Incidentally detected stenoses proximal to grafts originating below the common femoral artery: Do they affect graft patency or warrant repair in asymptomatic patients?☆☆☆
Article Outline
Abstract
Objective: Stenoses in infrainguinal arteries proximal to a lower extremity vein graft may reduce flow velocity through the bypass graft and are thought to predispose to graft occlusion. Repair of these lesions has been recommended to preserve graft function. This study was undertaken to better define the natural history of grafts below inflow lesions and to evaluate the necessity of repair to preserve graft patency. Methods: From 1994 through 1999, patients undergoing lower extremity vein grafts by a single surgeon at a university hospital and an affiliated teaching hospital were placed in a prospective protocol for proximal infrainguinal native artery and graft surveillance through use of duplex scanning. The records of those patients with grafts originating distal to the common femoral artery were evaluated; they form the basis for this report. Arteriograms were obtained before bypass grafting, and no patient had a stenosis greater than 50% diameter reduction proximal to the graft origin. Follow-up scans were obtained from the common femoral artery through the graft and outflow artery. The peak systolic velocity and velocity ratio in an infrainguinal native artery proximal to the graft origin were recorded, as were the location and the time interval since the bypass graft. Repair of these proximal lesions was not performed during the course of this study. Revision of the bypass graft or its anastomoses was undertaken according to preestablished duplex scan criteria. Results: During this time, 288 autogenous infrainguinal bypass grafts were performed, of which 159 originated below the common femoral artery; of these, 74 were from the superficial femoral artery, 29 from the profunda femoris artery, 49 from the popliteal artery, and 7 from a tibial artery. The maximum peak systolic velocity proximal to the graft origin was more than 250 in 38 arteries (25%) and more than 300 in 26 arteries (16%). The velocity ratio was 3.0 or more in 32 arteries at the same location as the peak systolic velocity and 3.5 or more in 23 arteries (15%), confirming hemodynamically significant stenoses at these sites. The location of peak systolic velocity was the common femoral artery in 81 patients (51%), the superficial femoral artery in 50 (31%), the popliteal artery in 22 (14%), and a tibial artery in 6 (4%). Follow-up ranged from 8 to 60 months (mean, 35 months). During follow-up, 19 patients died, 18 with patent grafts. Overall, nine grafts occluded. One of the occluded grafts had a velocity ratio greater than 3.0; this may have contributed to graft thrombosis. The other occlusions resulted from an unrepaired graft lesion in 2 patients, graft infection in 2 patients, and graft ligation necessitated by below-knee amputation in 2 patients. No cause for the occlusion could be identified in two of the grafts (neither had evidence of proximal arterial stenosis). Assisted primary patency rates were 95% and 91% at 3 and 5 years, respectively. Conclusions: For grafts originating distal to the common femoral artery, stenoses proximal to the graft do not affect bypass graft patency and do not require repair to prevent graft occlusion. Surveillance of these lesions may therefore be unnecessary, inasmuch as the repair of proximal lesions should not be undertaken to preserve graft function. (J Vasc Surg 2000;32:1180-9.)
Hemodynamically significant lesions in infrainguinal arteries proximal to the origin of lower extremity vein grafts are thought to predispose to graft occlusion. Identification and repair of these lesions have therefore been recommended to preserve graft function. However, the natural history of grafts below these lesions is not well defined, and no studies have routinely followed these grafts without intervention and evaluated graft patency.
Operative or endovascular treatment of proximal lesions, often with antecedent arteriography, is associated with significant cost and, in our experience, occasionally with significant morbidity. Iatrogenic arterial dissection or embolization may result in graft thrombosis, and infection may predispose to eventual graft rupture. In addition, routine surveillance of these proximal arteries with duplex scanning represents a time and cost burden to vascular laboratories evaluating significant numbers of grafts. Omission of surveillance and repair offers the potential to reduce the cost and morbidity of graft maintenance, but only if graft patency is not jeopardized. The current study was therefore undertaken to better define the natural history of grafts below infrainguinal native arteries that develop significant stenoses and to determine whether correction of these proximal arterial lesions can be safely omitted without affecting graft patency.
Methods and materials
From 1994 through 1999, patients undergoing infrainguinal bypass grafting through use of reversed saphenous vein at the University of Utah Medical Center and an affiliated teaching hospital staffed by the same surgeon were placed in a prospective protocol in which duplex scanning was used to evaluate the bypass graft and the proximal infrainguinal native arterial circulation. Preoperative evaluation included physical examination, ankle-brachial indices (ABIs) with arterial waveforms, and arteriography in all patients. Toe pressures were obtained in individuals with calcified or noncompressible arteries. Patients underwent arteriography before arterial reconstruction, and no patient had a stenosis proximal to the graft origin of greater than 50% diameter reduction on the basis of review of arteriograms by the attending vascular surgeon and a radiologist. Grafts were placed with the proximal anastomosis at the most distal site possible without a proximal lesion resulting in stenosis greater than 50%. This approach was used to minimize the length of requisite vein, minimize size discrepancy between the inflow artery and the reversed saphenous vein, and avoid dissection through scarred or previously dissected tissue. Most grafts were placed subcutaneously to allow for ease of surveillance and repair.
A completion duplex scan was obtained either intraoperatively or immediately postoperatively to ensure technical adequacy. These scans evaluated infrainguinal arteries and bypass grafts but did not routinely evaluate the aortoiliac segments. After surgery, patients underwent duplex scanning, physical examination, and ABIs every 3 months for the first 2 years, then every 6 months thereafter.
Postoperative duplex scans examined all infrainguinal arteries proximal to the graft origin, beginning with the common femoral artery and proceeding distally to include the bypass graft, both anastomoses, and the distal runoff artery. The iliac arteries were examined when common femoral waveforms suggested a hemodynamically significant iliac stenosis.
The duplex scan findings suggestive of possible iliac artery stenosis included a rounded upstroke on the waveform tracing and a clearly prolonged upstroke acceleration time. Each study determined the highest peak systolic velocity (PSV) and velocity ratio (VR) from a native infrainguinal artery proximal to the graft origin, as well as the location and time interval from the bypass graft to the highest PSV and VR. The VR was calculated by comparing the highest PSV in a native artery with the velocity in the same artery just proximal to the stenosis. In addition, the distal graft velocity and PSV within the bypass graft were recorded at each evaluation. The distal graft velocity was measured just proximal to the distal anastomosis because all grafts were reversed vein bypass grafts with the largest diameter segment placed distally. Appropriate angle correction was used in an attempt to ensure accurate velocities. Vein graft revision was performed for a VR greater than 3.0 involving the graft or its anastomoses. An elevated PSV without an elevated VR was not used as the basis for repair. Arteriograms were not routinely obtained before vein graft revision. In contrast to vein graft stenoses, stenoses in native infrainguinal arteries proximal to the graft origin, as determined by a PSV or VR elevated in comparison with the adjacent proximal arterial segment, were not treated to preserve graft patency. Therefore, these lesions were not routinely evaluated with arteriography or repaired in an attempt to prevent graft occlusion.
Patients with recurrent symptoms and individuals in whom no graft was present were managed in a similar manner. These few patients with recurrent symptoms and arterial insufficiency in the affected extremity despite a patent bypass were treated for limb preservation, not for graft salvage. Arterial insufficiency was determined by a reduction in ABI of more than 0.15 and either a change in distal arterial waveform configuration or a decrease in midgraft upstroke acceleration time. The presence of an infrainguinal inflow lesion without symptoms of arterial insufficiency was not evaluated with arteriography or treated. The proximal lesions were followed with serial duplex scans to better determine their natural history and effect on bypass graft patency.
Vascular laboratory technique
Studies were performed by a registered vascular technologist who used either an Acuson XP 128 ART (Acuson, Mountain View, Calif) or ATL 3000 (ATL, Bothell, Wash) scanner with a 7.5-MHz or 5.0-MHz probe for both the B-mode and the Doppler component. All studies were interpreted by a vascular surgeon with Registered Vascular Technologist (RVT) certification. Previously, we and others have validated the ability of duplex scanning to determine a hemodynamically significant arterial stenosis. At one institution, duplex scan evaluation of femoral arteries has been correlated with arteriography. At both institutions, selected patients undergoing both arteriography and duplex scanning have had the findings reviewed with good correlation, and no patient has been identified with a duplex scan–identified stenosis that was not confirmed by arteriography or at operation. Correlation of duplex scan velocity criteria with arteriography has been performed as part of Intersocietal Commission for the Accreditation of Vascular Laboratories (ICAVL) accreditation, although this has not been specifically done for proximal infrainguinal native arteries.
Data analysis
The indication for the initial procedure, details of the initial operation, date and location of any revisions, patient cardiovascular risk factors and demographic information, graft and native artery hemodynamic parameters, and outcome were prospectively recorded and entered into a computerized database (Filemaker Pro 4.0, Microsoft, Redmond, Wash). Statistical analysis to determine the relationship between inflow stenosis, as determined by PSV and VR, and subsequent graft patency and limb salvage was performed through use of an independent t test. Correlation of discrete cutoff levels for PSV (> 200, > 250, > 300 cm/s) and VR (> 2.0, > 2.5, > 3.0) with success or failure was determined through use of the Fisher t test. Results were reported in accordance with the suggested reporting standards for lower extremity arterial procedures as defined by the Society for Vascular Surgery and the North American Chapter of the International Society for Cardiovascular Surgery.1 The life table method was used to analyze graft patency. Each affected lower extremity was analyzed separately. The assisted primary patency reflected bypass grafts that have remained patent after one or more revisions.
Demographic information
From 1994 through 1999, 288 patients underwent infrainguinal bypass grafting. One hundred fifty-nine of these grafts were composed of reversed saphenous vein and originated distal to the common femoral artery; they form the basis of this report. The 159 patients (125 male, 34 female) ranged in age from 32 to 91 years (mean, 61 years). Cardiovascular risk factors included a history of cigarette use in 122 patients (77%), hypertension (defined as requiring medication for blood pressure control) in 84 (53%), symptomatic cardiac disease in 71 (45%), diabetes in 69 (43%), renal insufficiency (defined as a creatinine level > 1.5) in 66 (42%), and hyperlipidemia in 38 (24%). Forty-seven patients had undergone a previous arterial reconstruction, including a bypass graft on the contralateral limb (19), an inflow procedure above the inguinal ligament of the operated lower extremity (14), or a previous ipsilateral infrainguinal bypass graft (14). No patient in this study underwent a previous iliac or femoral endarterectomy. Patients undergoing prior aortoiliac reconstructions were included in this review, inasmuch as our study focused on infrainguinal native arterial stenoses.
Initial operation
The indication for the initial operation was gangrene or a nonhealing wound in 85 patients (53%), ischemic rest pain in 40 (25%), disabling intermittent claudication in 28 (18%), and a popliteal aneurysm in 6 (4%). The inflow artery was the superficial femoral artery (SFA) in 74 patients (47%), the profunda femoris artery in 29 (18%), the popliteal artery in 49 (31%), and a tibial or peroneal artery in 7 (4%). The outflow artery was the popliteal in 85 (53%), a tibial in 51 (32%), and a pedal in 23 (15%). The mean preoperative ABI was 0.49, although 12 patients had calcified tibial arteries that were associated with artificially elevated ABIs. The 12 patients with artificially elevated or inaccurate ABIs did not have their ABI values included in the analysis of ABI data.
Results
Hemodynamic parameters
In the 159 patients, the value for the highest PSV identified postoperatively in an infrainguinal artery proximal to the graft, as determined by sequential duplex scan measurements, ranged from 21 to 516 cm/s; the highest PSV was greater than 250 cm/s in 38 grafts (25%) and greater than 300 cm/s in 26 grafts (16%). No threshold value of PSV was used to determine that proximal revision was to be performed. One hundred fifty-one patients (94%) had the highest PSV and VR in a proximal artery within 12 months of graft placement (Fig 1); only eight patients had the maximum PSV more than 12 months after bypass grafting.
Fig. 1.
Interval to PSV from bypass graft (in months).
The number and percent of patients with an elevated VR for each PSV are listed in Table I.
Table I. PSV and VR in proximal artery: 159 grafts
| PSV* (cm/s) | No. of grafts (%) | VR | No. of grafts |
|---|---|---|---|
| > 200 | 43 (27) | > 2.0 | 43 |
| > 2.5 | 38 | ||
| > 3.0 | 30 | ||
| > 3.5 | 21 | ||
| > 250 | 38 (24) | > 2.5 | 28 |
| > 3.0 | 24 | ||
| > 3.5 | 21 | ||
| > 300 | 26 (16) | > 3.0 | 16 |
| > 3.5 | 16 | ||
| *Range, 21-516. | |||
Each of 16 grafts (10% of the study grafts) had a VR greater than 3.0 with a PSV greater than 300 cm/s. In all cases, these values were not present on the initial postoperative duplex scans. Five lesions were first detected at 3 months, 5 at 6 months, 3 at 9 months, 2 at 12 months, and 1 at 15 months, which suggests that these represent stenoses that developed rather than residual lesions. The progression of proximal stenosis in these patients was not associated with a change in distal graft velocity or graft velocity waveform. The distal graft velocity values in these patients ranged from 31 to 158 cm/s. Seven lesions occurred in the SFA, 7 in the common femoral artery, and 2 in the above-knee popliteal artery. Twelve (75%) of the 16 showed progression on serial scans; nine of these patients eventually underwent revision of the graft for a separate lesion within the bypass graft.
The location of the maximum PSV and VR was the common femoral artery in 81 patients, the SFA in 50, the popliteal artery in 22, and a tibial artery in 6. Most velocity elevations were in the common femoral artery, which reflects the evaluation of this vessel in all patients. In patients with grafts originating from the proximal SFA or deep femoral artery, the common femoral was usually the only inflow artery evaluated above the proximal anastomosis. For grafts originating from the popliteal or a tibial artery, the SFA was the more frequent site of PSV and VR elevation; this is a reflection of its smaller diameter, greater disposition for atherosclerotic disease, and greater length in comparison with the common femoral artery.
Follow-up
Follow-up for the 159 patients ranged from 8 to 60 months (mean, 35 months). The mean duration of follow-up after detection of the highest PSV and VR was 28 months. One hundred forty patients are alive; 19 patients died, 18 with patent grafts. Thirty-five (20%) of the 159 patients underwent revision of the vein graft during the course of this study. The revision performed was a patch angioplasty in 27 cases and an interposition graft in eight cases. The length of the interval from the initial operation to the revision was 3 months in 5 patients, 6 months in 7, 7 months in 4, 8 months in 3, 9 months in 5, 12 months in 8, and 15 months in 3. None of these patients had creation of a new proximal anastomotic site. Nine of these patients had a PSV greater than 250 and a VR greater than 3.0 (severe stenosis). Revision of an intrinsic graft stenosis was associated with an improvement in distal graft velocity (mean increase, 43 cm/s) and ABI (0.33). In accordance with the protocol, no patient underwent repair of a proximal native arterial stenosis. Of the 35 patients undergoing graft repair, 20 had PSVs greater than 150 in the proximal infrainguinal inflow artery. The proximal stenosis was not revised in any of these patients. All 20 of these grafts remained patent during follow-up.
During follow-up, nine grafts (6%) occluded. Five grafts that occluded had no hemodynamic evidence of an inflow stenosis, as determined by PSV or VR, before or at the time of graft thrombosis. Four of the nine grafts had an elevated PSV greater than 200 cm/s or a VR greater than 2.0 in an artery proximal to the bypass graft identified before graft thrombosis. One of these four grafts, with a PSV of 201 cm/s and a VR of 2.2, occluded and was excised because of graft infection. The other three grafts occluding with an inflow stenosis had PSVs of 106, 250, and 316 cm/s, with VRs of 1.8, 3.0, and 4.0, respectively.
The etiology for graft occlusion in 6 of these 9 patients was as follows: graft infection in 2 patients; unrepaired midgraft lesion as determined by elevated VR greater than 3.0 in 2 patients; graft ligation necessitated by below-knee amputation to control foot infection despite a patent bypass in 2 patients. Two patients had normal hemodynamic parameters in both the inflow artery and the graft. In one patient, inflow stenosis may have contributed to graft thrombosis, inasmuch as the PSV and VR were elevated in the native artery; this patient had no reduction in ABI, no recurrent symptoms, and no reduction in distal graft velocity to indicate that the graft was at risk for occluding.
For all 159 patients, the assisted primary patencies were 98%, 95%, and 91% at 1, 3, and 5 years, respectively (Table II and Fig 2).
Table II. Life table data analysis for assisted primary patency in 159 grafts originating distal to common femoral artery
| Interval (mo) | No. of grafts at risk at start | Failed | Withdrawn patent | Interval patency rate | Cumulative patency (%) | SE | |
|---|---|---|---|---|---|---|---|
| Because of duration | Because of death | ||||||
 0-1 | 159 | 1 | 0 | 0 | .993 | 100 | 0.00 |
| 1-2 | 158 | 0 | 0 | 1 | 1.0 | 99.3 | 0.66 |
| 3-5 | 157 | 1 | 0 | 5 | .994 | 99.3 | 0.66 |
| 6-8 | 151 | 0 | 0 | 0 | 1.0 | 98.7 | 0.89 |
| 9-11 | 151 | 1 | 1 | 1 | .993 | 98.7 | 0.89 |
| 12-14 | 148 | 0 | 0 | 1 | 1.0 | 98.0 | 1.10 |
| 15-17 | 147 | 0 | 4 | 2 | 1.0 | 98.0 | 1.10 |
| 18-20 | 141 | 2 | 10 | 2 | .985 | 98.0 | 1.10 |
| 21-23 | 127 | 1 | 10 | 0 | .992 | 96.5 | 1.42 |
| 24-26 | 116 | 1 | 10 | 2 | .991 | 95.8 | 1.56 |
| 27-29 | 103 | 0 | 12 | 0 | 1.0 | 94.9 | 1.70 |
| 30-32 | 91 | 0 | 8 | 2 | 1.0 | 94.9 | 1.70 |
| 33-35 | 81 | 0 | 6 | 1 | 1.0 | 94.9 | 1.70 |
| 36-38 | 74 | 0 | 7 | 0 | 1.0 | 94.9 | 1.70 |
| 39-41 | 67 | 1 | 11 | 1 | .984 | 94.9 | 1.70 |
| 42-44 | 54 | 1 | 13 | 0 | .980 | 93.4 | 1.90 |
| 45-47 | 40 | 0 | 10 | 0 | 1.0 | 91.5 | 2.11 |
| 48-50 | 30 | 0 | 6 | 0 | 1.0 | 91.5 | 2.11 |
| 51-53 | 24 | 0 | 5 | 0 | 1.0 | 91.5 | 2.11 |
| 54-56 | 19 | 0 | 10 | 0 | 1.0 | 91.5 | 2.11 |
| 57-60 | 9 | 0 | 9 | 0 | 1.0 | 91.5 | 2.11 |
Fig. 2.
Life table plot of assisted primary patency.
Table III. Correlation of graft occlusion with different levels of PSV and VR
| No. of grafts | |||
|---|---|---|---|
| Patent | Occluded | P value | |
| PSV | |||
| ≥ 250 | 28 | 1 | .4066 |
| ≥ 300 | 17 | 1 | .3955 |
| VR | |||
| ≥ 2.5 | 43 | 2 | .3257 |
| ≥ 3.0 | 30 | 2 | .2842 |
Patients were then stratified on the basis of the presence of an infrainguinal proximal arterial stenosis. Graft patency was calculated separately for the 50 patients with PSVs greater than 150 cm/s and VRs greater than 2.0 (proximal stenosis) and for the other 109 patients without proximal arterial stenosis. By life table analysis, assisted primary patencies were found to be 100%, 95%, and 92% at 1, 3, and 5 years for patients with stenosis and 99%, 93%, and 90% at the same intervals for patients without stenosis (Tables IV and V).
Table IV. Life table data analysis for assisted primary patency in 50 grafts originating distal to common femoral artery with proximal arterial stenosis (PSV > 150 cm/s and VR > 2.0)
| Interval (mo) | No. of grafts at risk at start | Failed | Withdrawn patent | Interval patency rate | Cumulative patency (%) | |
|---|---|---|---|---|---|---|
| Because of duration | Because of death | |||||
| 0-1 | 50 | 0 | 0 | 0 | 1.0 | 1.0 |
| 1-2 | 50 | 0 | 0 | 0 | 1.0 | 1.0 |
| 3-4 | 50 | 0 | 0 | 0 | 1.0 | 1.0 |
| 4-6 | 50 | 0 | 0 | 0 | 1.0 | 1.0 |
| 7-9 | 50 | 0 | 0 | 0 | 1.0 | 1.0 |
| 10-12 | 50 | 0 | 0 | 0 | 1.0 | 1.0 |
| 13-15 | 50 | 0 | 0 | 0 | 1.0 | 1.0 |
| 16-18 | 50 | 0 | 0 | 1 | 1.0 | 1.0 |
| 19-21 | 49 | 0 | 2 | 1 | 1.0 | 1.0 |
| 22-24 | 46 | 1 | 2 | 0 | .978 | .978 |
| 25-27 | 43 | 0 | 2 | 0 | 1.0 | .978 |
| 28-30 | 41 | 1 | 0 | 0 | .976 | .954 |
| 31-33 | 40 | 0 | 1 | 0 | 1.0 | .954 |
| 34-36 | 39 | 0 | 4 | 0 | 1.0 | .954 |
| 37-39 | 35 | 1 | 4 | 1 | .970 | .925 |
| 40-42 | 29 | 0 | 5 | 0 | 1.0 | .925 |
| 43-45 | 24 | 0 | 3 | 0 | 1.0 | .925 |
| 46-48 | 21 | 0 | 7 | 0 | 1.0 | .925 |
| 49-52 | 14 | 0 | 2 | 0 | 1.0 | .925 |
| 53-55 | 12 | 0 | 4 | 0 | 1.0 | .925 |
| 56-60 | 8 | 0 | 8 | 0 | 1.0 | .925 |
Table V. Life table data analysis for assisted primary patency in 109 grafts originating distal to common femoral artery with no proximal arterial stenosis
| Interval (mo) | No. of grafts at risk at start | Failed | Withdrawn patent | Interval patency rate | Cumulative patency rate (%) | |
|---|---|---|---|---|---|---|
| Because of duration | Because of death | |||||
| 0-1 | 109 | 0 | 0 | 0 | 1.0 | 1.0 |
| 1-3 | 109 | 0 | 0 | 0 | 1.0 | 1.0 |
| 4-6 | 109 | 0 | 0 | 0 | 1.0 | 1.0 |
| 7-9 | 109 | 1 | 0 | 6 | .991 | .991 |
| 10-12 | 102 | 0 | 1 | 1 | 1.0 | .991 |
| 13-15 | 100 | 0 | 1 | 2 | 1.0 | .991 |
| 16-18 | 97 | 1 | 6 | 0 | .990 | .981 |
| 19-21 | 90 | 1 | 8 | 1 | .988 | .970 |
| 22-24 | 80 | 1 | 8 | 2 | .987 | .957 |
| 25-27 | 69 | 0 | 9 | 0 | 1.0 | .957 |
| 28-30 | 60 | 0 | 11 | 3 | 1.0 | .957 |
| 31-33 | 46 | 0 | 6 | 0 | 1.0 | .957 |
| 34-36 | 40 | 1 | 1 | 0 | .975 | .933 |
| 37-39 | 38 | 0 | 8 | 0 | 1.0 | .933 |
| 40-42 | 30 | 1 | 4 | 0 | .964 | .900 |
| 43-45 | 25 | 0 | 9 | 0 | 1.0 | .900 |
| 46-48 | 16 | 0 | 4 | 0 | 1.0 | .900 |
| 49-51 | 12 | 0 | 1 | 0 | 1.0 | .900 |
| 52-54 | 11 | 0 | 4 | 0 | 1.0 | .900 |
| 55-57 | 7 | 0 | 3 | 0 | 1.0 | .900 |
| 58-60 | 4 | 0 | 4 | 0 | 1.0 | .900 |
Discussion
Stenoses in infrainguinal arteries proximal to grafts originating below the common femoral artery are frequently identified on the basis of serial duplex scan examinations. In our study, each of 30 proximal arteries (19% of extremities) had a PSV greater than 200 cm/s and a VR greater than 3.0, some lesions being greater than 80% to 90%.
Calligaro et al2 noted that 28% of grafts (19 patients) developed inflow lesions, whereas a University of Oregon study found that 18% of grafts (37 of 205 patients) developed inflow stenoses.3 However, Mills et al4 found that only 5% of grafts developed inflow lesions; this is similar to the findings of Mattos et al,5 who reported a 3.6% incidence, and other investigators.6, 7, 8 Criteria for identification of an inflow lesion, however, have often been unspecified or differed between reports. In addition, the number of proximal lesions invariably increases with longer follow-up.
Several reports have found that stenoses in inflow arteries proximal to infrainguinal vein grafts rarely result in graft thrombosis. Most occlusions result from lesions within the bypass graft or its anastomoses.5, 9, 10 At the University of Arizona, Westerband et al9 prospectively evaluated 101 grafts and found that no inflow lesions resulted in graft occlusion; several grafts remained patent from collateral blood flow despite complete proximal inflow artery or bypass occlusion. Taylor et al11 reported on 450 grafts; they too found no occlusions resulting from inflow lesions, the mean follow-up being 22 months (range, 1-22 months). Donaldson et al12 reported on 455 grafts with a mean follow-up of 18 months, defining graft failure as stenosis greater than 50%; only two of 92 failures resulted from inflow lesions. Grigg et al,13 reporting on 75 grafts, noted no graft occlusions resulting from extrinsic lesions. Caps et al14 and Mattos et al,5 in studies of infrainguinal vein grafts with duplex scanning, found that no occlusions resulted from inflow disease. Dougherty et al,15 in a study not limited to infrainguinal vein grafts, also found that progression of proximal disease did not result in any graft occlusions.
Smith concluded that inflow stenosis does not invariably lead to graft occlusion and used the term “pseudo-occlusion” to define a graft that is thought to be occluded but in fact is patent below an inflow occlusion.11 Fourteen grafts had proximal lesions identified on arteriography; all remained patent.11 Idu et al16 reported only one graft occlusion resulting from inflow obstruction, and Mattos,5 in an early report on 170 limbs, found that proximal lesions did not result in any graft occlusions.
These studies and our own experience prompt us to question the traditional view that a critical blood flow velocity in the inflow artery is necessary for graft patency. Whether there is an absolute threshold for blood flow velocity cannot be determined. Grafts originating from tibial or even pedal arteries, often with very low velocity in the inflow artery, have patency equivalent to grafts from more proximal vessels. Large-diameter grafts and grafts to distal arteries frequently have PSVs below 20 or even 15 cm/s, yet they usually remain patent.17, 18 In addition, our results are supported by other studies on graft revision, which have found that low global graft velocity (or distal graft velocity) is not a reliable indicator of impending graft occlusion and should not necessarily be the basis for graft revision.12, 19
Almost all of the studies described in the preceding paragraphs as evaluating the effect of proximal arterial stenosis on patency reviewed grafts that had thrombosed and then attempted to determine the etiology for graft occlusion. No study has prospectively identified proximal stenoses and followed them without intervention; the explanation for this is the traditional view that hemodynamically significant inflow lesions, by reducing blood flow velocity through the graft, will lead to graft occlusion. However, the long-term natural history of these lesions is unknown, because most investigators routinely repaired stenoses of greater than 50%8 or 70%.9 In addition, most of these lesions are in iliac rather than femoral, popliteal, or tibial arteries; whether lesions in these vessels represent the same or an increased risk has not been reported.
Because of the high patency of infrainguinal grafts associated with a policy of routine repair of proximal arterial stenoses, few institutions have been willing to undertake noninterventional treatment of these lesions. Taylor et al11 repaired 13 of the 16 inflow stenoses or occlusions after lesion identification by arteriography and reported excellent graft patency. Westerband et al9 noted that three inflow lesions were associated with an ABI reduction of greater than 0.15; all three lesions were repaired. In that series, 96 of 101 grafts remained patent during follow-up. Nehler et al20 have aggressively treated inflow lesions at the University of Oregon and have obtained greater than 90% 5-year vein graft patency.
Interestingly, despite concerns about progressive atherosclerosis in native arteries proximal to grafts originating from the SFA, the popliteal artery, or tibial arteries, the patency of distal origin grafts has been found to be equivalent to that of bypass grafts from the common femoral artery.21 In this study, the 91% 3-year patency for grafts originating distal to the common femoral is comparable to that for grafts originating more proximally, and these results are similar to reports from several other institutions.21 This suggests that the effect of these proximal arterial lesions is usually not significant and that efforts to identify these stenoses with either duplex scanning or arteriography may not be required. In addition, hemodynamically significant inflow lesions will often result in recurrent symptoms,9, 14 which should usually prompt diagnostic arteriography and treatment independently of concerns about distal bypass graft preservation. It is not surprising, given the high patency obtained with careful follow-up and revision of both proximal and graft lesions, that almost all reports advocate elective repair when these lesions are identified. Some reports recommend repair for all lesions greater than 50%,3 whereas others do not specify a specific threshold for revision.
On the basis of our experience and the reports of others, we note that routine evaluation and repair of these lesions may not be warranted. However, the development of recurrent symptoms, especially in the presence of a reduction in graft pulse or ABI, suggests a not only hemodynamically but also clinically significant lesion; such a lesion should be evaluated with arteriography and repaired if indicated for treatment of disabling claudication or limb-threatening ischemia. Our study suggests, however, that prophylactic repair of these lesions, in the absence of clinical symptoms and based solely on a reduction in global graft velocity (distal graft velocity), may not be necessary to prevent graft thrombosis.
Our study does not provide follow-up beyond 5 years, and many patients have not been followed up for more than 2 to 3 years. Longer follow-up will be necessary to further define the natural history of these lesions and confirm the safety of this approach.19 This study represents only intermediate-term data on graft patency and does not provide long-term information on the potential effect of inflow stenoses. However, if hemodynamically significant lesions, often present for more than 12 to 24 months, do not lead to graft occlusion, it is unclear that lesions present for a longer period of time will lead to graft thrombosis.
As an additional benefit, if repair of proximal lesions is not undertaken, then identification of these stenoses is unnecessary and evaluation of the proximal arterial circulation with duplex scanning may be unnecessary. This offers a potential significant cost benefit to institutions and vascular laboratories that scan significant numbers of grafts. In our experience, routine examination of the ipsilateral lower extremity for subcutaneously placed grafts requires approximately 45 to 60 minutes, depending on the location of the bypass graft and the size of the patient’s lower extremity. For grafts originating from the popliteal artery or a tibial artery, examination of the common femoral artery and SFA adds 20 to 30 minutes, which almost doubles the time and cost of the scan for the laboratory.
Because our laboratory now performs approximately 30 surveillance duplex scans each month, evaluation of the proximal arteries requires 9 to 10 hours of additional technologist time and often results in significant time delays in the vascular laboratory for other patients waiting to be examined. The time, and thus the cost, of the added surveillance varies depending on the technical difficulty of the study, extent of disease, and need for physician review. At a minimum, the additional cost to the vascular laboratory is $10 to $15 per study (> $2000 per month), the expense being even higher if the technologist is required to stay overtime or if other patients experience significant delays. There is intermittently significant cost to the institution if another study cannot be undertaken (because of equipment or personnel limitations) and hospital discharge or surgery therefore needs to be postponed or significantly delayed.
Conclusion
This study finds that stenoses developing in infrainguinal arteries proximal to distal origin vein grafts do not lead to bypass graft occlusion. Noninterventional management of these lesions does not jeopardize graft patency. Repair of these lesions, though associated with excellent graft patency, does not necessarily provide additional benefit. A policy of noninterventional treatment provides the additional significant benefit of potentially eliminating both diagnostic or prerevision arteriography and routine duplex scan surveillance of these vessels. This, in turn, provides significant benefits to vascular laboratories and institutions repairing large numbers of grafts. In addition, eliminating operative repair will reduce the cost and morbidity to the patients undergoing these procedures.
We caution that this report represents only an intermediate-term study and that our recommendations cannot be justified if the policy ultimately leads to a higher risk of graft occlusion. Nevertheless, our positive experience should prompt other prospective trials of noninterventional management and should form the basis for randomized studies comparing graft patency both with and without repair of proximal arterial lesions.
References
- Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg. 1986;4:80–94
- Selective use of duplex ultrasound to replace preoperative arteriography for failing arterial vein grafts. J Vasc Surg. 1998;27:89–95
- Duplex scanning alone is not sufficient imaging before secondary procedures after lower extremity reversed vein bypass graft. J Vasc Surg. 1999;29:270–281
- . The importance of routine surveillance of distal bypass grafts with duplex scanning: a study of 379 reversed vein grafts. J Vasc Surg. 1990;12:379–389
- . Does correction of stenosis identified with color duplex scanning improve infrainguinal graft patency?. J Vasc Surg. 1993;17:54–66
- . Impact of a color-flow duplex surveillance program on infrainguinal vein graft patency: a five-year experience. J Vasc Surg. 1993;17:42–53
- . Intraoperative color duplex scanning for infrainguinal vein grafts. Ann Vasc Surg. 1996;10:347–355
- . Long-term surveillance by duplex scanning of nonrevised infragenicular graft stenosis. Ann Vasc Surg. 1995;9:547–553
- . Prospective validation of threshold criteria for intervention in infrainguinal vein grafts undergoing duplex surveillance. Ann Vasc Surg. 1997;11:44–48
- . Natural history of infrainguinal vein graft stenosis relative to bypass grafting technique. J Vasc Surg. 1997;25:211–225
- . Does arterial inflow failure cause distal vein graft thrombosis?. Ann Vasc Surg. 1994;8:92–98
- . Causes of primary graft failure after in situ saphenous vein bypass grafting. J Vasc Surg. 1992;15:113–120
- . Detection and grading of femorodistal vein graft stenosis: duplex velocity measurements compared with angiography. J Vasc Surg. 1988;8:661–666
- . Vein graft lesions: time of onset and rate of progression. J Vasc Surg. 1995;22:466–475
- . The natural history of “failing” arterial bypass grafts in a duplex surveillance protocol. Ann Vasc Surg. 1998;12:255–259
- . Vein graft surveillance: is graft revision without angiography justified and what criteria should be used?. J Vasc Surg. 1998;27:399–413
- . Effect of outflow level and maximum graft diameter on the velocity parameters of reversed vein bypass grafts. J Vasc Surg. 1999;30:16–25
- . The variation in vein graft flow velocity with luminal diameter and outflow level. J Vasc Surg. 1992;15:991–999
- . Ongoing vascular laboratory surveillance is essential to maximize long-term in situ saphenous vein bypass patency. J Vasc Surg. 1996;23:18–27
- . Surgical treatment of threatened reversed infrainguinal vein grafts. J Vasc Surg. 1994;20:558–565
- . The popliteal artery as inflow for distal bypass grafting. Arch Surg. 1994;129:211–217
☆ Competition of interest: nil.
☆☆ Reprint requests: Gerald Treiman, MD, Vascular Surgery, Salt Lake City, VAMC, 500 Foothill Blvd, Salt Lake City, UT 84148.
PII: S0741-5214(00)06589-7
doi:10.1067/mva.2000.109770
© 2000 Society for Vascular Surgery and The American Association for Vascular Surgery, a Chapter of the International Society for Cardiovascular Surgery. Published by Elsevier Inc. All rights reserved.
