromb Vase Bioi. 1996;16:546-552.
THE AMERICAN JOURNAL OF SURGERY. VOLUME 174 AUGUST 1997
221
BACKGROUND: The cause of intrinsic vein graft stenosis, which develops in at least 20% of infrainguinal autogenous bypass grafts during the intermediate follow-up Interval, is unknown. We performed standard duplex surveillance of all lower extremity bypass grafts and evaluated the potential of comorbid patient risk factors that might predict development of vein graft flow disturbance or high-grade graft stenosis. METHODS: Patients with at least 6 months of postoperative duplex surveillance were identified through our vascular registry. The association of clinical and hemodynamic profiles of graft performance were compared with specific patient risk factors, including demographics, cigarette smokIng, antihypertensive medical therapy, type and quality of conduit, degree of Ischemia, bypass run-off, and presence of infection, using stepwise logistic regression analysis. RESULTS: Ninety-three patients (55 male, 38 female; mean age 69) underwent 100 Infrainguinal bypasses. Twenty-six high-grade graft stenoses (>70%) were identified in 26 patients during tollow-up (mean 21 months) by graft-flow peak systolic velocity (PSV) >300 cm/sec on more than one duplex examination, and were electively revised. Graft flow disturbances (180 cm/sec >PSV <300 cm/sec) were identified in an additional 13 grafts (6 regressed, 7 observed). The need for graft revision was associated with an early graft flow disturbance (P = 0.02), or drop In ankle-brachial index >0.15 (P = 0.03), and the use of an alternative conduit In 13 of 100 grafts (P = 0.04). Only smoking was associated with the development of a duplex detected graft flow disturbance during follow up (P = 0.03). CONCLUSION: Grafts with early flow disturbances warrant close duplex surveillance to identify graft-threatening stenosis. Risk factors that may predict future lower extremity bypass graft steno-
From the Division of Vascular Surgery (ATG, JLM, MAG, AW, SS8, GCH, JDH), Department of Surgery, and the Biometry Department (HC), Arizona Cancer Center, University of Arizona Health Sciences Center, Tucson. Arizona. Requests for reprints should be addressed to Joseph L. Mills, MD. Chief, Division of VaSCUlar Surgery, University of Arizona Health Sciences Center, PO 80x 245072. 1501 N. Campbell Ave., Tucson, Arizona 85724-5072. Presented at the 25th Annual Meeting of The Society for Clinical Vascular Surgery, Naples, Florida, March 12-16, 1997.
218
@ 1997 by Excerpta Medica. Inc. All rights reserved.
sis are smoking and the use of alternative bypass conduits. Am J Surg. 1997;174:218-221. © 1997 by Excerpta Medica, Inc. emodynamically significant intrinsic vein bypass graft stenosis has been consistently demonstrated to develop in 20% to 30% of infrainguinal grafts,l.z.3.4 usually within the intermediate (l to 18 month) follow-up interval. 5,6 Duplex surveillance of autogenous infrainguinal arterial reconstructions has proven accuracy in identifying these blood-flow-limiting vein graft lesions and directing bypass graft revision prior to thrombosis. Autogenous vein grafts revised while they remain patent have reported cumulative 5-year assisted primary patency rates ranging from 76% to 95%. These rates approach those of nonstenotic, never revised grafts.' Mattos et alB note a 90% 2-year graft patency rate for a subset of patients undergoing elective graft revision for stenosis versus a 33% 2-year patency rate for a similarly matched cohort of patients with nonrevised grafts. In addition to mechanical factors such as vein graft defects, retained fibrotic valves, sclerotic segments or areas of graft injury from clamps, handling or endarterectomy that can cause graft stenosis, patient comorbid risk factors that predispose to the development of intrinsic graft abnormalities have not been clearly identified. As the natural history of graft-threatening lesions is almost certain progression to thrombosis if not repaired," identification of factors that may predispose to the generation of these lesions is paramount. We undertook this study to identify which patient risk factors, if any, may predict early infrainguinal bypassgraft blood flow abnormalities or the need for graft revision for high-grade stenosis to prevent graft thrombosis.
H
METHODS All patients who underwent initial placement of an infrainguinal bypass graft for chronic lower extremity ischemia due to atherosclerosis and had at least 6 months of postoperative noninvasive bypass graft surveillance were identified through the computerized vascular registry of the University of Arizona Health Sciences Center. The last 100 consecutive bypass grafts with adequate follow-up were identified from the registry database and extensive review of patient comorbid risk factors and bypass graft performance was performed. Noninvasive Graft Surveillance Bypass graft stenoses were identified through a program of vascular laboratory surveillance that included serial history and physical examinations, ankle-brachial Doppler-derived systolic blood pressure indexes (ABIs), and 0002·9610/97/$17.00 PII S0002·961 0(97)00087-1
I PREDICTORS OF VEIN color flow duplex graft surveillance at 3-month intervals for the first year and 6-month intervals thereafter. Color duplex interrogation of the adjacent inflow and outflow segments, both anastomoses and the entire venous bypass conduit was performed with either a 5.0 or 7.5 MHz linear array probe. Representative center-stream velocity spectra were recorded at multiple graft sites, with attention to the anastomotic regions of the vein graft . Velocity spectral measurements were made at a Doppler angle of 60 degrees. If a focal color-flow disturbance (mosaic flow pattern) or "flow jet" was noted, careful interrogation with measurement of peak systolic velocity (PSV) and velocity ratio (Vr = PSV at stenosis, V 2. divided by normal adjacent PSV. VI) were noted ." The bypass graft was considered to have a graft flow disturbance if the PSV exceeded 180 cm/sec with a Vr > 1.5. These sites of flow disturbance were graded and monitored for progression, regression, or stabilization with more frequent graft surveillance (generally at 4 to 6 week intervals). Similarly, graft interrogation was performed for a clinically significant (>0.15) drop in ABI noted during follow-up. Preocclusive (high-grade, >70%) bypass graft stenoses were identified if a significant increase in graft PSV was noted (PSV >300 cm/sec, Vr > 3.5) or if a low flow state in the distal graft was detected (PSV <45 cm/sec).6 Patients suspected of having a high-grade vein graft stenosis or blood -flow-limiting inflow or outflow lesions by clinical findings, ABls, or duplex scan underwent arteriography prior to elective graft revision. Vein conduit quality was judged clinically as good, adequate, or poor based on size (>4mm), distensibility, and presence of gross defects, and analyzed using chi-square analysis with outcomes assigned as normal graft, graft flow disturbance, and graft-threatening stenosis. Similarly, extremity run-off was determined by preoperative arteriography, and a mod ified run-off index was applied for run-off versus outcomes with run-off grades based on the accepted criteria of the Ad Hoc Committee for the Society for Vascular Surgerv.'? good (run off index 1-3), adequate (4-6), or poor (7-10). Data Analysis Hospital records of all pat ients undergoing infrainguinal bypass were reviewed for operative factors including surgical indication, conduit type/source, and anastomotic sites. The usual patient comorbid risk factors assessed included patient age, gender, the presence of hypertension (systolic blood pressure> 160 mm Hg, diastolic blood pressure >90 mm Hg) or the current use of beta blockers or calcium channel blockers. coronary artery disease (defined by the presence of Q waves on electrocardiography or history of myocardial infarction. previous coronary artery bypass surgery, history of angina or congestive heart failure), peripheral vascular history, and the use of warfarin anticoagulation. Data for tobacco abuse were based on accepted risk factor index criteria of the Ad Hoc Committee. with positive tobacco use as a smoke index of 1 or greater (range 0 to 3).10 In addition, patient clinical symptoms such as claudication, rest pain or tissue loss. gangrene or ischemic ulcers, and wound/foot infection were noted for multivariate analysis. The clinical and hemodynamic profiles of graft performance were compared with patient comorbid risk factors
GRAFT STENOSIS/GENTILE ET AL
I
TABLE I Operative Results of 100 Infrainguinal Bypasses
Characteristic
Number
Above-knee femoral-popliteal Below-knee femoral-popliteal
13 35
~~
~
Pedal Reversed saphenous vein In situ saphenous vein Alternative conduit'
9
63 24 13
"P = 0.04 in predicting graft revision.
among patients with an entirely normal graft surveillance history (n = 61) , those with moderate early graft flow disturbance (I80 cm/sec >PSV <300 cm/sec, n = 13), and those undergoing elective revision for graft-threatening stenoses (PSV >300 cm/sec, n = 26) using stepwise multivariate logistic regression analysis. Data points consisting of the presence or absence of the above risk factors and objective measurements of noninvasive vascular testing were stored on a relati onal computer database (Microsoft Access). Patient demographic and continuous variables were expressed as descriptive statistics with normally distributed data and were expressed as the mean :t standard error of the mean. Baseline characteristics were compared using the chi-square test for categoric variables with continuous variables compared using Student's t test. All statistics and data analysis were performed by the Biometry Department of the Arizona Cancer Center at the University of Arizona. Data were entered into the SAS database for generation of logistic regression analysis results. A P value of less than 0.05 was considered significant.
RESULTS The study cohort consisted of 93 patients (55 male. 38 female; mean age 69 years, range 49 to 88) who underwent 100 infrainguinal revascularizations at the University of Arizona Health Sciences Center from 1991 through June 1996. All patients had at least 6 months postoperative follow-up with a mean follow-up ofll months (range 6 to 65) . These 93 patients had a mean of 3.6 duplex scans per patient during the follow-up period. The indication for placement of the original vein graft was critical ischemia in 64 patients, short distance claudication in 30 patients, and popliteal aneurysm in 6 patients. Neither the indication for lower extremity revascularizat ion nor the preoperative ABI correlated with early graft flow disturbance or need for subsequent graft revision for high grade stenosis by chi-square analysis. The distribution of distal anastomotic sites and graft conduits are presented in Table I. Twenty-six high-grade graft stenoses (> 70%) were identified in 26 patients during follow-up by graft-flow peak systolic velocity >300 em/sec, Vr >3.5. on more than one duplex examination with confirmatory arteriography. These 26 patients underwent elective graft revision at a mean of 7.1 months (range 1 to 17) following initial graft placement. Moderate bypass graft flow disturbances were identified in 13 grafts (6 regressed, 7 stable and under observation). The common coexistent medical risk factors present in this pa-
THE AMERICAN JOURNAL OF SURGERY· VOLUME 174 AUGUST 1997
219
PREDICTORS OF VEIN GRAFT STENOSIS/GENTILE ET AL
I
tient cohort were consistent with most other demographic reports describing patients requiring lower extremity revascularization (Table II). Of the risk factors analyzed, only smoking was associated with an early graft flow disturbance (P = 0.03) in the multivariate regression model. Similarly, the presence of a graftthreatening, high-grade stenosis was associated with an early graft flow disturbance (P = 0.02), drop in ABI >0.15 (P = 0.03), and the use of an alternative bypass conduit in 13 of 100 grafts (P = 0.04). An alternative conduit was defined as any graft not composed of a single segment of greater saphenous vein, and most often consisted of spliced segments of greater saphenous and arm vein. Revision to prevent graft thrombosis was required in 8 of 13 alternative bypass conduits, and 4 of 5 graft thromboses noted in the entire study cohort occurred in patients with alternative conduits. Other traditional atherosclerotic risk factors such as hypertension, diabetes, renal failure, and age were not associated with either graft flow disturbance or high-grade stenosis by multivariate modeling. Similarly, the current use of beta blockers (25%), calcium channel blockers (30%), or presence of wound/foot infection (12%) were not associated with hemodynamic graft abnormalities.
COMMENTS As biologic conduits, lower extremity vein grafts are ultimately subject to developing atherosclerotic degenerative changes when placed in the arterial circulation. However, these early developing graft abnormalities are not typically atherosclerotic in nature, but most often are focal, fibrotic intimal lesions that have been thought to be more consistent with myointimal hyperplasia. Due to the bypass grafts' accessability and readily studied hemodynamic characteristics, the factors that promote the development of these unique vein graft lesions are ideally suited for investigation. We have been interested in the proliferative cell types identified in these vein graft stenoses 11 as well as the composition of the extracellular matrix constituents of these lesions (unpublished data). Available data demonstrate that approximately 60% of vein graft failures result from intrinsic graft stenosis, with a lower proportion due to the progression of occlusive disease in the inflow and/or outflow vasculature.P:':' These progressive myointimal lesions are thought to be the most important cause of intermediate (1 to 18 months) infrainguinal vein graft stenosis and may represent a pathologic or maladaptive response of the vein graft to hemodynamic forces, which could be affected by individual patient risk factors. From these data, we conclude that early graft flow disturbances noted during graft surveillance and the quality of the bypass conduit are associated with the development of lower extremity vein graft stenosis. We consistently noted that in patients with entirely normal follow-up examinations virtually none developed significant graft abnormalities or spontaneous graft thrombosis. Of the 13 patients with low to moderate graft flow disturbances (180 cm/sec > PSV <300 cm/sed, none of these observed grafts with stable PSVs (7) or grafts with velocities noted to regress (6) progressed to graft thrombosis. Of the 13 grafts composed from alternative (not greater saphenous vein) sources of autogenous vein or composite 220
TABLE II Demographic Patient Risk Factors (n = 93)
Risk Factor Hypertension Diabetes Smoking Coronary disease Beta blocker Calcium channel blocker Quality vein (good) Quality run off (good) Wound/foot infection
Percent 65% 33% 67% 42% 25% 30% 49% 28% 12%
segments of vein, 8 of 13 grafts eventually required graft revision, and 4 of the 5 graft thromboses noted in this entire series occurred in patients with alternative vein conduits. The use of alternative venous conduits for lower extremity revascularization has been noted by others to double the risk of graft stenosis requiring revision. The Oregon group noted 26% of -altemarive infrainguinal bypass grafts required revision to maintain assisted patency compared with 12% of single-segment greater saphenous vein bypass conduits." Among the comorbid medical conditions considered in the multivariate model for predicting graft flow abnormalities, only smoking was independently associated with flow disturbances noted during follow-up (P = 0.03). Tobacco addiction remains a significant health problem in most industrialized countries, and is intimately linked to atherosclerotic disease of the entire cardiovascular system. Cigarette smoking is unquestionably the most preventable cause of cardiovascular morbidity and mortality and is known to be a potent risk factor for the development and progression of peripheral arterial disease. IS The association between smoking and cardiovascular disease is widely accepted," with smokers accounting for approximately 70% to 90% of patients with atherosclerosis obliterans. I? The relationship between risk factors and the presence of peripheral arterial disease has been well documented in several European studies.18 The effect of smoking on the extent of arterial disease has been reported in the Edinburgh Study of 617 patients with intermittent claudication who were compared with 722 control patients. This report identified several hemostatic and rheologic factors (fibrinogen) that were associated with intermittent claudication, and these factors were independent of smoking history. 19 Another large European study of more than 5,000 patients with peripheral arterial disease noted that the risk factors of smoking, diabetes, and increased age were associated with a higher prevalence of peripheral arterial dlsease.i? These population-based studies are mostly derived from risk factor assessment in patients with primary arteriosclerosis. Smoking has been assumed to have unfavorable effects on the patency of coronary and lower extremity saphenous bypass grafts and on the restenosis rate after carotid endarterectomy." Likewise, studies have revealed that smoking, among other comorbid risk factors, is implicated in vein bypass graft thrombosis.P:" In the British multicenter trial of antiplatelet drugs in patients undergoing lower extremity revascularization, multivariate analysis of risk factors re-
THE AMERICAN JOURNAL OF SURGERY. VOLUME 174 AUGUST 1997
PREDICTORS OF VEIN GRAFT STENOSIS/GENTILE ET AL
vealed that only ongoing cigarette use (as measured by serum thiocyanate levels) and an elevated fibrinogen level were associated with a higher rate of graft thrombosis.P In this study, patients were treated with either an antiplateler medication or a placebo. For vein grafts, 84% of the patients who were nonsmokers had patent grafts at 1 year compared with 63% of the smokers (P <0.02). Interestingly, these investigators also found ultrastructural changes of the vein walls, with thicker subendothelial basement membranes, occurred in smokers.P The mechanism by which tobacco accelerates atherosclerosis remain poorly understood but may include enhanced cellular proliferation in the vessel wall, loss of vasoactive or cytoprotecnve chemicals, and/or endothelial injury with increased permeability to lipoproteins or increased thrornbogenicity.2l.Z6 Smoking has been shown experimentally to inhibit the activity of endothelial nitric oxide synthase, a 26 potent vasodilator, in saphenous grafts. It seems logical that any vasoconstrictor may act synergistically with other medical or mechanical factors to promote exaggerated intimal remodeling in response to hemodynamic forces within arterialized bypass conduits. As most extremity vein graft stenoses occur during intermediate (l to 18 month) followup, recommendations for more compulsive early graft surveillance during this interval seem justified . There is little or no direct ev idence of the association of specific risk factors and early myointimal vein graft lesions . The forces initiating this process require investigation. In addition, identified risk factors (such as smoking) that may promote growth of these lesions independently or synergistically with other potential risk factors should be modified in the hopes of optimizing graft flow hemodynamics. We are unaware of any current pro spective follow -up studies documenting an association of specific risk factors on the hemodynamic performance of infrainguinal bypass grafts. In the present study, only smoking was associated with the development of early graft flow disturbances. We continue to recommend close graft surveillance during the first 1 to 2 years after infra inguinal revascularization when intrinsic vein graft abnormalities are most likely to appear and progress. Conduit quality and early developing graft flow abnormalities noted during graft surveillance continue to pro vide the most reliable information on the graft's current and expected performance.
REFERENCES
1. Szilagyi DE, Elliot JP, Hagemann JH, et al. Biologic fate of autogenous vein implants as arterial substitutes: clinical, angiographic, and histopathologic observations in femoro-popliteal operations for atherosclerosis. Ann Surg. 1973;178:232-246. 2. Mills [L, Bandyk OF, Gahtan V, Esses GE. The origin of intrainguinal vein graft stenosis : a prospective study based on duplex surveillance. ] Vase Surg. 1995;21 :16-25. 3. Taylor PR, Wolfe JHN, Tyrrell MR, et al. Graft stenosis-justification for l-year surveillance. BrJ Surg. 1990;77:1125-1128. 4. Erickson CA , Towne JB, Seabrook GR, et al. Ongoing vascular laboratory surveill ance is essential to maximize long-term in situ saphenous vein bypass patency. ] VaseSurg. 1996;23:18-27.
5. Nehler MR, Moneta Gl., Yeager RA, et al, Surgical treatment of threatened reversed infraingulnal vein grafts. ] Vase Surg.
1994;20:558-565. 6. Mills JL. The impact of duplex surveillance on infrainguinal bypass surgery. Adllanc Vasc Surg. 1996;4:201-210. 7. Sumner OS, Mattos MA, Hodgson KJ. Surveillance program for vascular reconstructive procedures. In: Yao JST , Pearce WH, eds. Long-term Results in Vascular Surgery. Norwalk, Conn: Appleton & Lange; 1993:33-59. 8. Mattos MA, Van Bemmelin PS, Hodgson KJ, et al. Does correction of stenoses identified with color duplex scanning improve infrainguinal graft patency? J Vase Surg. 1992;15:1059. Abstract. 9. Idu MM, Blankenstein JD, de Gier P, et al. Impact of a colorflow duplex surveillance program on infrainguinal vein graft patency: a five-year experience. ] Vase Surg. 1993;17:42-53. 10. Ad Hoc Committee on Reporting Standards, Society for Vascular Surgery. J Vase Surg. 1986;4:80-94. 11. Westerband A, Mills JL, Marek JM, et al. Immunocytochemical determination of cell type and proliferation rate in human vein graft stenoses. ] Vase Surg. 1997;25:64-67. 12. Whittemore AD, Clowes AW . Couch NP, Mannick JA . Secondary femoropopliteal reconstruction. Ann Surg. 1981 ;193:35-42. 13. Mills n, Fujitani RM. Taylor SM . The characteristics and anatomic distribution of lesions that cause reversed vein graft failure: a five-year prospective study. ] Vasc Surg . 1993;17:195-206. 14. Gentile AT. Lee RW , Moneta Gl., et al. Results of bypass to the popliteal and tibial arteries with alternative sources of autogenous vein. ] Vasc Surg. 1996;23:272-280. 15. Fowkes FGR . The aetiology of peripheral atherosclerosis. 8M] . 1989;298 :405-406. 16. Balkau B, Vray M, Eschwege E. Epidemiology of peripheral arterial disease. ] Cardiollase Pharm . 1994;23(suppI3):8S-16S. 17. lakier JB. Smoking and cardiovascular disease. Am ] Med . 1992;930A):8S-12S. 18. Novo S, Avellone G . DiGarbo V, et al. Prevalence of risk factors in patients with peripheral arterial disease. A clinical and epidemiological evaluation. Inl Angio. 1992;11:218-229. 19. Lee AJ, Fowkes FG, Rattray A, et al. Hemostatic and rheological factors in intermittent claud ication: the influence of smoking and extent of arterial disease . Brit] Haemat . 1996;92( 0 :226-230. 20. Newman AB, Siscovick OS, Manolio T A, er at. Ankle-arm index as a marker of atherosclerosis in the Cardiovascular Health Study. Cardiovascular Heart Study (CHS) Collaborative Research Group. Circulation . 1993;88:837-845. 21. Cooke JP, Ma AO. Medical therapy of peripheral arterial occlusive disease. In : Gerwertz Bl., ed. Surgical Clinics of North America. Philadelphia: WB Saunders; 1995;75:569-579. 22. Powell JT, Higman 0, Greenhalgh RM. Smoking and its influence on graft patency. In: Yao J, Pearce W, eds. Long-term Results ira Vascular Surgery . Norwalk, Conn: Appleton & Lange; 1993:917. 23. Cheshire NJ, Wolfe JH , Barradas MA, et al. Smoking and plasma fibrinogen, lipoprotein a and serotonin are markers for postoperative infrainguinal graft stenosis. Euro J Vasc Endooasc Surg . 1996;11:479-486. 24. Wiseman S, Kenchington G , Dain R, et al. Influence of smoking and plasma factors on femoropopliteal vein graft patency. 8M] . 1989;299:643-646. 25. Meade TW.lverson J, St irling Y. Effects of changes in smoking and other characteristics on clotting factors and the risks of ischemic heart disease. Lancet. 1987;ii:986-988. 26. Higman OJ. Srrachan AMJ. Buttery L, et al. Sm oking impairs the activity of endothelial nitric ox ide synthase in saphenous vein. Arterioscler Thromb Vase Bioi. 1996;16:546-552.
THE AMERICAN JOURNAL OF SURGERY. VOLUME 174 AUGUST 1997
221