- Email: [email protected]
Upper Extremity Hemodynamic Changes after Radial Artery Harvest for Coronary Artery Bypass Grafting
Upper Extremity Hemodynamic Changes after Radial Artery Harvest for Coronary Artery Bypass Grafting Joann M. Lohr, MD, RVT, Douglas S. Paget, MD, J. Michael Smith, MD, Jennifer L. Winkler, BS, RVT, and Alan R. Wladis, MD, Cincinnati, Ohio
Twenty-seven patients were studied with arterial duplex, photoplethysmography, segmental pressures, and pulse volume recordings both preoperatively and following radial artery harvesting. The average number of days to the follow-up visit was 66. Preoperative and postoperative data were compared using the matched Student’s t-test. There were no significant changes between preoperative and postoperative pressures in the brachial, radial, and ulnar arteries, and thumb, index, long, ring, or little fingers. Pressure changes in the thumb and index finger approached but did not achieve a statistical difference. Peak systolic velocity (PSV), end diastolic velocity (EDV), and resistive index (RI) in the distal ulnar artery changed significantly between preoperative and postoperative measurements. PSV changed from 0.50 ± 0.05 m/sec to 0.67 ± 0.04 m/sec (p = 0.02); EDV changed from 0.03 ± 0.03 m/sec to −0.10 ± 0.05 m/sec (p = 0.05); and RI changed from 0.97 ± 0.05 to 1.13 ± 0.05 (p = 0.02). Palmar arch evaluations revealed significant changes at rest and with ulnar compression between preoperative and postoperative measurements: (1) at rest EDV changed from 0.03 ± 0.02 m/sec to −0.05 ± 0.02 m/sec (p < 0.01); (2) at rest RI changed from 0.96 ± 0.05 to 1.12 ± 0.05 (p = 0.01); (3) with ulnar compression the PSV changed from 0.23 ± 0.05 m/sec to 0.005 ± 0.01 m/sec (p < 0.01); and (4) with ulnar compression the RI changed from 0.82 ± 0.11 to 0.27 ± 0.12 (p < 0.01). Eight patients had a variety of complaints at the follow-up visit, the majority being numbness and tingling. No patients reported symptoms of claudication or rest pain at the follow-up visit. The data suggest that while statistically significant changes in velocity and arterial resistance do occur, patients seem to tolerate radial artery harvesting without clinical consequences. The ideal method of preoperative evaluation remains to be determined. (Ann Vasc Surg 2000;14:56-62). DOI: 10.1007/s100169910010
INTRODUCTION Carpentier et al. first suggested the use of the radial artery as an arterial conduit for coronary artery bypass grafting in 1973.1 These initial results were dis-
From the John J. Cranley Vascular Laboratory, Good Samaritan Hospital, Cincinnati, OH. Presented at the Twenty-fourth Annual Meeting of the Peripheral Vascular Surgery Society, Washington, DC, June 5-6, 1999. Correspondence to: J.M. Lohr, MD, RVT, c/o Kim Hasselfeld, Research 11J, Good Samaritan Hospital, 375 Dixmyth Avenue, Cincinnati, OH 45220-2489, USA. 56
appointing and the technique was initially abandoned. The introduction of calcium channel blockers and antispasmodic agents has renewed interest in radial artery harvest for use as a conduit for coronary artery bypass grafting.2-5 Radial artery harvest techniques have also been changed to decrease artery handling.6 Clinical contraindications to radial artery harvest include a history of arterial trauma to the involved extremity, Raynaud’s syndrome, and ischemic syndromes, including claudication, rest pain, cold sensitivity, or history of digital ulceration. The upper extremity anatomy is complex and quite varied. A continuous superficial palmar arch (SPA) and digital arteries supplied primarily by the ulnar
Vol. 14, No. 1, 2000
artery is the most common configuration. However, the SPA may not always be complete or may not be based on the ulnar artery. A harvested radial artery in either of these last two situations can be disastrous and can lead to ischemic complications. A reliable method for the assessment of the SPA and the blood flow of the upper extremity prior to radial artery harvest has yet to be defined. The obvious concern is to avoid ischemic complications of the hand. Cardiothoracic surgeons have been extremely conservative as to which vessels are harvested. This study was an attempt to identify arterial dominance into the hand, provide information on the sufficiency of arterial flow and compensatory flow after radial arterial removal, and assess the potential for the development of ischemic complications.
MATERIALS AND METHODS Twenty-seven patients were evaluated with preoperative arterial duplex, photoplethysmography (PPG), segmental pressures, and pulse volume recordings (PVRs) prior to undergoing radial artery harvest during elective coronary artery bypass grafting and in the postoperative period. Data were analyzed using BMDP Statistical Software (BMDP Statistical Software Inc., Los Angeles, CA). Parametric statistical analysis was done with the Student’s t-test, and nonparametric comparisons were done with the Fisher’s exact test. A p-value of <0.05 was taken to represent a significant difference. Patients were referred for radial artery mapping (RAM) in their nondominant arm. Tests were performed in a temperature-controlled room at 70°F. Bilateral or dominant arm RAMs were performed at the request of the cardiovascular surgeon. RAM consisted of three phrases of testing: Doppler measurement of the upper extremity segmental artery pressures and digital pressures, Allen testing with PPG of all appendages, and arterial duplex imaging of the arm of interest including the SPA with and without radial artery compression as described previously by Winkler et al.7 Doppler measurement of segmental arterial pressures was performed bilaterally, with pressures and PVRs performed at the midhumeral region, where pressures were obtained at the brachial artery; in the midforearm, where pressures were obtained in both the radial and ulnar arteries; and in each digit. MVL Classic equipment (Health-watch Lifesciences, Vista, CA) was used. PVRs were made with a calibrated, preset volume of air for the segment being evaluated and cuff sizes. Cuff sizes were 12 × 23 cm for the humeral and forearm and 7 × 2 cm or 9 × 3
Hemodynamic effects of radial artery harvest 57
cm for the digits, depending on the size of the fingers. Allen testing with PPG was performed to obtain information on individual digits during compression maneuvers. The PPG probe was attached to each fingertip with double-stick tape or a Velcro威 strap, and waveforms were obtained at baseline and with radial artery and ulnar artery compression (RAC and UAC, respectively) at the wrist. Care was taken to have the patient’s hand relaxed without hyperextension of the fingers. If a Velcro strap was used to fix the PPG probe in place, it was wrapped snug enough to maintain contact with the patient’s skin yet loose enough so as not to cut off blood flow to the fingertip. Dependency on the radial artery for digital perfusion was demonstrated by a loss of signal on the PPG waveform with RAC and return of the signal with release. Arterial duplex imaging was performed on an ATL HDI 3000 (Advanced Technologies Laboratories, Bothell, WA) of all vessels from the clavicle to the wrist with either an L10-5, L7-4, or CL10-5 mHz extended operating frequency broadband linear array transducer. The selection of transducer was based on optimal visualization. In addition to the radial and ulnar arteries, the subclavian, axillary, and brachial arteries were imaged. All vessels were imaged in the transverse plane to evaluate for the presence or absence of atherosclerotic disease, turbulence in the color flow, and any anomalies. Vessel diameters were determined with use of electronic calipers. Difficulties were encountered in the visualization of the subclavian artery in some, resulting in very short segments of visualization. These difficulties included patient habitus and obstruction of visualization by the clavicle, sternum, and lateral border of the first rib. Peak systolic velocities (PSVs) were obtained from a sagittal view of each vessel with the flow indicator parallel to the vessel wall. In addition, the PSVs of the radial and ulnar arteries were evaluated at the wrist with manual compression of the other vessel. The SPA was imaged with the hand relaxed and the fingers slightly flexed. The SPA was differentiated from the deep palmar arch by anatomic location. The SPA lies volar to the flexor tendons, whereas the deep palmar arch lies more dorsally. Direction of flow was determined as forward for each patient at rest. If direction changed with either RAC or UAC compression, then this was considered reverse of flow in the SPA (Fig. 1). The vessel was assessed, again, for vessel diameter, the presence or absence of atherosclerotic disease and/or turbulence, and PSV. The PSV of the SPA was also evaluated with RAC and UAC. Because of the superficial nature of the upper
58
Lohr et al.
Fig. 1. SPA measurements were taken between the third and fourth digits at transverse palmar crease.
extremity arterial (UEA) circulation, atherosclerotic disease is readily and easily identifiable and quantifiable. Atherosclerotic disease was quantified by percentage of diameter reduction from measurements obtained by B-mode imaging. A residual diameter measurement was obtained at the point of maximal narrowing, and a normal arterial diameter was taken as the smallest normal diameter. Diameter reduction, or percentage of stenosis, was derived using the two previously mentioned variables in the following equation: [1 − (residual diameter/ normal diameter)] × 100. Because the difficulties in visualization of the subclavian artery previously described, the presence of disease was marked by the appearance of turbulence in the spectral and color Doppler images. Dominance of the SPA and digital artery blood flow was determined by the following criteria. Ulnar artery–dominant flow was demonstrated with obliteration of the PPG waveforms of all five digits with UAC. Additionally, loss of the Doppler signal of the SPA with UAC was also considered an indicator of ulnar artery dominance. Conversely, RAC produced the same effects in radial artery-dominant SPAs and digits. The findings of radial artery dominance absolutely contraindicated harvest of the radial artery. Mixed-dominant SPAs were those in-
Annals of Vascular Surgery
stances in which RAC and UAC did not produce complete loss of the Doppler signal. In these SPAs, both the radial and ulnar arteries were believed to contribute to SPA blood flow. In the ulnar artery- and mixed-dominant SPAs and digits, the next step was to determine whether other factors, such as atherosclerotic disease or vasospasm, that would contraindicate harvest of the radial artery were present. Removal of the radial artery was simulated with RAC maneuvers previously described. The suggestion of insufficient flow to the hands and digits by any of these maneuvers cautioned against harvest of the radial artery. All three components of RAM for each extremity were evaluated in patients with mixed or ulnar artery dominance. Radial artery harvest was ill advised when the following occurred during UEA segmental Doppler studies, suggesting the presence of atherosclerotic obstructive disease or vasospasm: (1) loss of the dicrotic notch, rounding and loss of amplitude, or absence of the PVR signal at any level; or (2) a >20mmHg pressure drop between cuff levels or between arms at the same level. If studies did not demonstrate either of these two results, the patient was considered to have adequate, unobstructed flow from the subclavian artery to the digits, and the radial artery was considered usable. Dependency on the radial artery for digital perfusion was demonstrated by loss of the signal on the PPG waveform with RAC and return of the signal with release. This obliteration of the digital PPG waveform was found to occur in one to all five digits. In these instances, collateral blood flow would be insufficient if the radial artery was removed. Mixed-dominant perfusion, with contribution from both the radial and ulnar arteries, was evident in the PPG waveforms with no obliteration in all digits with UAC or RAC. Digits were also considered to have mixed dominance if either RAC or UAC caused obliteration in some, but not all, of the digits. The SPA is completed by branches from the deep palmar arch (DPA), radial artery, or median artery in 78.5% of patients; the remaining 21.5% are incomplete. The DPA is completed by the ulnar artery in 98% of patients.8 Any blunting and/or turbulence noted in the baseline digital PPG waveform, in comparison to the other digits, was suggestive of digital small-vessel disease and/or vasospasm, and harvest was contraindicated. The presence of atherosclerotic disease of the UEA system contraindicated harvest of the radial artery. Flow to the SPA was considered radial artery dominant with obliteration of the SPA signal with RAC. Additionally, for flow to be compensatory in
Vol. 14, No. 1, 2000
Hemodynamic effects of radial artery harvest 59
Table I. Differences between preoperative and postoperative duplex data: distal ulnar valuesa
Value
n
Peak systolic velocity (m/sec)
27
End diastolic velocity (m/sec)
27
Resistive index
26
a
Preoperative value
Postoperative value
0.50 ± 0.05 (range −0.52-0.95) 0.03 ± 0.03 (range −0.21-0.23) 0.97 ± 0.05 (range 0.66-1.48)
0.67 ± 0.04 (range 0.34-1.24) −0.10 ± 0.05 (range −1.10-0.32) 1.13 ± 0.05 (range 0.70-1.44)
Difference between preop and postop values
p
0.17 ± 0.07
0.02
−0.12 ± 0.06
0.05
0.16 ± 0.06
0.02
Values are reported as mean ± standard error.
the ulnar artery, suggesting continuity between the radial and ulnar arteries at the SPA, the PSV of the ulnar artery should increase with RAC. A decrease or no change in the ulnar artery PSV is suggestive of insufficient flow into the hand following removal of the radial artery. Kamienski and Branes demonstrated that this increase in velocity correlates with the results of Allen testing and implies that the loss of the artery compressed (radial artery) would result in impaired circulation to the hand.9 Pola et al. suggested that this lack of an increase in ulnar artery PSV with disappearance of flow in the SPA with RAC contraindicates harvest of the radial artery.10 The absolute effects of this scenario are unknown but are implied by the results of Kamienski and Branes.9 In our experience and opinion, loss of SPA signal with RAC demonstrates radial artery dominance, and this, alone, strictly contraindicates harvest.
RESULTS Twenty-seven patients were studied between November 1996 and February 1997 with arterial duplex, PPGs, segmental pressures, and PVRs both preoperatively and following radial artery harvesting. Seven patients were women (average age 59.7 ± 7.0 years; range 48-66 years) and 20 were men (average age 59.0 ± 6.6 years; range 44-71). One right and 26 left radial arteries were harvested. Two patients were identified to have atherosclerosis on preoperative duplex scanning (one in the midulnar and one in the mid- and distal ulnar arteries). Both patients had <30% stenosis. The average postoperative length of time to study was 66 ± 22 days. Differences between preoperative and postoperative duplex data were compared. In looking at the distal ulnar artery values there was a significant difference in the PSV, end diastolic velocity (EDV), and the resistive index (RI) as shown in Table I. The
average diameter of the distal ulnar artery preoperative was 2.7 ± 0.1 mm and postoperative it was 2.6 ± 0.1 mm (not significantly different, p = 0.3275). At rest, no significant difference in the SPA peak systolic could be identified when comparing the preoperative and postoperative scans. The EDV, however, in the SPA and the RI were significantly different. The peak systolic SPA velocity was significantly changed with ulnar artery compression (Table II) as was the ulnar compressive RI. Analysis of the preoperative and postoperative digital pressures revealed no statistical difference in pressures in the long, ring, and little fingers at rest. The decrease in perfusion in both the thumb and index finger when comparing the preoperative and postoperative scans and pressures approached a significant difference (Table III). No significant differences were identified in the brachial, radial, or ulnar artery pressures. At the time of the postoperative visits, eight patients had complaints. Six patients complained of numbness and tingling with some grip and strength changes, however, these were not thought to be circulatory in nature but related to the technique of harvesting and localized nerve injury. One patient complained of tightness along the incision and one patient developed a wound infection at 4 weeks following radial artery harvesting. At the 1-year follow-up visit to the surgeon, no patient had complaints of hand claudication. In an effort to predict the changes that occur with radial artery harvest, we attempted to simulate radial artery harvesting by manually performing radial artery compression and then measuring the pressures in the digits. We randomly selected 27 control patients (10 females and 17 males, average age 62.9 ± 2.1) who had had only preoperative studies done with this additional technique. Using radial artery compression and obtaining digital
60
Lohr et al.
Annals of Vascular Surgery
Table II. Palmar arch evaluationa
Value
n
At rest Peak systolic velocity (m/sec) End diastolic velocity (m/sec) Resistive index Ulnar compression Peak systolic velocity (m/sec) End diastolic velocity (m/sec) Resistive index a
27 27 27
22 13 11
Difference between preop and postop values
Preoperative value
Postoperative value
0.35 ± 0.03 (range 0.16-0.84) 0.03 ± 0.02 (range −0.15-0.16) 0.96 ± 0.05 (range 0.67-1.52)
0.44 ± 0.05 (range −0.24-1.04) −0.05 ± 0.02 (range −0.30-0.17) 1.12 ± 0.05 (range 0.64-1.46)
0.23 ± 0.05 (range −0.06-1.05) 0.072 ± 0.06 (range −0.17-0.80) 0.82 ± 0.11 (range 0.00-1.47)
0.005 ± 0.01 (range −0.16-0.13) −0.0046 ± 0.01 (range −0.13-0.06) 0.27 ± 0.12 (range 0.00-1.00)
p
0.08 ± 0.06
0.16
−0.07 ± 0.02
<0.01
0.16 ± 0.06
0.01
−0.22 ± 0.05
<0.01
−0.077 ± 0.07
0.32
−0.55 ± 0.16
<0.01
Values are reported as mean ± standard error.
Table III. Differences between preoperative and postoperative pressuresa
Site
Arm Radial artery Ulnar artery Thumb Index finger Long finger Ring finger Little finger a
Preoperative pressures (mmHg)
Postoperative pressures (mmHg)
134 ± 4 (range 88-190) 132 ± 7 (range 114-142) 128 ± 4 (range 87-160) 131 ± 6 (range 67-195) 137 ± 6 (range 81-210) 139 ± 5 (range 88-207) 134 ± 6 (range 76-205) 135 ± 6 (range 78-195)
139 ± 4 (range 104-190) 129 ± 12 (range 108-160) 132 ± 4 (range 99-178) 117 ± 7 (range 35-207) 124 ± 5 (range 78-211) 130 ± 6 (range 83-212) 131 ± 5 (range 92-198) 128 ± 5 (range 79-180)
Difference between preop and postop pressures (mmHg)
p
5.1 ± 4.0
0.21
−2.8 ± 12.3
0.84
3.9 ± 4.1
0.36
−13.6 ± 7.6
0.09
−13.3 ± 6.8
0.06
−8.0 ± 6.6
0.24
−2.9 ± 4.9
0.56
−6.7 ± 6.4
0.30
Values are reported as mean ± standard error.
pressures we were able to predict the same magnitude of changes that were identified in the postoperative harvest scans at 2 to 4 months time (Table IV).
DISCUSSION The use of duplex arterial mapping in the upper extremities allows for the evaluation of arterial stenoses. The presence of arterial stenoses being diag-
nosed by pressure gradients and PVR changes is well documented. However, these techniques only allow for interpretation into broad ranges such as normal, diseased (mild, moderate, severe), or occluded. Use of conventional duplex imaging for digital arteries can be very time consuming and is highly technician dependent. The presence of calcifications in the radial artery does not contraindicate its use. Both patients in this study group who had atherosclerotic disease of <30% stenosis toler-
Vol. 14, No. 1, 2000
Hemodynamic effects of radial artery harvest 61
Table IV. Comparison of mean differences of digital pressures between the 27 patients who had a post-study and the control groupa
Digit
Mean difference for 27 patients who had a post-studyb
Control groupc preoperative pressures (mmHg)
Control groupc postoperative pressures (mmHg)
Mean difference for control group
p
Thumb Index Long Ring Little
13.6 ± 7.6 13.3 ± 6.8 8.0 ± 6.6 2.9 ± 4.9 6.7 ± 6.4
118.4 ± 7.3 123.0 ± 7.0 129.2 ± 6.8 129.1 ± 7.7 116.6 ± 6.0
103.2 ± 6.8 111.5 ± 7.2 119.2 ± 7.5 117.0 ± 7.4 110.2 ± 4.9
15.2 ± 4.2 11.4 ± 4.4 10.0 ± 5.2 12.1 ± 6.2 6.4 ± 5.4
0.8482 0.8162 0.8167 0.2442 0.9719
a
Values are reported as mean ± standard error, the control group is composed of patients who had had only preoperative studies done with this additional technique. b See Table III. c n = 27 harvests.
ated the harvest well but graft durability in these patients is unknown. Allen testing has been widely used since 1929 to diagnose disease and predict the presence of collateral circulation in the hands and fingers. But this clinical test is operator dependent and flawed by a significant number of false positives and false negatives.9 Little et al. suggested the use of a Doppler at the interdigitary clefts with associated Allen testing.11 In this study the interdigital arteries were insonnated with the Doppler at the interdigital clefts using either radial artery compression or ulnar artery compression and evaluating interdigital arterial blood flow. This, however, did not provide enough circulatory information for the individual digits. Some authors have suggested that an absolute thumb pressure of 20-40 mmHg or less predicts inadequate collateral supply for the radial artery.12-15 Other authors have suggested that prior to radial artery harvest a decrease in digital pressure of 40 mmHg or more with radial artery compression is consistent with inadequate collateral blood supply. The absolute number or exact gradient in digital pressures to determine inadequate collateral radial artery blood supply is still not known.13,16,17
CONCLUSION The ideal preoperative radial artery assessment method has not yet been identified and unfortunately, several patients who may tolerate radial artery harvest may be excluded. This ideal preoperative test should be quick, cost-effective, reliably exclude patients who would develop complications, and have low false-positive and false-negative rates. There is a great deal of difficulty in predicting postoperative changes. Some patients have increased
sympathetic tone a few days prior to undergoing a major operation and at the time preoperative noninvasive vascular studies are performed.18 The effect of postoperative medications and vascular tone may result in changes that are not predictable. Infection, local injury, and tendonitis may all alter the patient’s perception of wellness in the postoperative period. These 27 patients tolerated their harvest from a hemodynamic standpoint very well without ischemic changes at rest and at the 1-year followup. This study does not identify which variable can be used as a single method of assessment. Further evaluation in this group with a wide range of available vascular laboratory testing needs to continue until modification and specific contraindications are identified for the use of radial artery conduits. We would like to thank Dr. Donald Buckley, Dr. Richard Glaser, Dr. Ranjit Rath, Dr. Tracy Schreiber, and Dr. J. Michael Smith for referring patients to the study. We would also like to thank the staff of the John J. Cranley Vascular Laboratory and the Dr. E. Kenneth Hatton Research Center of Good Samaritan Hospital.
REFERENCES 1. Carpentier A, Guermonprez JL, Deloche A, et al. The aortato-coronary radial artery bypass graft: A technique avoiding pathological changes in grafts. Ann Thorac Surg 1973;16: 111-121. 2. Dietl C, Benoit CH. Radial artery graft for coronary revascularization: technical considerations. Ann Thorac Surg 1995;60:102-110. 3. Barner HB. Defining the role of the radial artery. Semin Thorac Cardiovasc Surg 1996;8:3-9. 4. de Costa FDA, de Costa IA, Poffo R, et al. Myocardial revascularization with the radial artery: a clinical and angiographic study. Ann Thorac Surg 1996;62:475-480. 5. Manasse E, Sperti G, Suma H, et al. Use of the radial artery
62
6.
7.
8.
9.
10.
11.
Lohr et al.
for myocardial revascularization. Ann Thorac Surg 1996;62: 1076-1083. Reyes AT, Frame R, Brodman RF. Technique for harvesting the radial artery as a coronary artery bypass graft. Ann Thorac Surg 1995;59:118-126. Winkler J, Lohr J, Bukhari RH, et al. Evaluation of the radial artery for use in coronary artery bypass grafting. J Vasc Technol 1998;22:23-29. Koman LA, Ruch DS, Paterson Smith B, Smith TL. Vascular disorders. In: Green DP, Hotchkiss RN, Pederson WC, eds. Green’s Operative Hand Surgery, 4th ed. New York Churchill Livingstone, 1999, pp 2254-2257. Kamienski RW, Branes RW. Critique of the Allen test for continuity of the palmar arch assessed by Doppler ultrasound. Surg Gynecol Obstet 1976;142:861-864. Pola P, Serricchio M, Flore R, et al. Safe removal of the radial artery for myocardial revascularization: a Doppler study to prevent ischemic complications to the hand. J Thorac Cardiovasc Surg 1992;112:737-744. Little JM, Zylstra PL, West J, et al. Circulatory patterns in the normal hand. Br J Surg 1973;60:652-655.
Annals of Vascular Surgery
12. Husum B, Palm T. Arterial dominance in the hand. Br J Anaesth 1978;50:913-916. 13. Starnes SL, Wolk SW, Lampman RM, et al. Noninvasive evaluation of hand circulation before radial artery harvest for coronary artery bypass grafting. J Thorac Cardiovasc Surg 1999;117:261-266. 14. Husum B, Berthelsen P. Allen’s test and systolic arterial pressure in the thumb. Br J Anaesth 1981;53:635-637. 15. Wolk SW, Moores HK, Lampman RM, et al. The use of preoperative noninvasive vascular studies for the evaluation of radial artery conduits for coronary artery bypass grafting. Vasc Surg 1998;32:249-253. 16. Gandhi SK, Reynolds AC. A modification of Allen’s test to detect aberrant ulnar collateral circulation. Anesthesiology 1983;59:147-148. 17. Newmeyer WL. Vascular disorders. In: Green DP, ed. Operative Hand Surgery, Vol 3, 2nd ed. New York: Churchill Livingstone, 1988, pp 2391-2458. 18. Dumanian GA, Segalman K, Mispireta LA, et al. Radial artery use in bypass grafting does not change digital blood flow or hand function. Ann Thorac Surg 1998;65:1284-1287.
Twenty-seven patients were studied with arterial duplex, photoplethysmography, segmental pressures, and pulse volume recordings both preoperatively and following radial artery harvesting. The average number of days to the follow-up visit was 66. Preoperative and postoperative data were compared using the matched Student’s t-test. There were no significant changes between preoperative and postoperative pressures in the brachial, radial, and ulnar arteries, and thumb, index, long, ring, or little fingers. Pressure changes in the thumb and index finger approached but did not achieve a statistical difference. Peak systolic velocity (PSV), end diastolic velocity (EDV), and resistive index (RI) in the distal ulnar artery changed significantly between preoperative and postoperative measurements. PSV changed from 0.50 ± 0.05 m/sec to 0.67 ± 0.04 m/sec (p = 0.02); EDV changed from 0.03 ± 0.03 m/sec to −0.10 ± 0.05 m/sec (p = 0.05); and RI changed from 0.97 ± 0.05 to 1.13 ± 0.05 (p = 0.02). Palmar arch evaluations revealed significant changes at rest and with ulnar compression between preoperative and postoperative measurements: (1) at rest EDV changed from 0.03 ± 0.02 m/sec to −0.05 ± 0.02 m/sec (p < 0.01); (2) at rest RI changed from 0.96 ± 0.05 to 1.12 ± 0.05 (p = 0.01); (3) with ulnar compression the PSV changed from 0.23 ± 0.05 m/sec to 0.005 ± 0.01 m/sec (p < 0.01); and (4) with ulnar compression the RI changed from 0.82 ± 0.11 to 0.27 ± 0.12 (p < 0.01). Eight patients had a variety of complaints at the follow-up visit, the majority being numbness and tingling. No patients reported symptoms of claudication or rest pain at the follow-up visit. The data suggest that while statistically significant changes in velocity and arterial resistance do occur, patients seem to tolerate radial artery harvesting without clinical consequences. The ideal method of preoperative evaluation remains to be determined. (Ann Vasc Surg 2000;14:56-62). DOI: 10.1007/s100169910010
INTRODUCTION Carpentier et al. first suggested the use of the radial artery as an arterial conduit for coronary artery bypass grafting in 1973.1 These initial results were dis-
From the John J. Cranley Vascular Laboratory, Good Samaritan Hospital, Cincinnati, OH. Presented at the Twenty-fourth Annual Meeting of the Peripheral Vascular Surgery Society, Washington, DC, June 5-6, 1999. Correspondence to: J.M. Lohr, MD, RVT, c/o Kim Hasselfeld, Research 11J, Good Samaritan Hospital, 375 Dixmyth Avenue, Cincinnati, OH 45220-2489, USA. 56
appointing and the technique was initially abandoned. The introduction of calcium channel blockers and antispasmodic agents has renewed interest in radial artery harvest for use as a conduit for coronary artery bypass grafting.2-5 Radial artery harvest techniques have also been changed to decrease artery handling.6 Clinical contraindications to radial artery harvest include a history of arterial trauma to the involved extremity, Raynaud’s syndrome, and ischemic syndromes, including claudication, rest pain, cold sensitivity, or history of digital ulceration. The upper extremity anatomy is complex and quite varied. A continuous superficial palmar arch (SPA) and digital arteries supplied primarily by the ulnar
Vol. 14, No. 1, 2000
artery is the most common configuration. However, the SPA may not always be complete or may not be based on the ulnar artery. A harvested radial artery in either of these last two situations can be disastrous and can lead to ischemic complications. A reliable method for the assessment of the SPA and the blood flow of the upper extremity prior to radial artery harvest has yet to be defined. The obvious concern is to avoid ischemic complications of the hand. Cardiothoracic surgeons have been extremely conservative as to which vessels are harvested. This study was an attempt to identify arterial dominance into the hand, provide information on the sufficiency of arterial flow and compensatory flow after radial arterial removal, and assess the potential for the development of ischemic complications.
MATERIALS AND METHODS Twenty-seven patients were evaluated with preoperative arterial duplex, photoplethysmography (PPG), segmental pressures, and pulse volume recordings (PVRs) prior to undergoing radial artery harvest during elective coronary artery bypass grafting and in the postoperative period. Data were analyzed using BMDP Statistical Software (BMDP Statistical Software Inc., Los Angeles, CA). Parametric statistical analysis was done with the Student’s t-test, and nonparametric comparisons were done with the Fisher’s exact test. A p-value of <0.05 was taken to represent a significant difference. Patients were referred for radial artery mapping (RAM) in their nondominant arm. Tests were performed in a temperature-controlled room at 70°F. Bilateral or dominant arm RAMs were performed at the request of the cardiovascular surgeon. RAM consisted of three phrases of testing: Doppler measurement of the upper extremity segmental artery pressures and digital pressures, Allen testing with PPG of all appendages, and arterial duplex imaging of the arm of interest including the SPA with and without radial artery compression as described previously by Winkler et al.7 Doppler measurement of segmental arterial pressures was performed bilaterally, with pressures and PVRs performed at the midhumeral region, where pressures were obtained at the brachial artery; in the midforearm, where pressures were obtained in both the radial and ulnar arteries; and in each digit. MVL Classic equipment (Health-watch Lifesciences, Vista, CA) was used. PVRs were made with a calibrated, preset volume of air for the segment being evaluated and cuff sizes. Cuff sizes were 12 × 23 cm for the humeral and forearm and 7 × 2 cm or 9 × 3
Hemodynamic effects of radial artery harvest 57
cm for the digits, depending on the size of the fingers. Allen testing with PPG was performed to obtain information on individual digits during compression maneuvers. The PPG probe was attached to each fingertip with double-stick tape or a Velcro威 strap, and waveforms were obtained at baseline and with radial artery and ulnar artery compression (RAC and UAC, respectively) at the wrist. Care was taken to have the patient’s hand relaxed without hyperextension of the fingers. If a Velcro strap was used to fix the PPG probe in place, it was wrapped snug enough to maintain contact with the patient’s skin yet loose enough so as not to cut off blood flow to the fingertip. Dependency on the radial artery for digital perfusion was demonstrated by a loss of signal on the PPG waveform with RAC and return of the signal with release. Arterial duplex imaging was performed on an ATL HDI 3000 (Advanced Technologies Laboratories, Bothell, WA) of all vessels from the clavicle to the wrist with either an L10-5, L7-4, or CL10-5 mHz extended operating frequency broadband linear array transducer. The selection of transducer was based on optimal visualization. In addition to the radial and ulnar arteries, the subclavian, axillary, and brachial arteries were imaged. All vessels were imaged in the transverse plane to evaluate for the presence or absence of atherosclerotic disease, turbulence in the color flow, and any anomalies. Vessel diameters were determined with use of electronic calipers. Difficulties were encountered in the visualization of the subclavian artery in some, resulting in very short segments of visualization. These difficulties included patient habitus and obstruction of visualization by the clavicle, sternum, and lateral border of the first rib. Peak systolic velocities (PSVs) were obtained from a sagittal view of each vessel with the flow indicator parallel to the vessel wall. In addition, the PSVs of the radial and ulnar arteries were evaluated at the wrist with manual compression of the other vessel. The SPA was imaged with the hand relaxed and the fingers slightly flexed. The SPA was differentiated from the deep palmar arch by anatomic location. The SPA lies volar to the flexor tendons, whereas the deep palmar arch lies more dorsally. Direction of flow was determined as forward for each patient at rest. If direction changed with either RAC or UAC compression, then this was considered reverse of flow in the SPA (Fig. 1). The vessel was assessed, again, for vessel diameter, the presence or absence of atherosclerotic disease and/or turbulence, and PSV. The PSV of the SPA was also evaluated with RAC and UAC. Because of the superficial nature of the upper
58
Lohr et al.
Fig. 1. SPA measurements were taken between the third and fourth digits at transverse palmar crease.
extremity arterial (UEA) circulation, atherosclerotic disease is readily and easily identifiable and quantifiable. Atherosclerotic disease was quantified by percentage of diameter reduction from measurements obtained by B-mode imaging. A residual diameter measurement was obtained at the point of maximal narrowing, and a normal arterial diameter was taken as the smallest normal diameter. Diameter reduction, or percentage of stenosis, was derived using the two previously mentioned variables in the following equation: [1 − (residual diameter/ normal diameter)] × 100. Because the difficulties in visualization of the subclavian artery previously described, the presence of disease was marked by the appearance of turbulence in the spectral and color Doppler images. Dominance of the SPA and digital artery blood flow was determined by the following criteria. Ulnar artery–dominant flow was demonstrated with obliteration of the PPG waveforms of all five digits with UAC. Additionally, loss of the Doppler signal of the SPA with UAC was also considered an indicator of ulnar artery dominance. Conversely, RAC produced the same effects in radial artery-dominant SPAs and digits. The findings of radial artery dominance absolutely contraindicated harvest of the radial artery. Mixed-dominant SPAs were those in-
Annals of Vascular Surgery
stances in which RAC and UAC did not produce complete loss of the Doppler signal. In these SPAs, both the radial and ulnar arteries were believed to contribute to SPA blood flow. In the ulnar artery- and mixed-dominant SPAs and digits, the next step was to determine whether other factors, such as atherosclerotic disease or vasospasm, that would contraindicate harvest of the radial artery were present. Removal of the radial artery was simulated with RAC maneuvers previously described. The suggestion of insufficient flow to the hands and digits by any of these maneuvers cautioned against harvest of the radial artery. All three components of RAM for each extremity were evaluated in patients with mixed or ulnar artery dominance. Radial artery harvest was ill advised when the following occurred during UEA segmental Doppler studies, suggesting the presence of atherosclerotic obstructive disease or vasospasm: (1) loss of the dicrotic notch, rounding and loss of amplitude, or absence of the PVR signal at any level; or (2) a >20mmHg pressure drop between cuff levels or between arms at the same level. If studies did not demonstrate either of these two results, the patient was considered to have adequate, unobstructed flow from the subclavian artery to the digits, and the radial artery was considered usable. Dependency on the radial artery for digital perfusion was demonstrated by loss of the signal on the PPG waveform with RAC and return of the signal with release. This obliteration of the digital PPG waveform was found to occur in one to all five digits. In these instances, collateral blood flow would be insufficient if the radial artery was removed. Mixed-dominant perfusion, with contribution from both the radial and ulnar arteries, was evident in the PPG waveforms with no obliteration in all digits with UAC or RAC. Digits were also considered to have mixed dominance if either RAC or UAC caused obliteration in some, but not all, of the digits. The SPA is completed by branches from the deep palmar arch (DPA), radial artery, or median artery in 78.5% of patients; the remaining 21.5% are incomplete. The DPA is completed by the ulnar artery in 98% of patients.8 Any blunting and/or turbulence noted in the baseline digital PPG waveform, in comparison to the other digits, was suggestive of digital small-vessel disease and/or vasospasm, and harvest was contraindicated. The presence of atherosclerotic disease of the UEA system contraindicated harvest of the radial artery. Flow to the SPA was considered radial artery dominant with obliteration of the SPA signal with RAC. Additionally, for flow to be compensatory in
Vol. 14, No. 1, 2000
Hemodynamic effects of radial artery harvest 59
Table I. Differences between preoperative and postoperative duplex data: distal ulnar valuesa
Value
n
Peak systolic velocity (m/sec)
27
End diastolic velocity (m/sec)
27
Resistive index
26
a
Preoperative value
Postoperative value
0.50 ± 0.05 (range −0.52-0.95) 0.03 ± 0.03 (range −0.21-0.23) 0.97 ± 0.05 (range 0.66-1.48)
0.67 ± 0.04 (range 0.34-1.24) −0.10 ± 0.05 (range −1.10-0.32) 1.13 ± 0.05 (range 0.70-1.44)
Difference between preop and postop values
p
0.17 ± 0.07
0.02
−0.12 ± 0.06
0.05
0.16 ± 0.06
0.02
Values are reported as mean ± standard error.
the ulnar artery, suggesting continuity between the radial and ulnar arteries at the SPA, the PSV of the ulnar artery should increase with RAC. A decrease or no change in the ulnar artery PSV is suggestive of insufficient flow into the hand following removal of the radial artery. Kamienski and Branes demonstrated that this increase in velocity correlates with the results of Allen testing and implies that the loss of the artery compressed (radial artery) would result in impaired circulation to the hand.9 Pola et al. suggested that this lack of an increase in ulnar artery PSV with disappearance of flow in the SPA with RAC contraindicates harvest of the radial artery.10 The absolute effects of this scenario are unknown but are implied by the results of Kamienski and Branes.9 In our experience and opinion, loss of SPA signal with RAC demonstrates radial artery dominance, and this, alone, strictly contraindicates harvest.
RESULTS Twenty-seven patients were studied between November 1996 and February 1997 with arterial duplex, PPGs, segmental pressures, and PVRs both preoperatively and following radial artery harvesting. Seven patients were women (average age 59.7 ± 7.0 years; range 48-66 years) and 20 were men (average age 59.0 ± 6.6 years; range 44-71). One right and 26 left radial arteries were harvested. Two patients were identified to have atherosclerosis on preoperative duplex scanning (one in the midulnar and one in the mid- and distal ulnar arteries). Both patients had <30% stenosis. The average postoperative length of time to study was 66 ± 22 days. Differences between preoperative and postoperative duplex data were compared. In looking at the distal ulnar artery values there was a significant difference in the PSV, end diastolic velocity (EDV), and the resistive index (RI) as shown in Table I. The
average diameter of the distal ulnar artery preoperative was 2.7 ± 0.1 mm and postoperative it was 2.6 ± 0.1 mm (not significantly different, p = 0.3275). At rest, no significant difference in the SPA peak systolic could be identified when comparing the preoperative and postoperative scans. The EDV, however, in the SPA and the RI were significantly different. The peak systolic SPA velocity was significantly changed with ulnar artery compression (Table II) as was the ulnar compressive RI. Analysis of the preoperative and postoperative digital pressures revealed no statistical difference in pressures in the long, ring, and little fingers at rest. The decrease in perfusion in both the thumb and index finger when comparing the preoperative and postoperative scans and pressures approached a significant difference (Table III). No significant differences were identified in the brachial, radial, or ulnar artery pressures. At the time of the postoperative visits, eight patients had complaints. Six patients complained of numbness and tingling with some grip and strength changes, however, these were not thought to be circulatory in nature but related to the technique of harvesting and localized nerve injury. One patient complained of tightness along the incision and one patient developed a wound infection at 4 weeks following radial artery harvesting. At the 1-year follow-up visit to the surgeon, no patient had complaints of hand claudication. In an effort to predict the changes that occur with radial artery harvest, we attempted to simulate radial artery harvesting by manually performing radial artery compression and then measuring the pressures in the digits. We randomly selected 27 control patients (10 females and 17 males, average age 62.9 ± 2.1) who had had only preoperative studies done with this additional technique. Using radial artery compression and obtaining digital
60
Lohr et al.
Annals of Vascular Surgery
Table II. Palmar arch evaluationa
Value
n
At rest Peak systolic velocity (m/sec) End diastolic velocity (m/sec) Resistive index Ulnar compression Peak systolic velocity (m/sec) End diastolic velocity (m/sec) Resistive index a
27 27 27
22 13 11
Difference between preop and postop values
Preoperative value
Postoperative value
0.35 ± 0.03 (range 0.16-0.84) 0.03 ± 0.02 (range −0.15-0.16) 0.96 ± 0.05 (range 0.67-1.52)
0.44 ± 0.05 (range −0.24-1.04) −0.05 ± 0.02 (range −0.30-0.17) 1.12 ± 0.05 (range 0.64-1.46)
0.23 ± 0.05 (range −0.06-1.05) 0.072 ± 0.06 (range −0.17-0.80) 0.82 ± 0.11 (range 0.00-1.47)
0.005 ± 0.01 (range −0.16-0.13) −0.0046 ± 0.01 (range −0.13-0.06) 0.27 ± 0.12 (range 0.00-1.00)
p
0.08 ± 0.06
0.16
−0.07 ± 0.02
<0.01
0.16 ± 0.06
0.01
−0.22 ± 0.05
<0.01
−0.077 ± 0.07
0.32
−0.55 ± 0.16
<0.01
Values are reported as mean ± standard error.
Table III. Differences between preoperative and postoperative pressuresa
Site
Arm Radial artery Ulnar artery Thumb Index finger Long finger Ring finger Little finger a
Preoperative pressures (mmHg)
Postoperative pressures (mmHg)
134 ± 4 (range 88-190) 132 ± 7 (range 114-142) 128 ± 4 (range 87-160) 131 ± 6 (range 67-195) 137 ± 6 (range 81-210) 139 ± 5 (range 88-207) 134 ± 6 (range 76-205) 135 ± 6 (range 78-195)
139 ± 4 (range 104-190) 129 ± 12 (range 108-160) 132 ± 4 (range 99-178) 117 ± 7 (range 35-207) 124 ± 5 (range 78-211) 130 ± 6 (range 83-212) 131 ± 5 (range 92-198) 128 ± 5 (range 79-180)
Difference between preop and postop pressures (mmHg)
p
5.1 ± 4.0
0.21
−2.8 ± 12.3
0.84
3.9 ± 4.1
0.36
−13.6 ± 7.6
0.09
−13.3 ± 6.8
0.06
−8.0 ± 6.6
0.24
−2.9 ± 4.9
0.56
−6.7 ± 6.4
0.30
Values are reported as mean ± standard error.
pressures we were able to predict the same magnitude of changes that were identified in the postoperative harvest scans at 2 to 4 months time (Table IV).
DISCUSSION The use of duplex arterial mapping in the upper extremities allows for the evaluation of arterial stenoses. The presence of arterial stenoses being diag-
nosed by pressure gradients and PVR changes is well documented. However, these techniques only allow for interpretation into broad ranges such as normal, diseased (mild, moderate, severe), or occluded. Use of conventional duplex imaging for digital arteries can be very time consuming and is highly technician dependent. The presence of calcifications in the radial artery does not contraindicate its use. Both patients in this study group who had atherosclerotic disease of <30% stenosis toler-
Vol. 14, No. 1, 2000
Hemodynamic effects of radial artery harvest 61
Table IV. Comparison of mean differences of digital pressures between the 27 patients who had a post-study and the control groupa
Digit
Mean difference for 27 patients who had a post-studyb
Control groupc preoperative pressures (mmHg)
Control groupc postoperative pressures (mmHg)
Mean difference for control group
p
Thumb Index Long Ring Little
13.6 ± 7.6 13.3 ± 6.8 8.0 ± 6.6 2.9 ± 4.9 6.7 ± 6.4
118.4 ± 7.3 123.0 ± 7.0 129.2 ± 6.8 129.1 ± 7.7 116.6 ± 6.0
103.2 ± 6.8 111.5 ± 7.2 119.2 ± 7.5 117.0 ± 7.4 110.2 ± 4.9
15.2 ± 4.2 11.4 ± 4.4 10.0 ± 5.2 12.1 ± 6.2 6.4 ± 5.4
0.8482 0.8162 0.8167 0.2442 0.9719
a
Values are reported as mean ± standard error, the control group is composed of patients who had had only preoperative studies done with this additional technique. b See Table III. c n = 27 harvests.
ated the harvest well but graft durability in these patients is unknown. Allen testing has been widely used since 1929 to diagnose disease and predict the presence of collateral circulation in the hands and fingers. But this clinical test is operator dependent and flawed by a significant number of false positives and false negatives.9 Little et al. suggested the use of a Doppler at the interdigitary clefts with associated Allen testing.11 In this study the interdigital arteries were insonnated with the Doppler at the interdigital clefts using either radial artery compression or ulnar artery compression and evaluating interdigital arterial blood flow. This, however, did not provide enough circulatory information for the individual digits. Some authors have suggested that an absolute thumb pressure of 20-40 mmHg or less predicts inadequate collateral supply for the radial artery.12-15 Other authors have suggested that prior to radial artery harvest a decrease in digital pressure of 40 mmHg or more with radial artery compression is consistent with inadequate collateral blood supply. The absolute number or exact gradient in digital pressures to determine inadequate collateral radial artery blood supply is still not known.13,16,17
CONCLUSION The ideal preoperative radial artery assessment method has not yet been identified and unfortunately, several patients who may tolerate radial artery harvest may be excluded. This ideal preoperative test should be quick, cost-effective, reliably exclude patients who would develop complications, and have low false-positive and false-negative rates. There is a great deal of difficulty in predicting postoperative changes. Some patients have increased
sympathetic tone a few days prior to undergoing a major operation and at the time preoperative noninvasive vascular studies are performed.18 The effect of postoperative medications and vascular tone may result in changes that are not predictable. Infection, local injury, and tendonitis may all alter the patient’s perception of wellness in the postoperative period. These 27 patients tolerated their harvest from a hemodynamic standpoint very well without ischemic changes at rest and at the 1-year followup. This study does not identify which variable can be used as a single method of assessment. Further evaluation in this group with a wide range of available vascular laboratory testing needs to continue until modification and specific contraindications are identified for the use of radial artery conduits. We would like to thank Dr. Donald Buckley, Dr. Richard Glaser, Dr. Ranjit Rath, Dr. Tracy Schreiber, and Dr. J. Michael Smith for referring patients to the study. We would also like to thank the staff of the John J. Cranley Vascular Laboratory and the Dr. E. Kenneth Hatton Research Center of Good Samaritan Hospital.
REFERENCES 1. Carpentier A, Guermonprez JL, Deloche A, et al. The aortato-coronary radial artery bypass graft: A technique avoiding pathological changes in grafts. Ann Thorac Surg 1973;16: 111-121. 2. Dietl C, Benoit CH. Radial artery graft for coronary revascularization: technical considerations. Ann Thorac Surg 1995;60:102-110. 3. Barner HB. Defining the role of the radial artery. Semin Thorac Cardiovasc Surg 1996;8:3-9. 4. de Costa FDA, de Costa IA, Poffo R, et al. Myocardial revascularization with the radial artery: a clinical and angiographic study. Ann Thorac Surg 1996;62:475-480. 5. Manasse E, Sperti G, Suma H, et al. Use of the radial artery
62
6.
7.
8.
9.
10.
11.
Lohr et al.
for myocardial revascularization. Ann Thorac Surg 1996;62: 1076-1083. Reyes AT, Frame R, Brodman RF. Technique for harvesting the radial artery as a coronary artery bypass graft. Ann Thorac Surg 1995;59:118-126. Winkler J, Lohr J, Bukhari RH, et al. Evaluation of the radial artery for use in coronary artery bypass grafting. J Vasc Technol 1998;22:23-29. Koman LA, Ruch DS, Paterson Smith B, Smith TL. Vascular disorders. In: Green DP, Hotchkiss RN, Pederson WC, eds. Green’s Operative Hand Surgery, 4th ed. New York Churchill Livingstone, 1999, pp 2254-2257. Kamienski RW, Branes RW. Critique of the Allen test for continuity of the palmar arch assessed by Doppler ultrasound. Surg Gynecol Obstet 1976;142:861-864. Pola P, Serricchio M, Flore R, et al. Safe removal of the radial artery for myocardial revascularization: a Doppler study to prevent ischemic complications to the hand. J Thorac Cardiovasc Surg 1992;112:737-744. Little JM, Zylstra PL, West J, et al. Circulatory patterns in the normal hand. Br J Surg 1973;60:652-655.
Annals of Vascular Surgery
12. Husum B, Palm T. Arterial dominance in the hand. Br J Anaesth 1978;50:913-916. 13. Starnes SL, Wolk SW, Lampman RM, et al. Noninvasive evaluation of hand circulation before radial artery harvest for coronary artery bypass grafting. J Thorac Cardiovasc Surg 1999;117:261-266. 14. Husum B, Berthelsen P. Allen’s test and systolic arterial pressure in the thumb. Br J Anaesth 1981;53:635-637. 15. Wolk SW, Moores HK, Lampman RM, et al. The use of preoperative noninvasive vascular studies for the evaluation of radial artery conduits for coronary artery bypass grafting. Vasc Surg 1998;32:249-253. 16. Gandhi SK, Reynolds AC. A modification of Allen’s test to detect aberrant ulnar collateral circulation. Anesthesiology 1983;59:147-148. 17. Newmeyer WL. Vascular disorders. In: Green DP, ed. Operative Hand Surgery, Vol 3, 2nd ed. New York: Churchill Livingstone, 1988, pp 2391-2458. 18. Dumanian GA, Segalman K, Mispireta LA, et al. Radial artery use in bypass grafting does not change digital blood flow or hand function. Ann Thorac Surg 1998;65:1284-1287.