Early Patency of in Situ Saphenous Vein Bypasses as Determined by Intraoperative Velocity Waveform Analysis

Early Patency of in Situ Saphenous Vein Bypasses as Determined by Intraoperative Velocity Waveform Analysis

Early Patency of In Situ Saphenous Vein Bypasses as Determined by Intraoperative Velocity Waveform Analysis David D. Schmitt, MD, Gary R. Seabrook, MD...

1MB Sizes 0 Downloads 17 Views

Early Patency of In Situ Saphenous Vein Bypasses as Determined by Intraoperative Velocity Waveform Analysis David D. Schmitt, MD, Gary R. Seabrook, MD, Dennis F. Bandyk, MD, Ruth F. Cato, RN, RVT, Janis W. Edwards, RN, Donna L. Karp BSN, RN, RVT, Judy L. Block, RN, Jonathan B. Towne, MD, Milwaukee, Wisconsin

Intraoperative velocity waveform analysis following in situ saphenous vein bypass grafting can identify abnormal hemodynamic conditions that correlate with the presence of a technical error or likelihood of perioperaUve thrombosis. Pulsed Doppler spectral analysis was used at operation to measure peak systolic blood flow velocity in the distal graft segment of 83 in situ saphenous vein bypasses to the poplReal (n = 35) or tibial (n = 48) arteries. Blood flow velocities were measured in the smallest diameter graft segment below the knee. Peak systolic blood flow velocity was greater than 40 cm/sec in 77 (93%) of grafts, and no early graft failures occurred. Low blood flow velocity (peak systolic blood flow velocity <40 cm/sec) was measured in six bypasses (7%) and was attributed to large (greater than 5 mm) vein diameter, residual hemodynamically significant lesions (intact valve leaflet, proximal arteriovenous fistula), or sclerosed vein segments. With the correction of these abnormalities, the 30 day patency for the entire series was 100%. The measurement of low blood flow velocity in the distal segment of an in situ saphenous vein bypass is an uncommon occurrence and mandates a thorough evaluation of the arterial reconstruction for correctable lesions. (Ann Vasc Surg 1990;4:270-275). KEY WORDS:

Saphenous vein bypass; intraoperative velocity waveform analysis.

tn situ saphenous vein bypass grafting has become a popular technique of lower limb revascularization, Its advantages include a tapered, converging conduit with close graft-artery size matching at anastomoses, intact vasa vasorum, and the ability to utilize veins less than 4 mm in diameter. Longterm patency rates exceeding 80% have been reFrom the Department of Surge~, The Medical College of Wisconsin, Milwaukee, Wisconsin. Presented at the Annual Meeting of the Peripheral Vascular Surgery. Society, New York, New York, June 17, 1989. Reprint requests: David D. Schmitt, MD, Department of Surgery, 8700 W. Wisconsin Avenue. Milwaukee, Wisconsin 53226.

ported [1]. Early graft patency is dependent on technical precision, since retained intact valve leaflets, missed arteriovenous fistulas, and injury to the endothelium during valve disruption may result in thrombosis. Upon completion of an arterial reconstruction, performance of hemodynamic assessment prior to leaving the operating room should lead to the identification of correctable lesions and prolonged primary patency. We have found pulsed Doppler spectral analysis of midstream flow to be a useful adjunct to arteriography for the intraoperative assessment of in situ saphenous vein grafts. This method has high sensitivity for the identification of undisrupted valve leaflets, persistent arteriovenous fistulas, and technical errors at anastomotic sites [2]. The purpose of 270

I N SITU S A P I I E N O U S VEIN B Y P A S S E S

VOLUME 4 N o 3 - 1990

271

this study was to determine the prognostic implications of intraoperative peak systolic blood flow velocity (Vp) determined by pulsed Doppler spectral analysis on graft patency following lower limb in situ saphenous vein bypass grafting.

PATIENTS AND METHODS Pulsed Doppler examination of midstream flow and arteriography were performed intraoperatively in 76 patients who underwent 83 in situ saphenous vein bypasses to the popliteal (n = 35) or tibial (n = 48) arteries. The study cohort included 51 men and 25 women whose ages ranged from 42 to 92 years (mean 64.5 years). The indications for bypass grafting were critical ischemia (n = 77, 92%), life-style limiting claudication (n = 3, 4%) and popliteal artery aneurysm (n = 3, 4%). All bypass grafts were constructed using the in situ technique previously described [3,4]. Briefly, following excision of the saphenous vein from the femoral vein, the proximal two or three vein valves were incised according to the technique of Leather and associates [5]. All proximal anastomoses were performed to the common femoral artery. Distal valves were incised with the Leather valvulotome inserted into vein side branches or through a puncture of the vein with an 18-gauge needle or through the transected end of the distal vein. Pulsed Doppler examination of midstream flow was used to detect abnormal flow patterns in the region of anastomotic and valve incision sites, to locate patent vein side branches, and to assess the outflow resistance of the runoff arteries. Completion arteriograms were obtained in all patients to verify technical adequacy of the distal anastomosis, to document valve leaflet ablation and the absence of arteriovenous fistulas, and to exclude the presence of thrombus and stricture in the outflow arteries. Instrumentation

Arterial flow pattern analysis was performed with a 20 mHz pulsed wave, direction sensitive Doppler velocity detector. This Doppler probe consisted of a small ultrasonic transducer mounted at the end of a 16-gauge needle*. The sample volume of the pulsed Doppler detector was approximately 0.2 mm 3 and could be positioned at any point between 1.1 and 11.5 mm from the end of the probe by range gating. A real-time, fast Fourier transform spectrum analyzert was used to process the back-scattered pulsed Doppler signal and display velocity spectra on a video monitor for interpretation and photographic recording. Spectral information was pre*CJ Hartley, PhD, Methodist Hospital, Houston, Texas. tModel D-10 Medasonics, Mountain View, California.

Fig. 1. Velocity spectra recorded at site of intact valve leaflet of in situ saphenous vein bypass graft. Increased frequency shifts recorded from disordered flow result in filling of spectral window.

sented graphically with frequency on the vertical axis, with time on the horizontal axis, and with amplitude indicated by intensity of the waveform.

Techniques

Pulsed Doppler flow pattern analysis and arteriography were performed following completion of the arterial reconstruction and restoration of blood flow. The gas-sterilized Doppler probe was placed directly on the surface of the vein graft and acoustically coupled to the vessel with sterile saline solution. The sample volume was located in the midstream of the flow by adjustment of the range control. An angle of approximately 60° was maintained between the Doppler probe and the longitudinal axis of the artery. The Doppler probe was first used to locate persistent venous side branches that result in arteriovenous fistulas. Beginning in the vein segment at the knee, the probe was placed on the vessel and the distal graft was occluded with digital pressure. The persistence of forward flow indicated a residual arteriovenous fistula distal to the probe. The entire graft was examined in this manner. Ligation of all arteriovenous fistulas should result in a staccatic, thumping Doppler signal in the graft with distal occlusion. The entire length of the graft was insonated with the Doppler probe to evaluate valveincision sites. Intact valve leaflets will result in disordered flow within the graft evidenced by increased frequency shifts resulting in spectral broadening (Fig. 1) and a decrease in the peak systolic frequency distal to the site. Similar changes in the velocity spectra will occur at anastomotic sites with technical defects.

1N SITU S A P H E N O U S VEIN B Y P A S S E S

272

ANNALS OF VASCULAR SURGERY

TABLE I.--intraoperative Vp and outcome of 83 in situ saphenous vein bypasses Distal graft Vp*

Number of grafts

Number not revised

>40 cm/sec <40 cm/sec

77 6

77 2

Number Number requiring thromrevision bosed 0 4

0 2

*Vp = peak systolic blood flow velocity

To evaluate the technical adequacy of the distal anastomosis and integrity of the runoff vessels, intraoperative arteriography was performed by direct needle puncture of the graft and injection by hand of 15 to 20 ml of contrast material with the proximal graft occluded. Single-exposure arteriograms were obtained with a portable roentgenographic unit. Following completion of the technical assessment of the in situ graft, peak systolic blood flow velocity was recorded in the below-knee graft segment of smallest diameter. Calculation of flow velocity requires a normal arterial flow pattern without turbulence, the accurate measurement of Doppler angle, and a vessel diameter that does not vary over short distances. Blood flow velocity was calculated from operator controlled cursor measurements of peaksystolic frequencies and the Doppler equation: Flow velocity = (C " Fs/2 • Fo cos (p~), where C is the speed of sound in tissue (1.54 x l 0 4 cm/sec); Fs is the measured frequency of the Doppler shifted signal; Fo is the beam frequency of the Doppler probe (20 x 106 Hz); and/z is the angle between the incident Doppler beam and the blood velocity vector (cos 60 ° = 0.5). Blood flow velocity was calculated under resting basal conditions approximately 20 minutes after restoration of flow.

RESULTS Peak systolic blood flow velocities (Vp) were recorded intraoperatively using pulsed Doppler spectral analysis in the distal below-knee graft segment of smallest diameter (Table I). Seventy-seven grafts (93%) had Vp greater than 40 cm/sec recorded in the distal graft segment (Fig. 2). All of these grafts had forward flow throughout the diastolic phase of the pulse cycle (hyperemic flow). No grafts in this group thrombosed or required revision in the early postoperative period (<30 days). Low blood flow velocity (Vp < 40 cm/sec) was observed in six grafts intraoperatively following completion of the revascularization (Table II). In two of the grafts the cause of the low Vp was attributed to large (>5 mm) vein graft diameter. Measurements from intraoperative arteriography of the first case documented the diameter of the vein

Fig. 2. Intraoperative spectral waveform of in situ saphenous vein bypass graft. Forward flow throughout pulse cycle (hyperemic waveform) is characteristic of reperfusion of vascular bed with low peripheral vascular resistance. Peak systolic blood flow velocity is 68 cm/sec.

graft to be 5.5 mm in the smallest below-knee segment, and the spectral waveform revealed a Vp of 33 cm/sec (Fig. 3). By duplex scan and spectral waveform analysis performed 34 months postoperatively this graft measured 7 mm in the below-knee segment with a Vp of 40 cm/sec (Fig. 4). The other vein graft with a conduit diameter of 6.5 mm also remains patent at 27 months without revision. The low Vp in four of the grafts could retrospectively be attributed to lesions that were not recognized at the initial operation by either the intraoperative arteriogram or Doppler interrogation of the graft. These four grafts all had diastolic flow with a characteristic hyperemic waveform for a revascularized extremity. In one graft (Vp of 26 cm/sec) an arteriovenous fistula which was felt to be insignificant was not ligated at the initial operation (Fig. 5). Two weeks postoperatively, the patient had persistent low graft flow velocities and underwent reoperation with ligation of the fistula. The graft Vp increased to 64 cm/sec. A second graft, with an intraoperative Vp of 37 cm/sec, occluded on the twelfth postoperative day. Subsequent review of the intraoperative arteriogram identified an intact valve leaflet. The patient underwent graft thrombectomy and disruption of the valve leaflet with an increase in graft Vp to 59 cm/sec (Fig. 6). Poor quality vein was felt to be the etiology of failure in two grafts whose intraoperative Vp were 21 and 33 cm/sec. In one patient the distal in situ vein graft was less than 2 mm for a length of 27 cm. This graft occluded 24 hours postoperatively. No evidence of intraoperative vein injury was noted at the time of revision; however, there were focal areas of intimal irregularity and adherent thrombus

VOLUME 4 NO 3 - 1990

I N S I T U S A P H E N O U S VEIN B YPASSES

273

TABLE II.--Outcome of six in situ saphenous vein grafts with Vp < 40 cm/sec Case 1 2

Intraoperative Vp* 33 cm/sec 37 cm/sec

Etiology Large diameter vein Large diameter vein

Outcome Graft patent at 34 months Graft patent at 27 months

26 cm/sec

Arteriovenous fistula

Reoperation, fistula ligated VP to 64 cm/sec Graft thrombosed Reoperation, valve leaflet disruption VP to 59 cm/sec Graft thrombosed first postoperative day Conversion to fem-pop isolated segment Graft patent at 48 months Replaced stenotic segment with reversed interposition vein graft Graft patent at 19 months

3

Graft type Femoropopliteal Femoroposterior tibial Femoropopliteal

4

Femoropopliteal

37 cm/sec

Persistent valve leaflet

5

Femoroposterior tibial

33 cm/sec

Poor vein quality

6

Femoropopliteal

21 cm/sec

Early vein graft stenosis, poor vein quality

*Vp = peak systolic blood flow velocity

within the graft. The bypass was salvaged by converting the anastomosis to an isolated femoropopliteal segment which remains patent at 48 months with a Vp of 62 cm/sec (Fig. 7). A critical stenosis of the other poor quality vein graft was identified by arteriography one week postoperatively. The luminal surface of the excised vein segment exhibited longitudinal furrows and calcification extending into the medial layer, suggesting the sequelae of previous thrombophlebitis. A reversed saphenous vein interposition graft was placed which remains patent at 19 months postoperatively (Fig. 8).

Fig. 3. lntraoperative arteriogram of femoral-popliteal in situ saphenous vein bypass graft utilizing conduit with varicosities. Large vein diameter resulted in low Vp.

DISCUSSION The intraoperative determination of in situ saphenous vein graft peak systolic blood flow velocity is

Fig. 4. Duplex scan of graft in Figure 3 obtained 34 months postoperative. Velocity spectra has triphasic waveform configuration with Vp of 40 cm/sec. Maximal cross-sectional diameter of vein graft was 13 mm.

274

I N S1TU S A P H E N O U S VEIN B Y P A S S E S

Fig. 5. Intraoperative arteriogram (a) of femoropopliteal in situ saphenous vein bypass graft demonstrates small arteriovenous fistula (arrow) initially thought to be insignificant. Vp in below-knee segment is 26 cm/sec. Arteriogram (b) 20 days postoperative demonstrates that fistula has enlarged (arrow), requiring Ugation.

an accurate predictor of early patency following lower extremity revascularization. All grafts in which intraoperative waveform analysis demonstrated a peak systolic blood flow velocity greater than 40 cm/sec remained patent without revision at 30 days. However, four of six grafts with a peak systolic blood flow velocity of less than 40 cm/sec

Fig. 6. Intraoperative arteriogram (a) of femoropopliteal in situ saphenous vein bypass graft with retained valve leaflet (arrow). Graft segment with intact valve leaflet magnified (b). Arteriogram on fourteenth )ostoperative day (c) shows no flow in in situ bypass arrow).

~

ANNALS OF

VASCULARSURGERY

Fig. 7. Intraoperative arteriogram (a) of femoroposterior tibial in situ saphenous vein bypass graft with low Vp (33 cm/sec) in distal graft with diameter less than 2 mm. Graft was converted to femoropopliteel bypass (b) following graft occlusion on first postoperative day.

thrombosed or required revision for technical errors in the early postoperative period. Although the routine use of intraoperative arteriography demonstrates unsuspected technical errors

Fig. 8. Intraoperative (a) femoropopliteal in situ saphenous vein bypass with residual distal graft stenosis (arrow). Reconstruction was revised with interposition saphenous vein graft (b).

VOLUME 4

No 3 - 1990

IN SITU SAPHENOUS VEIN BYPASSES

in 2 to 10% of in situ vein grafts [6-8], we believe that a roentgenographic study is enhanced with the adjunctive use of intraoperative waveform analysis. Pulsed Doppler spectral analysis has been shown to be a sensitive modality for the evaluation of anastomotic and valve incision sites for unsuspected technical error and for the location of patent vein side branches for ligation [2,9]. The use of the pulsed Doppler spectral analysis enables much of the graft evaluation to be performed before obtaining an arteriogram, resulting in decreased operative time and contrast load administered to the patient. Waveform analysis permits hemodynamic evaluation of abnormalities demonstrated by the operative arteriogram. We feel that waveform analysis and intraoperative arteriograms are complementary studies that permit a more critical intraoperative evaluation of the in situ saphenous vein bypass. Retrospective review of the intraoperative arteriograms in this series identified an etiology for the low blood flow velocity in the four grafts that required revision. The hemodynamics of in situ saphenous vein bypasses have been extensively studied. Bandyk and colleagues showed mean intraoperative peak systolic blood flow velocities for femoropopliteal and femorotibial grafts to be 81 _ 18 cm/sec and 74 -+ 21 cm/sec, respectively [9]. The four grafts that required revision in our series had intraoperative flow velocities two standard deviations below these means. Hyperemic blood flow (forward blood flow through both systole and diastole of the pulse cycle) is an important initial hemodynamic characteristic of a successful in situ bypass graft. Absence of hyperemic flow or a low peak systolic blood flow velocity has been found to be a predictor of early failure of in situ bypass grafts [10]. All four grafts in this study that required revision initially exhibited a hyperemic flow pattern on the intraoperative study. However, the presence of a low peak blood flow velocity may be a more sensitive indicator of a residual technical error than the presence of a hyperemic flow pattern.

mmm

275

CONCLUSIONS We found low blood flow velocity to be a predictor of the need for graft revision in the early postoperative period. The low peak systolic blood flow velocity led to the retrospective identification of lesions not found on completion arteriography and intraoperative Doppler assessment for residual technical errors. This study identified three etiologies for low peak systolic blood flow in the operating room: large diameter vein (>5 mm); residual technical error (intact valve leaflet, residual arteriovenous fistula); and poor quality vein conduit. The identification of low blood flow velocity in the distal graft segment, although an uncommon occurrence, requires an exhaustive search for correctable lesions to insure prolonged graft patency. REFERENCES 1. BANDYK DF, SCHMITT DD, SEABROOK GR, et al. Monitoring functional patency of in situ saphenous vein bypasses: the impact of a surveillance protocol and elective revision. J Vasc Surg 1989;9(2):286-296. 2. BANDYK DF, JORGENSEN RA, TOWNE JB. Intraoperative assessment of in situ saphenous vein arterial grafts using pulsed Doppler spectral analysis. Arch Surg 1986; 121:292- 299. 3. LEVINE AW, BANDYK DF, BONIER PH, TOWNE JB. Lessons learned in adopting the in situ saphenous vein bypass. J Vasc Surg 1985;1:145-153. 4. BANDYK DF, KAEBNICK HW, STEWART GW, TOWNE JB. Durability of in situ saphenous vein bypass: a comparison of primary and secondary patency. J Vasc Surg 1987 ;2: 256-268. 5. LEATHER RP, SHAH DM, KARMODY AM. Infrapopliteal arterial bypass for limb salvage: increased patency and utilization of the saphenous vein used "in situ." Surgery 1981 ;90:100(O1008. 6. BANDYK DF, ZIERLER RE, THIELE BL. Detection of technical error during arterial surgery by pulsed Doppler spectral analysis. Arch Surg 1984;119:421-428. 7. PLECHA FR, PORIES WJ. Intraoperative angiography in the immediate assessment of arterial reconstruction. Arch Surg 1972;105:907-909. 8. SIGEL B, COELHO JC, FLAN1GAN DP, et al. Detection of vascular defects during operation by imaging ultrasound. Ann Surg 1982;196:473-480. 9. BANDYK DF, KAEBNICK HW, BERGAMINI TM, et al. Hemodynamics of in situ saphenous vein arterial bypass. Arch Surg 1988;123:477-482. t0. BANDYK DF, CATO RF, TOWNE JB. A low flow velocity predicts failure of femoropopliteal and femorotibial bypass grafts. Surgery 1985;98:799--809.