Selection of Patients for Lumbar Sympathectomy

Symposium on Surgical Practices at the Lahey Clinic II

Selection of Patients for Lumbar Sympathectomy Alfred V. Persson, M.D., * Lori A. Anderson, B.S.,t and Frank T. Padberg, Jr., M.D.*

Sympathectomy has been used for many years in the treatment of patients with critical ischemia of the limbs. 1. 6. 7. 12. 13, 17 The value of lumbar sympathectomy has been controversial primarily because of its unpredictable results. 3 Some physicians will not recommend the procedure because they believe it has not provided beneficial clinical and experimental results,4, 5, 9, 10, 14 Other physicians consider sympathectomy appropriate in the treatment of causalgia but rarely elect sympathectomy for the treatment of peripheral occlusive disease. 16 The patients included in our study presented with critical ischemia of a limb manifested as pain at rest, ulcerations, deep infection, gangrene, or blue-toe syndrome. Success or failure of sympathectomy was predicted on the basis of arm-ankle Doppler ratios. We predicted that patients with an arm-ankle Doppler ratio greater than or equal to 0.3 would benefit from lumbar sympathectomy and that patients with an arm-ankle Doppler ratio less than 0.3 would receive no benefit from the procedure, The selection of a Doppler ratio of 0.3 was based on a retrospective 1976 study of 12 patients showing that ischemic limbs with a Doppler ratio greater than or equal to 0.3 benefited from lumbar sympathectomy.ll Similar findings have been reported by others.! We summarize a prospective study of carefully selected patients with severe ischemia of the limb who underwent lumbar sympathectomy, Beneficial results were predicted and realized in 86 per cent of patients.

*Head, Section of Peripheral Vascular Surgery, Lahey Clinic Medical Center, Burlington, Massachusetts tChief Technician, Noninvasive Peripheral Vascular Laboratory, Lahey Clinic Medical Center, Burlington, Massachusetts :j:Assistant Professor of Surgery, Section of Vascular Surgery, University of Medicine and Dentistry of New Jersey, Newark, New Jersey

Surgical Clinics of North America-VoL 65, No.2, April 1985

393

394

ALFRED

V.

PERSSON, LoRI A. ANDERSON AND FRANK

T.

PADBERG, JR.

MATERIALS AND METHODS From October 1976 through January 1982, 42 sympathectomies were performed in 38 patients (23 men and 15 women). Of the 42 sympathectomies, nine were performed in eight diabetic patients, and 18 patients smoked tobacco. The average age was 69 years. Pain at rest was present in 37 limbs, ulcers were apparent on 22 limbs, and deep infection was present in 6 limbs. Three limbs were gangrenous, and one limb also had blue-toe syndrome. Patients with pain at rest had had symptoms of typical nocturnal discomfort for at least 1 month. Symptoms were usually relieved or improved by dependence of the limb. We considered ulcerations (wet or dry) to be superficial if they did not penetrate the dermis. Draining ulcers, osteomyelitis, and exposure of subdermal structures were considered deeptissue involvement or deep infection. Patients with pregangrenous digits or blue-toe syndrome were also candidates for sympathectomy.8 Specimens of all superficial ulcerations and deep infections were submitted for culture, and appropriate antibiotics were administered to patients before and after operation. Intensive local management of these ulcerations was an integral part of patient care. A split-thickness skin graft was beneficial in the care and healing of one large superficial ulceration. All patients in this series underwent lumbar sympathectomy as the only surgical procedure during their hospitalization. Patients who required concomitant or subsequent vascular surgery were excluded from this series. Some patients had previously undergone unsuccessful vascular reconstruction, and 80 per cent of patients presented with nonreconstructible lesions as demonstrated by angiography. The remaining patients underwent lumbar sympathectomy because their general medical condition precluded major vascular reconstructive surgery. Arm-ankle Doppler ratios were obtained before operation for all patients. The ratio is calculated by dividing the systolic blood pressure in the brachial artery by the systolic blood pressure at the ankle. The systolic blood pressure is obtained by using a standard blood pressure cuff and a Doppler velocity meter as a stethoscope. Group 1 (35 limbs) had Doppler ratios greater than or equal to 0.3. Group 2 (seven limbs) had Doppler ratios less than 0.3. We predicted that patients in Group 1 would derive beneficial results from sympathectomy and that patients in Group 2 would receive no benefit. Calculated popliteal-brachial ratios did not provide definitive screening data for these patients . ..

, SURGICAL PROCEDURE

We perform lumbar sympathectomy under general anesthesia by way of a muscle-splitting retroperitoneal approach. The side to be operated on is elevated on a small rolled towel, and the patient is placed in the supine position with the head and feet lowered slightly (Fig. 1). The three abdominal muscles are split in the line of their fibers as in the classic McBurney incision (Fig. 2). Care is taken to avoid cutting the intercostal nerve.

LUMBAR SYMPATHECTOMY

395

Figure 1. Patient is placed in reverse jackknife position with a small rolled towel under the back to elevate the operative side.

When the peritoneum has been exposed, blunt finger dissection is used to separate the peritoneum from the retroperitoneal space. This allows exposure of the psoas muscle, aorta or vena cava, and sympathetic chain as it runs along the spinal column (Fig. 3). The sympathetic nerve is dissected upward with the use of a nerve hook (Fig. 4). The sympathetic chain must not be confused with the genitofemoral nerve that runs laterally over the body of the psoas muscle. The best results are obtained when at least two ganglia and the intervening nerves are excised (Fig. 5). Marking both ends of the sympathetic chain with metal hemoclips is an easy way to maintain hemostasis and forever identifies the absence of the nerve.

Figure 2. Abdominal muscles are incised in the line of their fibers. Care is taken to avoid injuring blood vessels and nerves that lie between transverse muscles and internal oblique muscles.

396

ALFRED

V.

PERSSON, LoRI A. ANDERSON AND FRANK

T.

PADBERG, JR.

Figure 3. Once the peritoneum has been exposed, it is freed from the retroperitoneum by blunt finger dissection. Dissection is carried superiorly to psoas muscle to expose aorta or vena cava and spinal column. Arrow shows position of sympathetic chain in relation to spinal column (in groove between great vessels and psoas muscle).

The wound is closed in layers using nonabsorbable sutures (Fig. 6). The skin is approximated with a running subcuticular suture. The patient is given nothing by mouth for 36 hours after operation and is usually on a regular diet within 3 days. The average period of hospitalization is 5 days. RESULTS The follow-up period was inclusive to 82 months, either by interview or physical examination. End points for follow-up study were amputation or death. Results were based on the degree of resolution of the patient's presenting indications. Complete resolution was considered good. Symptomatic relief with incomplete resolution was regarded as improved. If the procedure failed to relieve the patient's presenting indications, the result was considered unchanged. Forty-two lumbar sympathectomies were performed in 38 patients. Follow-up study was carried out in 100 per cent of patients; the longest survival period is now 82 months after sympathectomy. Approximately 50 per cent of patients died during the follow-up period, most of myocardial infarction. No deaths resulted from complications of sympathectomy procedures, and the 30-day perioperative mortality rate was zero.

LUMBAR SYMPATHECTOMY

397

Figure 4. Sympathetic chain between great vessels and psoas muscle. A nerve hook is used to facilitate dissection.

Of the 37 limbs with pain at rest, complete resolution of pain was attained immediately after operation in 78 per cent. All these patients had Doppler ratios greater than or equal to 0.3. In three patients, rest pain was relieved but some component of pain remained. Five of the 37 limbs with rest pain received no benefit from sympathectomy or actually repre-

Figure 5. Chain is dissected proximally and distally to include at least two ganglia and their branches. Clips are placed prOximally and distally for hemostasis and to mark extent of nerve resection.

398

ALFRED

V.

PERSSON, LoRI

A.

ANDERSON AND FRANK

T.

PADBERG, JR.

Figure 6. Closure of wound by reapproximating muscle layers with interrupted sutures. Skin is closed with running subcuticular Dexon suture.

sented failures in treatment. Of these five, four limbs that also had deep infection were later amputated. In two of the five failures, the patients had Doppler ratios less than 0.3 (Table 1). All patients whose rest pain was initially relieved retained their good result during the follow-up period. The three limbs with initial improved results remained stable during the follow-up period, No amputations were performed in either of these two groups of patients. Of the 22 limbs that contained ulcerations on presentation, 77 per cent were completely healed immediately after operation. No amputations were performed in this group. Six limbs had ulcers that were unchanged after sympathectomy, and all six limbs were amputated. Five of the six also had deep infection, and two had Doppler ratios less than 0.3. One ulcer recurred, and the patient was later found to have osteomyelitis. With the exception of this one patient, all ulcers that were initially healed retained their good result during the follow-up period (Table 2). Of the patients in Group 1 (35 limbs with arm-ankle Doppler ratios greater than or equal to 0.3), 86 per cent had complete resolution of their presenting indications. Of these 35 sympathectomies, five failed; all five had deep infection in addition to rest pain or superficial ulceration or both. Four of these five limbs were amputated (Table 3). Of the seven limbs in Group 2 in patients with arm-ankle ratios less than 0.3, sympathectomy failed in three, which were amputated; Of these three, two also had deep infection. Three successful sympathectomies were performed in this group in patients with Doppler ratios between 0.25 and 0.3 without deep infection. One patient had no detectable Doppler signals before operation but derived a successful result from lumbar sym-

399

LUMBAR SYMPATHECTOMY

Table l. DATE OF FOLLOW-UP SYMPATHECTOMY (MONTHS)

05-75 12-76 01-77 02-77 0&-77 08-77* 10--77 II-77 II-77 12-77 12-77 12-77 02-78 04--78 05-78 10--78 01-79 03--79 03--79 0&-79 07-79 09-79 03--80 04-81 05-81 05-81 05-81 0&-81 10--81 II-81 01-82 02-82 03--82 II-82 II-82 12-82 01-83

78 30 I

82 76 46 74 48 43 2 6 18 40 52 67 42 59 57 58 24 53 25 45 5 27 20 I

30 I

3 23 22 21 14 II 3 II

Rest Pain in 37 Limbs

DOPPLER RATIO

INITIAL RESULT

0.40 0.28 0.00 0.25 0.73 0.36 0.48 0.93 0.35 0.00 0.27 0.37 0.59 0.34 0.42 0.42 0.50 1.00 1.00 0.30 1.00 0.49 0.42 0.47 0.48 0.30 0.40 0.47 0.30 0.34 0.43 0.57 0.82 0.50 0.75 0.29 0.34

Good Good Unchanged Good Good Improved Good Good Improved Improved Good Good Improved Good Good Good Good Good Good Good Improved Good Good Good Good Good Improved Good Unchanged Improved Good Good Good Good Good Unchanged Good

DEEP DATE OF INFECTION DEATH

Yes

0&-79 01-79

Yes

0&-81

DATE OF AMPUTATION

02-77

0&-81 0&-81 02-78 0&-78 0&-79 0&-81 0~2

12-83 04--82

0&-80 12-79 09-81 08-83 Yes Yes Yes

0&-81 03--83

II-81 02-82

02-83

*U]cer recurred.

pathectomy. Arm-ankle ratio was 0.4 after operation. This suggests that a gray zone (less than 0.3) may exist and indicates another group of patients who may benefit from lumbar sympathectomy (Table 4). Of the nine sympathectomies performed in the eight diabetic patients, five were successful and four failed. Four of the five failures also had deep infection. Sympathectomy failed in two of the three patients with gangrene; both of these patients had deep infection. One patient with bluetoe syndrome received a good result from sympathectomy. This patient had normal pedal pulses and no deep infection.

400

ALFRED

V.

PERSSON, LORI A. ANDERSON AND FRANK

Table 2. FOLLOW-UP DATE OF SYMPATHECTOMY (MONTHS)

12-76 08-77 12-77 12-77 01-78 02-78 04-78 05-78 09-78 10-78 03-79 07-79 07-79 03-81 05-81 05-81 05-81 10-81 11-81 03-82 01-82 01-83

30 46 6 2 71

40 52 13 63 42 58 53 3 1 27 20 1 1 3 21 23 11

T.

PADBERG, JR.

Ulcers in 22 Limbs

DOPPLER RATIO

INITIAL RESULT

0.28 0.36 0.27 0.00 1.00 0.59 0.34 0.51 1.00 0.42 1.00 1.00 0.20 0.50 0.48 0.30 0.40 0.30 0.34 0.82 0.43 0.34

Good Improved Good Improved Good Improved Good Good Good Good Good Improved Unchanged Unchanged Good Good Improved Unchanged Improved Good Good Good

DATE OF DEEP INFECTION DEATH

Yes

DATE OF AMPUTATION

06-79 06-81 06-78 02-78 01-84 06-81 08-82 06-79 01-84 04--82 10-79 04--81

Yes Yes

08-83 Yes Yes Yes

03-83

06-81 11-81 02-82

DISCUSSION When lumbar sympathectomy was first introduced, it represented the only alternative to amputation for patients with peripheral vascular disease. Initial enthusiasm ran high, but its role declined with the development of successful surgical techniques for vascular reconstruction. According to two series,6, 15 patients who are not candidates for vascular reconstruction have responded favorably to lumbar sympathectomy. Early advocates were aware that patients with adequate inflow manifested as palpable pulses did better than patients without palpable pulses. 2 During the last decade, measurement of Doppler ankle pressures has provided an objective parameter for determining the critical minimum amount of arterial flow necessary for the successful application of lumbar sympathectomy,ll, 17, 18 By combining careful selection of patients with noninvasive criteria, Walker and Johnston 17 described positive response rates of up to 90 per cent, such as those reported in our group of patients. Arm-ankle Doppler ratios appear to be the most effective method of determining which patients can be managed successfully with lumbar sympathectomy alone. In our group of patients, 86 per cent of limbs with armankle ratios greater than or equal to 0.3 had complete resolution of presenting indications. As recognized earlier, some patients with lower ratios may have successful results. 18 However, the number of patients in this series with ratios less than 0.3 and a successful result is too small to draw

401

LUMBAR SYMPATHECTOMY

Table 3. DOPPLER RATIO

DEEP INFECTION

0.34 0.50 1.00 1.00 0.40 0.50 0.57 0.51 1.00 1.00 0.42 0.43 0.34 0.40 0.50 0.82 0.47 0.75 0.73 0.93 0.36* 0.47 0.35 0.42 0.30 0.48 0.30 0.49 0.48 0.37 0.34 0.30 0.59 0.42 1.00

Yes

Yes Yes

Yes

Yes

Group 1, Success Rate of 86 Per Cent INITIAL RESULT

Good Unchanged Good Improved Good Good Good Good Good Good Good Good Improved Improved Good Good Improved Good Good Good Improved Good Good Good Unchanged Good Good Good Good Good Good Good Good Good Good

FOLLOW-UP RESULT

Success Failure Success Success Success Success Success Success Success Success Success Success Failure Failure Success Success Success Success Success Success Failure Success Success Success Failure Success Success Success Success Success Success Success Success Success Success

AMPUTATION

Yes

Yes Yes

Yes

*Ulcer recurred.

Table 4. DOPPLER RATIO

0.20 0.27 0.28 0.25 0.00 0.29 0.00

Group 2, Failure Rate of 43 Per Cent

DEEP INFECTION

INITIAL RESULT

Yes

Unchanged Good Good Good Unchanged Unchanged Improved

Yes

FOLLOW-UP RESULT

Failure Success Success Success Failure Failure Success

AMPUTATION

Yes

Yes Yes

402

ALFRED

V. PERSSON, LORI A. ANDERSON AND FRANK T. PADBERG, JR.

Table 5.

Deep Infection in Seven Limbs

DATE OF SYMPATHECTOMY

FOLLOW-UP (MONTHS)

DOPPLER RATIO

INITIAL RESULT

DATE OF DEATH

AMPUTATION (DATE)

01-77 08--77 07-79 03--in

1 46 3 1 1 1 3

0.00 0.36 0.20 0.50 0.40 0.30 0.34

Unchanged Improved Unchanged Unchanged Improved Unchanged Improved

01-79

02-77

05--B1

10--81 11-B1

06--B1

10--79

0~3

04--81 06--B1 11-B1 02-B2

any reliable conclusions. Thus, the decision regarding lumbar sympathectomy should be weighed against other risk factors. An absolute ankle pressure of 60 mm Hg or greater has correlated with a high likelihood of successful results from sympathectomy.17 Popliteal-brachial ratios were also obtained for our group of patients, but we did not find the data useful; the recommended criterion of 0.7 would have denied the vast majority of patients in our series the benefit of clinically successful therapy. 12 In this prospective study, we achieved a success rate of 86 per cent in predicting positive results from lumbar sympathectomy in patients with arm-ankle Doppler ratios greater than or equal to 0.3. When predicting failure of the procedure in patients with ratios of less than 0.3, our success rate was less than 50 per cent. Of the eight failed limbs in the entire series, seven occurred in patients whose presenting indications included deep infection (Table 5). Sympathectomy was unsuccessful in all patients with deep infection. The one patient without a deep infection whose sympathectomy failed was a mild diabetic who had an arm-ankle ratio of 0.29.

SUMMARY We successfully predicted that patients presenting with critical ischemia of a limb and Doppler ratios greater than or equal to 0.3 would benefit from lumbar sympathectomy alone. However, we found that the procedure failed in 14 per cent of limbs whose ratios were greater than or equal to 0.3. In retrospect, all these patients were found to have deep infection. Had we known this fact prospectively, our predictions for success would have been close to 100 per cent. We were less than 50 per cent successful in predicting failure of the procedure, but the number of patients in this group is too small to draw reliable conclusions. We believe that patients with arm-ankle Doppler ratios greater than or equal to 0.3 whose manifestations of ischemia are limited to the skin will have a greater than 95 per cent chance of receiving a good result from lumbar sympathectomy alone and that this result will be maintained for many years.

403

LUMBAR SYMPATHECTOMY

REFERENCES 1. Baker,· W. H.: Lumbar sympathectomy for peripheral vascular disease of the lower extremities. In Bergan, J. J., and Yao, J. S. (eds.): Gangrene and Severe Ischemia of the Lower Extremities. New York, Grune & Stratton, 1978, pp. 303-315. 2. Edwards, E. A., and Crane, C.: Ten-year status after sympathectomy for arteriosclerosis. J.A.M.A., 175:677--679, 1961. 3. Ewing, M.: The history oflumbar sympathectomy. Surgery, 70:790-796, 1971. 4. Froysaker, T.: Lumbar sympathectomy in impending gangrene and foot ulcer. Scand. J. Clin. Lab. Invest., 31 (Suppl. 128):71-72, 1973. 5. Fulton, R L., and Blakeley, W. R: Lumbar sympathectomy: A procedure of questionable value in the treatment of arteriosclerosis obliterans of the legs. Am. J. Surg., 116:735-744, 1968. 6. Haimovici, H., Steinman, C., and Karson, I. H.: Evaluation of lumbar sympathectomy. Arch. Surg., 89:1089-1095, 1964. 7. Imparato, A. M.: Lumbar sympathectomy: Role in the treatment of occlusive arterial disease in the lower extremities. Surg. Clin. North Am., 59:719-735, 1979. B. Karmody, A. M., Powers, S. R, Monaco V. J., et al.: "Blue toe" syndrome: An indication for limb salvage surgery. Arch. Surg., 111:1263-1268, 1976. 9. Lindenauer, S. M., and Cronenwett, J. L.: What is the place oflumbar sympathectomy? Br. J. Surg., 69(Supp!.):S32-S33, 1982. 10. Myers, K. A., and Irvine, W. T.: An objective study oflumbar sympathectomy: I. Intermittent claudication. Br. Med. J., 1:879-883, 1966. 11. Persson, A. V., Griffey, S. P., and Kopreski, M.: Use of the noninvasive vascular laboratory as an adjunct to clinical vascular surgery. In Puel, P., Beccalon, H., and Enjabert, A. (eds.): Hemodynamics of the Limbs. Toulouse, France, La Societe de Nouvelle Imprimerie Fournie it Toulouse, 1979, pp. 535-541. 12. Plecha, F. R, Bomberger, R A., Hoffman, M., et al.: A new criterion for predicting response to lumbar sympathectomy in patients with severe arteriosclerotic occlusive disease. Surgery, 88:375-381, 1980. 13. Rutherford, R B.: Lumbar sympathectomy: Indication and technique. In Moore, W. S. (ed.): Vascular Surgery. Philadelphia, W. B. Saunders Co., 1977, pp. 555-562. 14. Strandness, D. E., Jr., and Bell, J. W.: Critical evaluation of the results of lumbar sympathectomy. Ann. Surg., 160:1021-1029, 1964. 15. Szilagyi, D. E., Smith, R F., Scerpella, J. R, et al.: Lumbar sympathectomy: Current role in the treatment of arteriosclerotic occlusive disease. Arch. Surg., 95:753-761, 1976. 16. Thompson, f E., and Garrett, W. V.: Peripheral-arterial surgery. N. Eng!. J. Med., 302:491-503, 1980. 17. Walker, P. M., and Johnston, K. W.: Predicting the success of a sympathectomy: A prospective study using discriminant function and multiple regression analysis. Surgery, 87:216-221, 1980. lB. Yao, J. S. T., and Bergan, J. J.: Predictability of vascular reactivity relative to sympathetic ablation. Arch. Surg., 107:676-680, 1973.

SUGGESTED READINGS Friedman, S. A., Frieberg, P., and Colton, J.: Vasomotor tone in diabetic neuropathy. Ann. Intern. Med., 77:353-356, 1972. Moorhouse, J. A., Carter, S. A., and Doupe, J.: Vascular responses in diabetic peripheral neuropathy. Br. Med. J., 1:883-888, 1966. Ozeran, R S., Wagner, G. R, Reimer, T. R, et al.: Neuropathy of the sympathetic nervous system associated with diabetes mellitus. Surgery, 68:953-958, 1970. Section of Peripheral Vascular Surgery Lahey Clinic Medical Center 41 Mall Road, Box 541 Burlington, Massachusetts 01805