HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Balci, A. E. Articles by Eren, M. N. Search for Related Content PubMed PubMed Citation Articles by Balci, A. E. Articles by Eren, M. N. Related Collections Chest wall

Ann Thorac Surg 2003;75:1091-1096
© 2003 The Society of Thoracic Surgeons

---

Original article: general thoracic

Surgical treatment of thoracic outlet syndrome: effect and results of surgery

Akn Eraslan Balc, MD^a*, Tansel Ansal Balc, MD^b, Ömer Çakr, MD^a, evval Eren, MD^a, Mehmet Nesimi Eren, MD^a

^a Department of Thoracic & Cardiovascular Surgery, Diyarbakr, Turkey
^b Department of Nuclear Medicine, Dicle University School of Medicine, Diyarbakr, Turkey

Accepted for publication November 1, 2002.

^* Address reprint requests to Dr Balc, Dicle University School of Medicine, Department of Thoracic and Cardiovascular Surgery, 21280 Diyarbakr, Turkey
e-mail: abalci{at}dicle.edu.tr

	Abstract

Top Abstract Introduction Patients and methods Results Comment References

 
BACKGROUND: Because of the difficulty in diagnosis and different treatment options, debate on thoracic outlet syndrome (TOS) has continued. Our aim is to report our surgical experience.

METHODS: Forty-seven patients with thoracic outlet syndrome were operated on between 1985 and 2000. Mean age was 37.9 years (range, 17 to 58 years); female/male ratio was 41/6. The most frequent symptom was paresthesia (72.3%). Seventeen patients (36%) had bilateral symptoms. Of all, 89.3% (42 cases) were neurologic thoracic outlet syndrome, and 10.7% (five cases) were vascular. Lower plexus (C8-T1/ulnar nerve) compression was present in 36 patients and upper plexus (C5-C7/median nerve) compression in 6 patients. Doppler ultrasonography in 11 patients, angiography in 8, and lymph node scintigraphy in 1 patient were also performed. Main operative indications were persistence of symptoms after conservative therapy and reduced (< 60 m/s) ulnar nerve conduction velocity.

RESULTS: Fifty-five operations were performed on the 47 patients. First (59.6%) and cervical costae (21.3%) resections were the most frequent operations. Mean ulnar nerve conduction velocity was 54.8 m/s (range, 43 to 68 m/s) preoperatively and 69.4 m/s (range, 47 to 70 m/s) postoperatively (p < 0.05). The morbidity rate was 17% (8 of 47). No difference was observed between transaxillary and supraclavicular incisions. No brachial plexus injuries occurred. The most frequent cause of morbidity was incisional pain. Two reoperations were performed for recurrences. Follow-up was 4.6 years, and 75% of lower plexus and 50% of upper plexus compressions remained asymptomatic. Severe and long-term pain occurred in 1 patient.

CONCLUSIONS: Surgical decompression for thoracic outlet syndrome is efficient and dependable, but results deteriorate over time.

	Introduction

Top Abstract Introduction Patients and methods Results Comment References

 
Thoracic outlet syndrome (TOS), which is currently being recognized and diagnosed with increasing frequency, is the name given to various clinical manifestations characterized by abnormal compression of the subclavian artery, subclavian vein, or brachial plexus. Upper plexus irritation includes C5, C6, and C7; lower plexus irritation includes C8 and T1 nerve root compression. Because of the difficulty in diagnosis, TOS has been inadequately understood, and treatments have therefore varied [1–3]. Unfortunately, many physicians simply will not make the diagnosis or recognize the syndrome and are reluctant to suggest surgical intervention [4–6]. Selection of patients likely to benefit from surgery and timing of surgery are matters still being debated [2]. Furthermore, some patients who were reported by their surgeons as having improved were subsequently described by others as not having been helped by surgery [7]. Independent observers have to be invited to do long-term follow-up examinations [4, 7, 8]. This retrospective study presents our surgical experience with TOS.

	Patients and methods

Top Abstract Introduction Patients and methods Results Comment References

 
We operated on 47 patients with thoracic outlet syndrome (TOS) between 1985 and 2000. Medical records were studied for patients who had undergone surgical removal of the first or cervical rib. The mean age of the patients was 37.9 years (range, 17 to 58 years), and the female/male ratio was 41/6. All initially had undergone a vascular examination. A cardiologist, a neurologist, and an orthopedic surgeon were consulted in selected cases. To exclude the presence of a ruptured disc or spinal stenosis in the cervical region, computed tomography or magnetic resonance imaging was done. The patients were staged clinically, and their ulnar nerve conduction velocities (UNCV) were recorded preoperatively and postoperatively. Two groups were created relative to preoperative UNCV: group A consisted of 19 patients whose UNCV was 43 to 55 m/s, and group B had 28 patients with UNCV of 56 to 68 m/s. Initial evaluations for all patients were performed at least 4 weeks after the operation. Subsequent follow-up examinations were performed by physicians not involved in the operation.

The following guide for clinical staging was developed: stage 0 = asymptomatic; stage I = appearance of symptoms or reduction in or loss of radial artery pulsations with provocative tests (eg, Adson, Roos); stage II = symptoms in daily life (eg, pain when carrying a dish or dropping light objects); and stage III = deprivational symptoms that hinder the performance of daily tasks, or acute or severe symptoms (eg, severe pain when brushing teeth). Staging is based completely on clinical evaluation and the patient’s history.

The associations between various preoperative and postoperative factors and postoperative outcome were assessed retrospectively. The value of preoperative tests in terms of differential outcomes was assessed using the ² or Fisher’s exact test as appropriate. If p was less than 0.05, the difference was accepted as significant.

Symptoms and diagnosis
The mean duration of symptoms was 28.3 ± 4.4 months (range, 4 to 48 months) for all patients, but 4 to 9 months in 8 patients (arterial, 2; neurologic, 6) whose symptoms were severe. All patients had undertaken unsuccessful medical therapy previously or were severely symptomatic. All had taken painkillers for more than 4 months (nonsteroid antiinflammatory drugs, mostly naproxen sodium or diclofenac sodium), and physiotherapy was given for a median of 4 months (range, 1 to 18 months). The percentage of patients who had been treated for longer than 1 year was 86% and longer than 2 years was 65%.

The most frequent symptom was paresthesia in the arms and hands (72.3%). The most prominent symptoms are shown in Table 1. Twenty-six (53%) patients complained that their symptoms were aggravated when lifting objects above the head or when placing the arms in an exaggerated hyperabduction position. Of all patients, 89.3% (42 patients) had neurologic symptoms, and 10.7% (5 patients) had vascular ones (two arterial, three venous). C8-T1/ulnar nerve compression was present in 36 patients, and C5-C7/median nerve compression was present in 6. Therefore, the lower plexus in 85.7% and the upper plexus in 14.3% of patients were compressed in neurologic cases.

Three-dimensional cervical x-rays were studied routinely, and cervical rib was detected in 19 patients. The Adson, Roos (provocation of symptoms when the arms are abducted to 90 degrees in external rotation), Spurling (tenderness over the scalene muscles), hyperabduction, and costoclavicular tests were positive only 25% to 50% of the time. Indeed, the Adson test was positive in 12 patients, hyperabduction in 7, costoclavicular tests in 9, and tenderness over the scalene muscles in 23. Scalene muscle block with 4 mL of 1% xylocaine infiltration was performed in seven cases and five were positive (loss of tenderness).

Diminished or lost upper extremity pulsations were present in 9 (19%) patients. Two of them had a history of ischemia of the arm at rest or in abduction. Claudication of the affected arm was observed in 33 patients (70.2%). Doppler ultrasonography was performed on 14 patients, angiography on 8, and phlebography on 6. In 2 patients, upper extremity lymph scintigraphy was also done. One of the patients continued to take lymphedema treatment with the diagnosis based on lymph scintigraphy. These studies suggested another diagnosis in only 3 patients, 2 with axillary artery obstruction with angiography and the lymphedema patient who was misdiagnosed, as determined postoperatively.

Upon preoperative evaluation, the most frequent (40.4%) cause of TOS was cervical costae. According to the study by Pollak [9], the number of patients with type I cervical rib was 8; type II, 8; and type III (complete), 3. A considerable percentage (34%) of patients had unknown etiology. Trauma had been sustained by 8 patients, blow to the shoulder area in 5, traffic accidents in 2, and whiplash injury in 1. None of these patients had TOS symptoms before their accidents. The etiology of TOS is shown in Table 2.

Follow-up examination
All patients were followed up. Data were obtained during a clinical visit performed for the purpose of the study in 22 patients or by phone and by letter in the remaining patients. Controls included a physical medicine specialist’s evaluation. The examination consisted of systematic assessment of the movements and mobility of the cervical spine, shoulder girdles, elbows, wrists, and hands. The sensory modalities and reflexes of the upper arms were also tested. In addition to the classic TOS provocation tests, Adson’s, Wright’s, and Eden’s tests and the Roos exercise test were also performed. Both subclavian arteries were auscultated and distal pulses palpated. Status was defined as abnormal if there were definite abnormal findings in either the active or passive movements or in the muscle and nerve functions tested.

	Results

Top Abstract Introduction Patients and methods Results Comment References

 
Fifty-five operations were performed (thromboembolectomy n = 2, bilateral n = 5, reoperation n = 1). First (59.6%) and cervical costae (21.3%) resections were the most frequent operations in all 47 patients (Table 3). Supraclavicular incision for the cervical rib along with transaxillary incision for the first rib were performed preferentially. Of 9 patients whose cervical and first ribs were removed together, 5 were performed by supraclavicular, 2 by transaxillary, and 2 by both supraclavicular and transaxillary routes. No transthoracic route was used. Of patients with cervical rib, 9 underwent a first rib resection in addition to cervical rib removal. Patients were usually discharged on the fifth to seventh postoperative day.

The failure rate of the operation was 10% (1 of 10) for stage I, 17.4% (3 of 23) for stage II, and 14.3% (3 of 14) for stage III. The failure rate of medical treatment was 31% (23 of 74) for stage II and 32.5% (14 of 43) for stage III. Surgical treatment had a lower failure rate than medical treatment in stages II and III (p = 0.037 and 0.004, respectively). Of 8 patients in stage III with severe or acute symptoms, 5 (62.5%) became asymptomatic (stage 0) after the operation, 2 (25%) of them were in stage I, and 1 (12.25%) continued to be in stage III.

The morbidity rate was 17% (8 of 47 patients). Three patients with incisional pain responded to diclofenac sodium (75 mg intramuscularly per day) within 3 days. Pneumothorax (one case) was diagnosed with immediate postoperative checks; this patient needed chest tube drainage for 4 days with negative suction (20 cm H₂O). B vitamins and naproxen sodium were prescribed to 1 patient with aching pain from intercostobrachial nerve damage. The aching pain on the outer surface of the arm continued for 2 months. Staphylococcus aureus was identified from wound infection in 1 patient, and cefotaxime (2 g per day) was prescribed intravenously according to culture antibiogram. Wound hematoma (one case) was drained with intermittent negative suction through a suture line. The complications had no detrimental effects on surgical results. No brachial plexus or vascular injury occurred. The first rib had been resected instead of the cervical rib in a pseudorecurrence patient. The cervical rib was seen with immediate postoperative control x-ray and was removed 2 weeks later.

The long-term follow-up examination was done on average 4.6 years (range, 1.5 to 8 years) later for all patients. All patients were examined routinely 1 and 2 months after the operation. Reoperation was performed for recurrence in 2 patients (4.2%): only one true recurrence (2.1%) occurred 1.5 years after an operation for lower plexus compression. Scar tissue was removed and thoracic sympathectomy added. An improvement in symptoms was observed. The same incisions as the initial operation (transaxillary) were used for recurrences. Severe and long-term pain occurred in 1 patient. At long-term follow-up, 75% of lower plexus and 50% of upper plexus compression remained asymptomatic. In symptomatic patients, symptoms were mostly minimal, and all symptomatic patients were in lower stages than their preoperative stages. Mean recurrence time for symptoms was 3.2 years after operation. No brachial plexus damage or significant bleeding occurred. The follow-up period of vascular cases was uneventful.

Mean UNCV by the Caldwell technique was 54.8 m/s (range, 43 to 68 m/s) preoperatively and 69.4 m/s (range, 47 to 70 m/s) postoperatively (p < 0.05). One patient in stage II continued to have symptoms postoperatively despite improvement of UNCV (56.3 m/s preoperatively, 67 m/s postoperatively). All patients had improvement of UNCV, except for 1 patient with pseudorecurrence. Table 4 shows the postoperative outcome of group A and B patients. Analysis showed no difference between groups A and B for operative failure, morbidity, and recurrence.

	Comment

Top Abstract Introduction Patients and methods Results Comment References

Vascular laboratory studies, functional angiography, and the classic TOS provocation tests (ie, Adson’s, Eden’s and Wright’s tests) are all aimed at disclosing vascular compression, which is present only in 1% to 2% of all patients with TOS. The vascular TOS occurrence rate was 10.6% in our study. None of the preoperative tests used in this study seemed to be of important clinical value. This was already noted by others [4, 10]. The most useful clinical test, with a 70.2% positive result, was the arm claudication test. A lack of uniform neurophysiologic findings means also that there is no reliable neurophysiologic method for diagnosing TOS [7].

Cervical ribs have been reported to be present in about 0.5% of the general population; 10% of cervical ribs give rise to symptoms, and 7.5% to 9% are present in surgical cases [5, 11]. In our study 40.4% of patients had cervical rib. This excessive difference may be due to relatively less frequent referring of patients for an operation by neurologists or physical medicine specialists, contrary to a symptomatic patient with cervical x-rays indicating a cervical rib. Thus, indications for surgical treatment of TOS, associated with the results, must be emphasized. History taking, physical examination, and conservative management of patients initially are the most important aspects of the surgical treatment of TOS. A cervical rib accompanied by vascular symptoms is also an indication for arteriography [12]. No patients with arterial TOS with cervical rib were in our study.

Angiography was performed in 8 patients. The main indication for arteriography was diminished or lost pulsations. Of 8 patients, 2 had arterial obstruction as determined by angiography. Doppler ultrasonography was included in the routine examination of patients after it became available, so no specific indication was present. No thrombosis was observed in our patients. Lymphatic scans of the upper extremity were used to diagnose obstruction of the lymphatic circulation in the region of the cervico-axillary channel and could show the loss of the obstruction postoperatively.

Eight patients with severe or acute symptoms underwent relatively early surgical intervention because they were unable to tolerate conservative (medical) treatment.

Thoracic outlet syndrom affects women more frequently than men. Women comprised approximately 75% of the patients, although they comprised 83% of our study group. Several reasons for the preponderance of female patients have been postulated, including relative underdevelopment of the muscles inserting around the shoulder girdle or relatively lower origin of the brachial plexus with frequent contributions from the second thoracic nerve root.

Resection of the first rib is the procedure most frequently used [11]. Rates of successful outcome after decompression with first rib or cervical rib resections are similar. The patient with a UNCV exceeding 60 m/s usually improves from conservative physical therapy [11]. There was significant increase in UNCV postoperatively (54.8 m/s versus 69.4 m/s; p < 0.05). Preoperative UNCV could not predict postoperative failure. Resection of the first or cervical rib opens the space under the clavicle and loosens the anterior and middle scalene muscles, which are no longer attached to the first rib. In the reported results of surgical treatment of TOS, 55% to 90% of the patients had good or excellent relief [1, 2, 6, 12].

Of 1508 procedures on patients with symptoms of ulnar nerve compression (lower plexus), 76% achieved an excellent result, 20% a good result, 4% a fair result, and fewer than 1% a poor result [13]. With first rib resection and scalenectomy in upper plexus TOS, 77% reported good or excellent results [14], 92% had long-term relief, and 2.2% had recurrent symptoms requiring reoperation [5]. In long-term follow up, good or excellent results of the operation decreased up to 55% [2, 8, 15]. In the present study, 74.5% of operated patients had an improvement in the symptoms. Although relatively less improvement in traumatic TOS has been reported [3], it was 100% for Toso and associates [2]. The percentage of asyptomatic patients was only 25% in our study.

Staging of the patients’ symptoms led to more objective surgical results. This method is easy and universally applicable. We were unable to find a similar staging system in the literature. It might be useful for comparing different studies on TOS.

The major complications of TOS operations include chylothorax from injury to the thoracic duct, phrenic nerve, or brachial plexus and injury or life-threatening hemorrhage of the subclavian artery and vein [1, 5]. The most serious complication in this study was pneumothorax, at 2%, which can occur at the rate of 4.3% to 11.3% [8]. Intercostobrachial nerve damage causes a burning, aching pain on the outer aspect of the arm [12] as in 1 patient, where it persisted for 2 months.

The supraclavicular approach involves a difficult and potentially dangerous dissection around the brachial plexus and the vessels covering the rib. In the infraclavicular approach only the anterior two thirds of the rib can be removed safely [12]. With longer follow-up periods, transthoracic and axillary approaches have been found to have similar outcomes [2]. However, one report [8] indicated that recurrences after the transaxillary approach required a supraclavicular approach for meticulous dissection in order to avoid damage to fragile nerves. We observed no difference between the supraclavicular and transaxillary approaches and no need for supraclavicular incision in cases of recurrences. With the supraclavicular route, no neurologic or vascular compression occurred, but there was difficulty resecting the rib. Although the operative field is deep and tunnel shaped in the transaxillary approach, it offers the best possibility for complete resection of the first rib and all congenital bands [12].

First or cervical rib resection, along with total scalenotomy, was curative regardless of the mechanism of compression [1]. Scalenectomy alone gave good results less than 50% of the time. It has been argued that the transaxillary approach was usually performed as an initial procedure because no muscle division was required [11]. Some surgeons do careful searches for congenital bands, as excision of all bands is necessary for relief of symptoms [1, 14]. There were many anomalies in all cases, which were difficult to classify, such as strong fibrous bands or ligaments, fusion of the first rib with the second, interdigitating of the anterior or middle scalene muscles between the nerves of the plexus, and mostly flat bands over the first rib and cervical nerves. We did not search closely for congenital bands, nor did we attempt to classify them. The main aim of the operation was to enhance and decompress the outlet area. Some surgeons advise shortening the anterior scalene muscle 1 to 2 cm in order to prevent reattachment [1, 8]. In the present study, complete removal of the anterior scalene muscle was tried in association with first rib resection, especially in upper plexus compression. Resection of the posterior third of the rib is crucial, as the lower trunk of the brachial plexus passes the rib in this region [12]. Two patients had incomplete resection of the first rib. One of them became asymptomatic, the other improved from stage III to stage II. We speculate that enhancement of the cervicoaxillary channel should be the main aim of surgical intervention rather than complete resection.

Surgical decompression for TOS is efficient and dependable in selected cases; conservative treatment should be attempted first. Severe symptoms that are resistant to medical treatment should initially bring to mind the beneficial effects of a surgical approach rather than exploration of further medical treatment modalities in order to resolve them, although surgical results deteriorate over time.

	References

Top Abstract Introduction Patients and methods Results Comment References

 

1. Hempel G.K., Rusher A.H., Wheeler C.G., et al. Supraclavicular resection of the first rib for thoracic outlet syndrome. Am J Surg 1981;141:213-215.[Medline]
2. Toso C., Robert J., Berney T., et al. Thoracic outlet syndrome: influence of personal history and surgical technique on long-term results. Eur J Cardiothorac Surg 1999;16:44-47.[Abstract/Free Full Text]
3. Urschel H.C., Razzuk M.A. The failed operation for thoracic outlet syndrome: the difficulty of diagnosis and management. Ann Thorac Surg 1986;42:523-528.[Abstract]
4. Mackinnon S.E. Thoracic outlet syndrome. Ann Thorac Surg 1994;58:287-289.[Medline]
5. Roos D.B. The place for scalenectomy and first-rib resection in thoracic outlet syndrome. Surgery 1982;92:1077-1085.[Medline]
6. Dale W.A. Thoracic outlet compression syndrome. Arch Surg 1982;117:1437-1445.[Abstract]
7. Carroll R.E., Hurst L.C. The relationship of thoracic outlet syndrome and carpal tunnel syndrome. Clin Orthop 1982;164:149-153.
8. Lepantalo M, Lindgren KA, Leino E, et al. Long term outcome after resection of the first rib for thoracic outlet syndrome
9. Pollak E.W. Surgical anatomy of the thoracic outlet syndrome. Surg Gynecol Obstet 1980;150:97-103.[Medline]
10. Qvarfordt P.G., Ehrenfeld W.K., Stoney R.J. Supraclavicular radical scalenectomy and transaxillary first rib resection for the thoracic outlet syndrome. A combined approach. Am J Surg 1984;148:111-116.[Medline]
11. Urschel H.C., Razzuk M.A. Neurovascular compression in the thoracic outlet. Changing management over 50 years. Ann Surg 1998;228:609-617.[Medline]
12. Sallström J, Gjöres JE. Surgical treatment of the thoracic outlet syndrome. Acta Chir Scand 1983;149:555–60
13. Urschel H.C., Razzuk M.A. Upper plexus thoracic outlet syndrome: optimal therapy. Ann Thorac Surg 1997;63:935-939.[Abstract/Free Full Text]
14. Wood V.E., Ellison D.W. Results of upper plexus thoracic outlet syndrome operation. Ann Thorac Surg 1994;58:458-461.[Abstract]
15. Sanders R.J. Results of the surgical treatment for thoracic outlet syndrome. Semin Thorac Cardiac Surg 1996;8:221-228.

This article has been cited by other articles:

				P. Baltopoulos, C. Tsintzos, G. Prionas, and M. Tsironi Exercise-Induced Scalenus Syndrome Am. J. Sports Med., February 1, 2008; 36(2): 369 - 374. [Abstract] [Full Text] [PDF]

				S. Yavuzer, C. Atinkaya, and O. Tokat Clinical predictors of surgical outcome in patients with thoracic outlet syndrome operated on via transaxillary approach Eur. J. Cardiothorac. Surg., February 1, 2004; 25(2): 173 - 178. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Balci, A. E. Articles by Eren, M. N. Search for Related Content PubMed PubMed Citation Articles by Balci, A. E. Articles by Eren, M. N. Related Collections Chest wall