- Email: [email protected]
Endoscopic transthoracic sympathectomy for upper limb hyperhidrosis: Limited sympathectomy does not reduce postoperative compensatory sweating
Endoscopic transthoracic sympathectomy for upper limb hyperhidrosis: Limited sympathectomy does not reduce postoperative compensatory sweating Guy Lese`che, MD,a Yves Castier, MD,a Gabriel Thabut, MD,b Marie-Dominique Petit, MD,a Myriam Combes, MD,a Olivier Cerceau, MD,a and Mathieu Besnard, MD,a Clichy, France Objective: Compensatory sweating is the most common and troublesome complication of thoracodorsal sympathectomy. Whether the magnitude of compensatory sweating is related to the extent of sympathectomy is unclear. We investigated the association between the extent of sympathectomy and the occurrence and severity of compensatory sweating after endoscopic transthoracic sympathectomy for upper limb hyperhidrosis. Methods: From September 1992 to June 2000, data from patients undergoing thoracoscopic sympathectomy to treat primary upper limb hyperhidrosis in our department were prospectively collected. Routine follow-up with clinical examination was performed at 1, 3, and 6 months for the first postoperative year and every year thereafter. Late follow-up (February 2001) was with a standardized questionnaire by mail or telephone concerning compensatory sweating and patient satisfaction. Associations between the extent of sympathectomy and the occurrence and severity of compensatory sweating were analyzed with logistic regression and adjusted for age, gender, and relevant confounding factors. Results: Two hundred sixty-eight sympathectomies were consecutively performed in 134 patients (99 female, 35 male; mean age, 27.8 ⴞ 6.7 years). In the 84 patients with palmar hyperhidrosis, eight underwent T1-T2 resection, four T1-T3 resection, eight T2-T3 resection, and 64 T2-T4 resection. In the 43 patients with palmar and axillary hyperhidrosis, eight underwent T1-T5 resection and 35 T2-T5 resection. The seven patients with isolated axillary hyperhidrosis underwent T3-T5 sympathectomy. No deaths occurred; one conversion for bleeding, one permanent Horner’s syndrome, and six minor complications did occur. The initial cure rate was 99.2%. The initial satisfaction rate was 97%. The mean follow-up period was 44.3 months (range, 7 to 100 months), and complete follow-up was available in 132 patients (98.5%). Ninety-five patients (71.9%) had compensatory sweating develop. Seventy patients (53%) judged their compensatory sweating to be minor and intermittent, and 25 patients (19%) judged it severe (16% embarrassing, 3% disabling). On univariate and multivariate analysis, the extent of denervation was not associated with the occurrence or the severity of compensatory sweating. The late satisfaction rate was 91.5%. Compensatory sweating and temporary relief/recurrence were equally considered to be the main causes of dissatisfaction. Conclusion: Compensatory sweating was the most common long-term complication of thoracodorsal sympathectomy for primary hyperhidrosis. Its incidence and severity were not associated with the extent of sympathectomy. (J Vasc Surg 2003;37:124-8.)
Endoscopic transthoracic sympathectomy is a popular surgical technique to treat severe refractory essential hyperhidrosis because it is a safe, effective, and minimally invasive method. However, this is sometimes offset by the development of disturbing side effects, especially compensatory sweating (CS).1 In most series, CS seems to be the most common symptom in patients with postoperative disatisfaction despite successful surgery. According to Shelley and Florence,2 postsympathectomy CS serves a thermoregulatory From the Service de Chirurgie Vasculaire et Thoraciquea and the Service de Pneumologie,b Hoˆpital Beaujon. Competition of interest: nil. Reprint requests: Guy Lese`che, MD, Service de Chirurgie Vasculaire et Thoracique, Hoˆpital Beaujon, Assistance Publique des Hoˆpitaux de Paris, 100 Boulevard du Ge´ne´ral Leclerc, 92110 Clichy, France (e-mail: [email protected]). Published online Nov 27, 2002. Copyright © 2002 by The Society for Vascular Surgery and The American Association for Vascular Surgery. 0741-5214/2002/$30.00 ⫹ 0 doi:10.1067/mva.2002.23
124
function. Although they did not show it, they suggested that the magnitude of CS reflects the extent of denervation. On the basis of this theory, several investigators have advocated the use of limited or selective sympathectomy and have reported their midterm results with mixed conclusions.3-6 On the other hand, Wilkinson7 suggested that patients were more likely to have severe CS if their preexisting hyperhidrosis included more widespread pathologic sweating. Other factors that may account for a high incidence rate of severe CS include hot and humid climate,5 bilateral versus unilateral sympathectomy,8 and a positive family history.7,9 In this study, we especially aimed to investigate whether the extent of denervation was predictive of the occurrence and severity of postoperative CS. We also studied the influence of age, gender, family history, and combined plantar hyperhidrosis with multivariate analysis. PATIENTS AND METHODS From September 1992 to June 2000, data from patients undergoing thoracoscopic sympathectomy to treat
JOURNAL OF VASCULAR SURGERY Volume 37, Number 1
primary upper limb hyperhidrosis in our vascular and thoracic surgery department (Beaujon University Hospital, Clichy, France) were prospectively collected. Data included: patient demographics, family history, clinical indications for surgery, previous therapies, surgical details, immediate and late outcome, and complications. The immediate success and satisfaction with the procedure was evaluated with a standardized questionnaire on the day of discharge from the hospital. Routine follow-up with clinical examination was performed at 1, 3, and 6 months for the first postoperative year and every year thereafter. For late follow-up (February 2001), patients completed a standardized questionnaire by mail or telephone concerning CS and patient satisfaction. To quantify the amount of CS at various sites (chest, abdomen, back, thighs), CS was graded as follow: 1, absent; 2, minor and intermittent, 3, embarassing (ie, excessive sweating in a hot environment or during exercise); and 4, disabling (ie, almost permanent, need to change during the day, asked for complementary treatment). For statistical analysis, patients in whom embarrassing or disabling CS developed were regrouped as having severe CS. Surgical technique. Endoscopic transthoracic sympathectomy was performed according to a previously described technique.10 Briefly, all patients underwent onestage bilateral endoscopic transthoracic sympathectomy with general anesthesia with a double lumen endotracheal tube with no insufflation of carbon dioxide. All patients had three ports (two for instruments and one for a video camera). Port entries are illustrated in the Fig. In the first 20 patients with palmar hyperhidrosis, the upper limit of sympathetic chain resection included the lower part of the stellate ganglia, and Horner’s syndrome was observed. Subsequently, T1 was not excised, and the sympathetic chain was divided and removed (including the fibers of Kuntz) en bloc (2 cm along) with sharp dissection (no electrocautery) from T2 to T3 or T2 to T4 in patients with palmar hyperhidrosis and from T2 to T5 in patients with a combination of palmar and axillary hyperhidrosis. Patients with isolated axillary hyperhidrosis underwent T3-T5 sympathectomy. Histologic confirmation of the sympathetic chain was obtained in all cases. Statistical analysis. Results are expressed as the mean ⫾ the standard deviation for quantitative variables. Comparisons of continuous and categoric variables were made with the unpaired Student t test, analysis of variance, and Fisher exact test when indicated. To avoid bias linked to potential confounding variables, we used multiple logistic regression analysis to assess the effect of the extent of surgical resection (two, three, or more than four ganglia) on the occurrence or severity of CS. In a first step, age, gender, familial history, combined plantar hyperhidrosis, and extent of surgical resection were screened with univariate analysis for their association with the occurrence and severity of CS. In a second step, all these variables were introduced into the multiple logistic regression model to produce adjusted estimates of the effect of these variables
Lese`che et al 125
Port placement: one in third intercostal space, two in sixth intercostal space. 1 and 2, Instrumental ports; 3, video camera port.
on the occurrence and severity of CS. A P value of less than .05 was considered significant. RESULTS Two hundred sixty-eight sympathectomies were consecutively performed in 134 patients (99 female, 35 male; mean age, 27.8 ⫾ 6.7 years). The locations of excessive sweating indicating surgery were palmar in 84 patients, axillary in seven patients, and a combination in 43 patients. Pedal hyperhidrosis affected 91 patients (approximately 80% of whom had widespread hyperhidrosis). In all cases, hyperhidrosis was bilateral. Ninety-two patients (68.6%) had had hyperhidrosis since childhood, 28 (20.8%) since puberty, and 14 (10.4%) for less than 10 years. Fifty-five subjects (41%) had a family history of hyperhidrosis. Conservative treatment had been tried without effect in all cases and included numerous topical (antiperspirants, aluminium chloride hexahydrate) and alternative therapies, including 100% iontophoresis. In the 84 patients with palmar hyperhidrosis, eight underwent T1-T2 resection, four T1-T3 resection, eight T2-T3 resection, and 64 T2-T4 resection. In the 43 patients with palmar and axillary hyperhidrosis, eight underwent T1-T5 resection and 35 T2-T5 resection. The seven patients with isolated axillary hyperhidrosis un-
JOURNAL OF VASCULAR SURGERY January 2002
126 Lese`che et al
Table I. Baseline characteristics of 134 patients according to extent of denervation Variable
Two ggl (n ⫽ 16)
Three ggl (n ⫽ 75)
Four or more ggl (n ⫽ 43)
P value
26.3 (5.1) 13 (81.2%) 6 (37.5%) 11 (68.8%)
27.5 (5.8) 54 (72.0%) 34 (46.6%) 57 (78.1%)
28.9 (8.5) 32 (74.4%) 15 (34.9%) 23 (53.5%)
.35 .88 .47 .02*
Age (y) (SD) Female gender Positive family history Plantar hyperhidrosis *Statistical significance. SD, Standard deviation; ggl, ganglia.
Table II. Compensatory sweating (n ⫽ 95; 71.9%) in 132 patients with complete follow-up Variable Age (per year increase) Female gender Positive familial history Combined plantar hyperhidrosis Extent of denervation Two ganglia (n ⫽ 16) Three ganglia (n ⫽ 73) Four or more ganglia (n ⫽ 43)
Odds ratio
95% CI
P value
0.92 0.59 0.89 1.31
0.87-0.98 0.22-1.60 0.39-2.00 0.55-3.10
.015* .30 .77 .54
1 1.51 1.22
– 0.45-5.12 0.33-4.45
.51 .77
*Statistical significance.
Table III. Severe compensatory sweating (n ⫽ 25, 19%; embarrassing, n ⫽ 21; disabling, n ⫽ 4) in 132 patients with complete follow-up Variable Age (per year increase) Female gender Positive familial history Combined plantar hyperhidrosis Extent of denervation Two ganglia (n ⫽ 16) Three ganglia (n ⫽ 73) Four or more ganglia (n ⫽ 43)
Odds ratio
95% CI
P value
0.95 1.52 1.04 1.25
0.89-1.02 0.50-4.56 0.42-2.59 0.46-3.38
.20 .46 .93 .67
1 0.99 1.33
– 0.24-4.08 0.31-5.82
.99 .70
derwent T3-T5 sympathectomy. Table I shows patient characteristics according to the extent of denervation. No deaths occurred; one conversion for bleeding, one permanent Horner’s syndrome, and six minor complications (three temporary Horner’s syndrome, two pneumothorax, one intercostal neuralgia) that resolved with no further surgery did occur. All patients except one were initially cured (cure rate, 99.2%). Failure to achieve anhydrosis correlated with the absence of ganglia in the histologic specimen led to successful reoperation in this patient 48 hours after the initial procedure. Initial satisfaction rate was 97%. Four patients were dissatisfied with the immediate outcome of the operation (one with conversion, two with Horner’s syndrome, one with neuralgia). The mean hospital stay was 2.8 days (range, 1 to 7 days). The mean follow-up period was 44.3 months (range, 7 to 100 months). In February 2001, 119 patients answered the postal questionnaire, and the 13 who did not were contacted by phone, resulting in a 98.5% complete followup. Ninety-five patients (71.9%) had CS develop 2 to 7 months after endoscopic transthoracic sympathectomy
(median, 3 months) CS affected the back (n ⫽ 72), chest (n ⫽ 63), abdomen (n ⫽ 48), or buttocks/thighs (n ⫽ 53). Seventy patients judged their CS to be minor and intermittent, 21 embarrassing in a hot environment or during exercise, and four disabling. Tables II and III show the association of CS (Table II) or embarrassing and disabling CS (Table III) with the extent of denervation and four general baseline characteristics. On univariate and multivariate analysis, age was the sole variable associated with a statistically significant increase in the occurrence of CS but not with severe CS. The extent of denervation was not associated with the occurrence or the severity of CS. The number of groups and the heterogeneous sample sizes made it impossible to analyze the relationship between the level of resection and CS. Four patients who underwent limited sympathectomies (T1-T2, n ⫽ 3; T2-T3, n ⫽ 1) for palmar hyperhidrosis had temporary relief/recurrence at 2, 6, 9, and 45 months, respectively, after the initial procedure, and three of them had postoperative axillary CS. Of these four patients, three underwent redo thoracoscopic sympathectomy extended
JOURNAL OF VASCULAR SURGERY Volume 37, Number 1
to T5 with successful early and late outcome. The remaining patient has chosen not to have further surgery. One patient who underwent T3-T5 sympathectomy for axillary hyperhidrosis had some return of axillary sweating 2 years after the initial procedure (recurrence rate, 3.7%). Regarding late patient satisfaction, 87% (115/132) were satisfied or very satisfied with the surgical results and would recommend the operation to their close relatives. Six patients (4.5%) were partially satisfied and would again agree to the operation. Of the 11 patients who were dissatisfied with the operative results, CS was considered to be the main reason in four patients (36.3%). Other causes of dissatisfaction were conversion, Horner’s syndrome, and neuralgia in one patient each and temporary relief/recurrence in four patients.
DISCUSSION Compensatory sweating is one of the most troublesome and commonly noted complications after thoracic sympathectomy. The reported incidence rate of CS after thoracic sympathectomy varies from 24% to 98.6%,3,5 and the rate of severe CS is reported to be as high as 43%.5 This wide range may be partly from the intensity of questioning and the thoroughness of the follow-up examination. In this study on CS, this complication affected 72% of 132 patients with complete follow-up and was classified as severe in 19% (embarrassing in 16%, disabling in 3%). It has been suggested but not yet proven that the incidence of CS is correlated with the extent of denervation and that extensive resection beyond the fourth ganglion could aggravate compensatory hyperhidrosis.1,11 In this series, although sympathectomy was bilateral in all cases and although 88% of patients had three or more ganglia resected, the incidence rate of CS was in the range of that reported after limited T2 only or T2-T3 ganglionectomy. Furthermore, no significant difference was seen in the incidence or severity of CS according to the extent of denervation in our series. This finding supports Lai et al,5 who noted no significant difference in the incidence rate (ie, 98.6%) or severity of CS between the T2 and T2-T3 sympathectomy groups in their series. On the other hand, Hederman3 reported a reduced incidence rate of CS after T2 sympathectomy compared with T2-T4 sympathectomy to treat palmar hyperhidrosis. However, Chiou and Chen9 and Hsu and coworkers12 reported an incidence rate of CS as high as 97% and 81%, respectively, after limited T2 sympathectomy. These large studies from Taiwan5,9,12 described a higher rate of CS than is usually reported in Western series, suggesting that genetic factors or the hot and humid climate in Taı¨wan may play an important role. However, Reisfeld, Nguyen, and Pnini13 in a recent Californian study reported an 83% incidence rate of CS in 650 patients who underwent bilateral T2 sympathectomy for primary palmar hyperhidrosis. Despite the limited resection, CS was considered to be disabling in 8% of 321 patients available for follow-up. Other investigators who limited the level of sectioning to
Lese`che et al 127
the T2-T3 ganglion and its connection did not reduce the occurrence or severity of CS.14-16 In this study, neither the extent of denervation nor a positive family history of hyperhidrosis were correlated with the incidence or severity of CS. This suggests that the value of limited sympathectomy to minimize the magnitude of CS requires further evaluation. On the other hand, more extensive resection of the sympathetic chain seems to provide more effective treatment of the hyperhidrosis. Limited sympathectomy may only provide temporary relief because of incomplete treatment or recurrence from nerve regeneration.17-19 After limited T2 sympathectomy, Fox, Hands, and Collin20 reported some return of palmar sweating and recurrent symptoms in 12 of 50 patients (24%) after a median follow-up period of 1.8 years, and Chiou and Chen9 reported a 16% recurrence rate with a mean follow-up period of 4.9 weeks. In this series, four of 16 patients (25%) who underwent limited sympathectomy (T1-T2 or T2-T3) for palmar hyperhidrosis also had temporary relief/recurrence at 2, 6, 9, and 45 months, respectively, after the initial procedure. In addition, three of them had embarrassing compensatory axillary hyperhidrosis develop. The latter finding has been observed by others16,21 and provides further evidence against not excising the fourth ganglion in patients with palmar hyperhidrosis. To minimize CS, Wittmoser et al4 recommended dividing the rami communicantes while not excising the sympathetic trunk (selective sympathectomy). On this basis, Gossot et al6 retrospectively compared conventional truncal sympathectomy and selective sympathectomy. Both procedures were found to be effective in the early outcome with an equal incidence of CS, and selective sympathectomy resulted in an unacceptable level of recurrence. The authors acknowledged that not excising the sympathetic trunk may lead to functional regeneration of the sympathetic ganglia as suggested by Smithwick22 more than 60 years ago. Our study has certain limitations. The use of welldefined validated scales23,24 would have been more appropriate than the use of our own standardized questionnaires to evaluate success and satisfaction with the procedure, to define the severity of CS, and to assess the effect of surgery on quality of life. Secondly, our inability to detect a difference between extent of sympathectomy and CS may be the results of statistical power issues. On the basis of the previous data and 10 years of experience with thoracoscopic sympathectomy, we have progressively extended sympathectomies to resect T2-T4 in patients with palmar hyperhidrosis. Like others,25 our policy was to excise T5 in patients with axillary hyperhidrosis to denervate the axilla. Neither our results nor studies in the literature support this approach. In conclusion, CS was the most common long-term complication of thoracodorsal sympathectomy for primary hyperhidrosis. Its incidence and severity were not associated with the extent of sympathectomy.
JOURNAL OF VASCULAR SURGERY January 2002
128 Lese`che et al
REFERENCES 1. Moran KT, Brady MP. Surgical management of primary hyperhidrosis. Br J Surg 1991;78:279-83. 2. Shelley W, Florence R. Compensatory hyperhidrosis after sympathectomy. N Engl J Med 1960;24:1056-8. 3. Hederman WP. Present and future trends in thoracoscopic sympathectomy. Eur J Surg 1994;572:17-9. 4. Wittmoser R. Thoracoscopic sympathectomy and vagotomy. In: Cuschieri GB, J Perissat, editors. Operative manuel of endoscopic surgery. New York: Springer; 1992. p. 110-33. 5. Lai YT, Yang LH, Chio CC, Chen HH. Complications in patients with palmar hyperhidrosis treated with transthoracic endoscopic sympathectomy. Neurosurgery 1997;41:110-3. 6. Gossot D, Toledo L, Fritsch S, Celerier M. Thoracoscopic sympathectomy for upper limb hyperhidrosis: looking for the right operation. Ann Thorac Surg 1997;64:975-8. 7. Wilkinson HA. Percutaneous radiofrequency upper thoracic sympathectomy. Neurosurgery 1996;38:715-25. 8. Andrews BT, Rennie JA. Predicting changes in the distribution of sweating following thoracoscopic sympathectomy. Br J Surg 1997;84): 1702-4. 9. Chiou TS, Chen SC. Intermediate-term results of endoscopic transaxillary T2 sympathectomy for primary palmar hyperhidrosis. Br J Surg 1999;86:45-7. 10. Lese`che G, Nicolet J, Andre´ assian B. Treatment of primary hyperhidrosis of the upper limbs by transthoracic endoscopic sympathectomy. Presse Med 1995;24:1569-73. 11. O’Riordain DS, Maher M, Waldron DJ, O’Donovan B, Brady MP. Limiting the anatomic extent of upper thoracic sympathectomy for primary palmar hyperhidrosis. Surg Gynecol Obstet 1993;176:151-4. 12. Hsu CP, Chen CY, Lin CT, Wang JH, Chen CL, Wang PY. Videoassisted thoracoscopic T2 sympathectomy for hyperhidrosis palmaris. J Am Coll Surg 1994;179:59-64.
13. Reisfeld R, Nguyen R, Pnini A. Endoscopic thoracic sympathectomy for treatment of essential hyperhidrosis syndrome: experience with 650 patients. Surg Laparosc Endosc Percutan Tech 2000;10:5-10. 14. Gjerris F, Olesen HP. Palmar hyperhidrosis. Long-term results following high thoracic sympathectomy. Acta Neurol Scand 1975;51:167-72. 15. Drott C, Gothberg G, Claes G. Endoscopic transthoracic sympathectomy: an efficient and safe method for the treatment of hyperhidrosis. J Am Acad Dermatol 1995;33:78-81. 16. Kopelman D, Hashmonai M, Ehrenreich M, Bahous H, Assalia A. Upper dorsal thoracoscopic sympathectomy for palmar hyperhidrosis: improved intermediate-term results. J Vasc Surg 1996;24:194-9. 17. van Rhede van der Kloot EJ, Jorning PJ. Resympathectomy of the upper extremity. Br J Surg 1990;77:1043-5. 18. Hsu CP, Chen CY, Hsia JY, Shai SE. Resympathectomy for palmar and axillary hyperhidrosis. Br J Surg 1998;85:1504-5. 19. Lin TS. Video-assisted thoracoscopic “resympathicotomy” for palmar hyperhidrosis: analysis of 42 cases. Ann Thorac Surg 2001;72:895-8. 20. Fox AD, Hands L, Collin J. The results of thoracoscopic sympathetic trunk transection for palmar hyperhidrosis and sympathetic ganglionectomy for axillary hyperhidrosis. Eur J Vasc Endovasc Surg 1999;17: 343-6. 21. Greenhalgh RM, Rosengarten DS, Martin P. Role of sympathectomy for hyperhidrosis. Br Med J 1971;1:332-4. 22. Smithwick R. The rationale and technic of sympathectomy for relief of vascular spasm of the extremities. N Engl J Med 1940;22:699-704. 23. Streiner D, Norman G. Health measurement scale: a practical guide to their development and use. New-York: Oxford Medical Publications; 1998. 24. Cina C, Clase C. The illness instrutiveness rating scale: a measure of severity in individuals with hyperhydrosis. Qual Life Res 1999;8:693-8. 25. Kux M. Thoracic endoscopic sympathectomy in palmar and axillary hyperhydrosis. Arch Surg 1978;113:264-6. Submitted May 9, 2002; accepted Jul 30, 2002.
Receive table of contents by e-mail To receive the tables of contents by e-mail, sign up through our Web site at http://www.mosby.com/jvs Choose E-mail Notification Simply type your e-mail address in the box and click the Subscribe button Alternatively, you may send an e-mail message to [email protected] Leave the subject line blank and type the following as the body of your message: subscribe jvs_toc You will receive an e-mail to confirm that you have been added to the mailing list. Note that TOC e-mails will be sent out when a new issue is posted to the Web site.
Endoscopic transthoracic sympathectomy is a popular surgical technique to treat severe refractory essential hyperhidrosis because it is a safe, effective, and minimally invasive method. However, this is sometimes offset by the development of disturbing side effects, especially compensatory sweating (CS).1 In most series, CS seems to be the most common symptom in patients with postoperative disatisfaction despite successful surgery. According to Shelley and Florence,2 postsympathectomy CS serves a thermoregulatory From the Service de Chirurgie Vasculaire et Thoraciquea and the Service de Pneumologie,b Hoˆpital Beaujon. Competition of interest: nil. Reprint requests: Guy Lese`che, MD, Service de Chirurgie Vasculaire et Thoracique, Hoˆpital Beaujon, Assistance Publique des Hoˆpitaux de Paris, 100 Boulevard du Ge´ne´ral Leclerc, 92110 Clichy, France (e-mail: [email protected]). Published online Nov 27, 2002. Copyright © 2002 by The Society for Vascular Surgery and The American Association for Vascular Surgery. 0741-5214/2002/$30.00 ⫹ 0 doi:10.1067/mva.2002.23
124
function. Although they did not show it, they suggested that the magnitude of CS reflects the extent of denervation. On the basis of this theory, several investigators have advocated the use of limited or selective sympathectomy and have reported their midterm results with mixed conclusions.3-6 On the other hand, Wilkinson7 suggested that patients were more likely to have severe CS if their preexisting hyperhidrosis included more widespread pathologic sweating. Other factors that may account for a high incidence rate of severe CS include hot and humid climate,5 bilateral versus unilateral sympathectomy,8 and a positive family history.7,9 In this study, we especially aimed to investigate whether the extent of denervation was predictive of the occurrence and severity of postoperative CS. We also studied the influence of age, gender, family history, and combined plantar hyperhidrosis with multivariate analysis. PATIENTS AND METHODS From September 1992 to June 2000, data from patients undergoing thoracoscopic sympathectomy to treat
JOURNAL OF VASCULAR SURGERY Volume 37, Number 1
primary upper limb hyperhidrosis in our vascular and thoracic surgery department (Beaujon University Hospital, Clichy, France) were prospectively collected. Data included: patient demographics, family history, clinical indications for surgery, previous therapies, surgical details, immediate and late outcome, and complications. The immediate success and satisfaction with the procedure was evaluated with a standardized questionnaire on the day of discharge from the hospital. Routine follow-up with clinical examination was performed at 1, 3, and 6 months for the first postoperative year and every year thereafter. For late follow-up (February 2001), patients completed a standardized questionnaire by mail or telephone concerning CS and patient satisfaction. To quantify the amount of CS at various sites (chest, abdomen, back, thighs), CS was graded as follow: 1, absent; 2, minor and intermittent, 3, embarassing (ie, excessive sweating in a hot environment or during exercise); and 4, disabling (ie, almost permanent, need to change during the day, asked for complementary treatment). For statistical analysis, patients in whom embarrassing or disabling CS developed were regrouped as having severe CS. Surgical technique. Endoscopic transthoracic sympathectomy was performed according to a previously described technique.10 Briefly, all patients underwent onestage bilateral endoscopic transthoracic sympathectomy with general anesthesia with a double lumen endotracheal tube with no insufflation of carbon dioxide. All patients had three ports (two for instruments and one for a video camera). Port entries are illustrated in the Fig. In the first 20 patients with palmar hyperhidrosis, the upper limit of sympathetic chain resection included the lower part of the stellate ganglia, and Horner’s syndrome was observed. Subsequently, T1 was not excised, and the sympathetic chain was divided and removed (including the fibers of Kuntz) en bloc (2 cm along) with sharp dissection (no electrocautery) from T2 to T3 or T2 to T4 in patients with palmar hyperhidrosis and from T2 to T5 in patients with a combination of palmar and axillary hyperhidrosis. Patients with isolated axillary hyperhidrosis underwent T3-T5 sympathectomy. Histologic confirmation of the sympathetic chain was obtained in all cases. Statistical analysis. Results are expressed as the mean ⫾ the standard deviation for quantitative variables. Comparisons of continuous and categoric variables were made with the unpaired Student t test, analysis of variance, and Fisher exact test when indicated. To avoid bias linked to potential confounding variables, we used multiple logistic regression analysis to assess the effect of the extent of surgical resection (two, three, or more than four ganglia) on the occurrence or severity of CS. In a first step, age, gender, familial history, combined plantar hyperhidrosis, and extent of surgical resection were screened with univariate analysis for their association with the occurrence and severity of CS. In a second step, all these variables were introduced into the multiple logistic regression model to produce adjusted estimates of the effect of these variables
Lese`che et al 125
Port placement: one in third intercostal space, two in sixth intercostal space. 1 and 2, Instrumental ports; 3, video camera port.
on the occurrence and severity of CS. A P value of less than .05 was considered significant. RESULTS Two hundred sixty-eight sympathectomies were consecutively performed in 134 patients (99 female, 35 male; mean age, 27.8 ⫾ 6.7 years). The locations of excessive sweating indicating surgery were palmar in 84 patients, axillary in seven patients, and a combination in 43 patients. Pedal hyperhidrosis affected 91 patients (approximately 80% of whom had widespread hyperhidrosis). In all cases, hyperhidrosis was bilateral. Ninety-two patients (68.6%) had had hyperhidrosis since childhood, 28 (20.8%) since puberty, and 14 (10.4%) for less than 10 years. Fifty-five subjects (41%) had a family history of hyperhidrosis. Conservative treatment had been tried without effect in all cases and included numerous topical (antiperspirants, aluminium chloride hexahydrate) and alternative therapies, including 100% iontophoresis. In the 84 patients with palmar hyperhidrosis, eight underwent T1-T2 resection, four T1-T3 resection, eight T2-T3 resection, and 64 T2-T4 resection. In the 43 patients with palmar and axillary hyperhidrosis, eight underwent T1-T5 resection and 35 T2-T5 resection. The seven patients with isolated axillary hyperhidrosis un-
JOURNAL OF VASCULAR SURGERY January 2002
126 Lese`che et al
Table I. Baseline characteristics of 134 patients according to extent of denervation Variable
Two ggl (n ⫽ 16)
Three ggl (n ⫽ 75)
Four or more ggl (n ⫽ 43)
P value
26.3 (5.1) 13 (81.2%) 6 (37.5%) 11 (68.8%)
27.5 (5.8) 54 (72.0%) 34 (46.6%) 57 (78.1%)
28.9 (8.5) 32 (74.4%) 15 (34.9%) 23 (53.5%)
.35 .88 .47 .02*
Age (y) (SD) Female gender Positive family history Plantar hyperhidrosis *Statistical significance. SD, Standard deviation; ggl, ganglia.
Table II. Compensatory sweating (n ⫽ 95; 71.9%) in 132 patients with complete follow-up Variable Age (per year increase) Female gender Positive familial history Combined plantar hyperhidrosis Extent of denervation Two ganglia (n ⫽ 16) Three ganglia (n ⫽ 73) Four or more ganglia (n ⫽ 43)
Odds ratio
95% CI
P value
0.92 0.59 0.89 1.31
0.87-0.98 0.22-1.60 0.39-2.00 0.55-3.10
.015* .30 .77 .54
1 1.51 1.22
– 0.45-5.12 0.33-4.45
.51 .77
*Statistical significance.
Table III. Severe compensatory sweating (n ⫽ 25, 19%; embarrassing, n ⫽ 21; disabling, n ⫽ 4) in 132 patients with complete follow-up Variable Age (per year increase) Female gender Positive familial history Combined plantar hyperhidrosis Extent of denervation Two ganglia (n ⫽ 16) Three ganglia (n ⫽ 73) Four or more ganglia (n ⫽ 43)
Odds ratio
95% CI
P value
0.95 1.52 1.04 1.25
0.89-1.02 0.50-4.56 0.42-2.59 0.46-3.38
.20 .46 .93 .67
1 0.99 1.33
– 0.24-4.08 0.31-5.82
.99 .70
derwent T3-T5 sympathectomy. Table I shows patient characteristics according to the extent of denervation. No deaths occurred; one conversion for bleeding, one permanent Horner’s syndrome, and six minor complications (three temporary Horner’s syndrome, two pneumothorax, one intercostal neuralgia) that resolved with no further surgery did occur. All patients except one were initially cured (cure rate, 99.2%). Failure to achieve anhydrosis correlated with the absence of ganglia in the histologic specimen led to successful reoperation in this patient 48 hours after the initial procedure. Initial satisfaction rate was 97%. Four patients were dissatisfied with the immediate outcome of the operation (one with conversion, two with Horner’s syndrome, one with neuralgia). The mean hospital stay was 2.8 days (range, 1 to 7 days). The mean follow-up period was 44.3 months (range, 7 to 100 months). In February 2001, 119 patients answered the postal questionnaire, and the 13 who did not were contacted by phone, resulting in a 98.5% complete followup. Ninety-five patients (71.9%) had CS develop 2 to 7 months after endoscopic transthoracic sympathectomy
(median, 3 months) CS affected the back (n ⫽ 72), chest (n ⫽ 63), abdomen (n ⫽ 48), or buttocks/thighs (n ⫽ 53). Seventy patients judged their CS to be minor and intermittent, 21 embarrassing in a hot environment or during exercise, and four disabling. Tables II and III show the association of CS (Table II) or embarrassing and disabling CS (Table III) with the extent of denervation and four general baseline characteristics. On univariate and multivariate analysis, age was the sole variable associated with a statistically significant increase in the occurrence of CS but not with severe CS. The extent of denervation was not associated with the occurrence or the severity of CS. The number of groups and the heterogeneous sample sizes made it impossible to analyze the relationship between the level of resection and CS. Four patients who underwent limited sympathectomies (T1-T2, n ⫽ 3; T2-T3, n ⫽ 1) for palmar hyperhidrosis had temporary relief/recurrence at 2, 6, 9, and 45 months, respectively, after the initial procedure, and three of them had postoperative axillary CS. Of these four patients, three underwent redo thoracoscopic sympathectomy extended
JOURNAL OF VASCULAR SURGERY Volume 37, Number 1
to T5 with successful early and late outcome. The remaining patient has chosen not to have further surgery. One patient who underwent T3-T5 sympathectomy for axillary hyperhidrosis had some return of axillary sweating 2 years after the initial procedure (recurrence rate, 3.7%). Regarding late patient satisfaction, 87% (115/132) were satisfied or very satisfied with the surgical results and would recommend the operation to their close relatives. Six patients (4.5%) were partially satisfied and would again agree to the operation. Of the 11 patients who were dissatisfied with the operative results, CS was considered to be the main reason in four patients (36.3%). Other causes of dissatisfaction were conversion, Horner’s syndrome, and neuralgia in one patient each and temporary relief/recurrence in four patients.
DISCUSSION Compensatory sweating is one of the most troublesome and commonly noted complications after thoracic sympathectomy. The reported incidence rate of CS after thoracic sympathectomy varies from 24% to 98.6%,3,5 and the rate of severe CS is reported to be as high as 43%.5 This wide range may be partly from the intensity of questioning and the thoroughness of the follow-up examination. In this study on CS, this complication affected 72% of 132 patients with complete follow-up and was classified as severe in 19% (embarrassing in 16%, disabling in 3%). It has been suggested but not yet proven that the incidence of CS is correlated with the extent of denervation and that extensive resection beyond the fourth ganglion could aggravate compensatory hyperhidrosis.1,11 In this series, although sympathectomy was bilateral in all cases and although 88% of patients had three or more ganglia resected, the incidence rate of CS was in the range of that reported after limited T2 only or T2-T3 ganglionectomy. Furthermore, no significant difference was seen in the incidence or severity of CS according to the extent of denervation in our series. This finding supports Lai et al,5 who noted no significant difference in the incidence rate (ie, 98.6%) or severity of CS between the T2 and T2-T3 sympathectomy groups in their series. On the other hand, Hederman3 reported a reduced incidence rate of CS after T2 sympathectomy compared with T2-T4 sympathectomy to treat palmar hyperhidrosis. However, Chiou and Chen9 and Hsu and coworkers12 reported an incidence rate of CS as high as 97% and 81%, respectively, after limited T2 sympathectomy. These large studies from Taiwan5,9,12 described a higher rate of CS than is usually reported in Western series, suggesting that genetic factors or the hot and humid climate in Taı¨wan may play an important role. However, Reisfeld, Nguyen, and Pnini13 in a recent Californian study reported an 83% incidence rate of CS in 650 patients who underwent bilateral T2 sympathectomy for primary palmar hyperhidrosis. Despite the limited resection, CS was considered to be disabling in 8% of 321 patients available for follow-up. Other investigators who limited the level of sectioning to
Lese`che et al 127
the T2-T3 ganglion and its connection did not reduce the occurrence or severity of CS.14-16 In this study, neither the extent of denervation nor a positive family history of hyperhidrosis were correlated with the incidence or severity of CS. This suggests that the value of limited sympathectomy to minimize the magnitude of CS requires further evaluation. On the other hand, more extensive resection of the sympathetic chain seems to provide more effective treatment of the hyperhidrosis. Limited sympathectomy may only provide temporary relief because of incomplete treatment or recurrence from nerve regeneration.17-19 After limited T2 sympathectomy, Fox, Hands, and Collin20 reported some return of palmar sweating and recurrent symptoms in 12 of 50 patients (24%) after a median follow-up period of 1.8 years, and Chiou and Chen9 reported a 16% recurrence rate with a mean follow-up period of 4.9 weeks. In this series, four of 16 patients (25%) who underwent limited sympathectomy (T1-T2 or T2-T3) for palmar hyperhidrosis also had temporary relief/recurrence at 2, 6, 9, and 45 months, respectively, after the initial procedure. In addition, three of them had embarrassing compensatory axillary hyperhidrosis develop. The latter finding has been observed by others16,21 and provides further evidence against not excising the fourth ganglion in patients with palmar hyperhidrosis. To minimize CS, Wittmoser et al4 recommended dividing the rami communicantes while not excising the sympathetic trunk (selective sympathectomy). On this basis, Gossot et al6 retrospectively compared conventional truncal sympathectomy and selective sympathectomy. Both procedures were found to be effective in the early outcome with an equal incidence of CS, and selective sympathectomy resulted in an unacceptable level of recurrence. The authors acknowledged that not excising the sympathetic trunk may lead to functional regeneration of the sympathetic ganglia as suggested by Smithwick22 more than 60 years ago. Our study has certain limitations. The use of welldefined validated scales23,24 would have been more appropriate than the use of our own standardized questionnaires to evaluate success and satisfaction with the procedure, to define the severity of CS, and to assess the effect of surgery on quality of life. Secondly, our inability to detect a difference between extent of sympathectomy and CS may be the results of statistical power issues. On the basis of the previous data and 10 years of experience with thoracoscopic sympathectomy, we have progressively extended sympathectomies to resect T2-T4 in patients with palmar hyperhidrosis. Like others,25 our policy was to excise T5 in patients with axillary hyperhidrosis to denervate the axilla. Neither our results nor studies in the literature support this approach. In conclusion, CS was the most common long-term complication of thoracodorsal sympathectomy for primary hyperhidrosis. Its incidence and severity were not associated with the extent of sympathectomy.
JOURNAL OF VASCULAR SURGERY January 2002
128 Lese`che et al
REFERENCES 1. Moran KT, Brady MP. Surgical management of primary hyperhidrosis. Br J Surg 1991;78:279-83. 2. Shelley W, Florence R. Compensatory hyperhidrosis after sympathectomy. N Engl J Med 1960;24:1056-8. 3. Hederman WP. Present and future trends in thoracoscopic sympathectomy. Eur J Surg 1994;572:17-9. 4. Wittmoser R. Thoracoscopic sympathectomy and vagotomy. In: Cuschieri GB, J Perissat, editors. Operative manuel of endoscopic surgery. New York: Springer; 1992. p. 110-33. 5. Lai YT, Yang LH, Chio CC, Chen HH. Complications in patients with palmar hyperhidrosis treated with transthoracic endoscopic sympathectomy. Neurosurgery 1997;41:110-3. 6. Gossot D, Toledo L, Fritsch S, Celerier M. Thoracoscopic sympathectomy for upper limb hyperhidrosis: looking for the right operation. Ann Thorac Surg 1997;64:975-8. 7. Wilkinson HA. Percutaneous radiofrequency upper thoracic sympathectomy. Neurosurgery 1996;38:715-25. 8. Andrews BT, Rennie JA. Predicting changes in the distribution of sweating following thoracoscopic sympathectomy. Br J Surg 1997;84): 1702-4. 9. Chiou TS, Chen SC. Intermediate-term results of endoscopic transaxillary T2 sympathectomy for primary palmar hyperhidrosis. Br J Surg 1999;86:45-7. 10. Lese`che G, Nicolet J, Andre´ assian B. Treatment of primary hyperhidrosis of the upper limbs by transthoracic endoscopic sympathectomy. Presse Med 1995;24:1569-73. 11. O’Riordain DS, Maher M, Waldron DJ, O’Donovan B, Brady MP. Limiting the anatomic extent of upper thoracic sympathectomy for primary palmar hyperhidrosis. Surg Gynecol Obstet 1993;176:151-4. 12. Hsu CP, Chen CY, Lin CT, Wang JH, Chen CL, Wang PY. Videoassisted thoracoscopic T2 sympathectomy for hyperhidrosis palmaris. J Am Coll Surg 1994;179:59-64.
13. Reisfeld R, Nguyen R, Pnini A. Endoscopic thoracic sympathectomy for treatment of essential hyperhidrosis syndrome: experience with 650 patients. Surg Laparosc Endosc Percutan Tech 2000;10:5-10. 14. Gjerris F, Olesen HP. Palmar hyperhidrosis. Long-term results following high thoracic sympathectomy. Acta Neurol Scand 1975;51:167-72. 15. Drott C, Gothberg G, Claes G. Endoscopic transthoracic sympathectomy: an efficient and safe method for the treatment of hyperhidrosis. J Am Acad Dermatol 1995;33:78-81. 16. Kopelman D, Hashmonai M, Ehrenreich M, Bahous H, Assalia A. Upper dorsal thoracoscopic sympathectomy for palmar hyperhidrosis: improved intermediate-term results. J Vasc Surg 1996;24:194-9. 17. van Rhede van der Kloot EJ, Jorning PJ. Resympathectomy of the upper extremity. Br J Surg 1990;77:1043-5. 18. Hsu CP, Chen CY, Hsia JY, Shai SE. Resympathectomy for palmar and axillary hyperhidrosis. Br J Surg 1998;85:1504-5. 19. Lin TS. Video-assisted thoracoscopic “resympathicotomy” for palmar hyperhidrosis: analysis of 42 cases. Ann Thorac Surg 2001;72:895-8. 20. Fox AD, Hands L, Collin J. The results of thoracoscopic sympathetic trunk transection for palmar hyperhidrosis and sympathetic ganglionectomy for axillary hyperhidrosis. Eur J Vasc Endovasc Surg 1999;17: 343-6. 21. Greenhalgh RM, Rosengarten DS, Martin P. Role of sympathectomy for hyperhidrosis. Br Med J 1971;1:332-4. 22. Smithwick R. The rationale and technic of sympathectomy for relief of vascular spasm of the extremities. N Engl J Med 1940;22:699-704. 23. Streiner D, Norman G. Health measurement scale: a practical guide to their development and use. New-York: Oxford Medical Publications; 1998. 24. Cina C, Clase C. The illness instrutiveness rating scale: a measure of severity in individuals with hyperhydrosis. Qual Life Res 1999;8:693-8. 25. Kux M. Thoracic endoscopic sympathectomy in palmar and axillary hyperhydrosis. Arch Surg 1978;113:264-6. Submitted May 9, 2002; accepted Jul 30, 2002.
Receive table of contents by e-mail To receive the tables of contents by e-mail, sign up through our Web site at http://www.mosby.com/jvs Choose E-mail Notification Simply type your e-mail address in the box and click the Subscribe button Alternatively, you may send an e-mail message to [email protected] Leave the subject line blank and type the following as the body of your message: subscribe jvs_toc You will receive an e-mail to confirm that you have been added to the mailing list. Note that TOC e-mails will be sent out when a new issue is posted to the Web site.