- Email: [email protected]
Neurogenic thoracic outlet decompression: rationale for sparing the first rib
Cardiovascular INEMANN
VASCULAR PAPERS
Surgery, Vol. 3, No. b. pp. hll-623, 1095 Copyright @ 1995 Blsevier Science Ltd Printed in Great Britain. All rights reserved OYh7--2lIWY5 $lO.OO + 0.w
Neurogenicthoracic outlet decompression: rationale for sparing the first rib __-.._-.II .-..._-___ -.._. --S. W. K. Cheng*, L. M. Reilly, N. A. Nelken, W. V. Ellis and R. J. Stoney ReparTmentofSurgery, Divisionof VascularSurgery, M-488 Un~ve~i~/ofCalifornia,505Parn~-~(~u~ Avenue,SanFranckco, CA 94143 USA
A total of 168 primary supraclaviculardecompressionswere performed on 146 patients with neurogenic thoracic outlet syndrome. This report compares the results of rib resection (supraclavicular anterior and middle scalenectomy and first rib resection) with rib-sparing (supraclavicularanterior and middle scalenectomyalone) operations. All patients with cervical ribs were excluded.In total, 125 rib resections and 43 rib-sparing procedures were performed between 1983 and 1992 by a single surgeon. The patients were otherwise comparable in symptoms and physical signs. During surgery there was a significantly higher proportion of pleural injury associatedwith rib resection (59%) than with rib-sparing (40%) procedures, The mean hospital stay was also prolonged by 1 day in patients undergoing rib resection (P= 0.005). There was no significant difference in early successbetween the two groups (83% for rib resection, 91% for rib sparing) and no difference in those resuming employment (52% and 63% respectively).Life-table analysisshowed that the two groups have timPar long-term results (69% and 76% at 2 years). The only important factor determining clinical outcome in primary supraclavicular thoracic outlet syndrome decompression was the duration of symptoms before operation. Some 83% of patients with symptoms less than 2 years had a successful result compared with only 68% in those with symptoms longer than 2 years (PC 0.05). Spontaneous or post-traumatic neurogenic symptoms responded to operation identically.The theoretical benefit of first rib resectionto relievemechanicalcompressionof the brachial plexus is not evident from this review. Thorough removal of the scalene musculature and other myofascialanomalies,preferably through the supraclavicularapproach, leads to less patient morbidity, shortens hospitalization,and is recommended for patients with neurogenic thoracic outlet syndrome requiring operative intervention. -___
Keywords: thoracic outlet syndrome, supraclavicular. rib sparing _- ..__._.. -_
Mechanical compression of the brachial plexus is presumed to initiate the neurogenic symptoms of thoracic outlet syndrome. It has never been established whether the bony (first rib) or the muscular (scalene muscles)elementsof the thoracic outlet are the principal factors contributing to pathogenesis.Much controversy still exists concerning the ideal operation for this group of patients. Removing the first thoracic rib via the
Correspondence to: Dr R. J. Stoney *Present address: Department of Surgery, Queen Mary University of Hong Kong, Hong Kong.
CARDIOVASClJL4R SURGERY DECEMBER IQ95 VOL 3 NO 6
Hospital,
----
--.-..-1-.._ .._.-. -.____-__.___
transaxillary route’ is still the most popular operation. The authors have employed the supradavicular approach because of irs superior exposure, safety and consistently good results2. As experience was gained with the supraclavicular operation, it was observed that the first rib seldom caused direct brachial plexus irritation and its removal was therefore unnecessary. In an attempt to decreasepatient morbidity, the operation has since been modified such that the first thoracic rib is now preserved. This article reports the results of 168 consecutive primary supraclavicular rhoracic outlet decompression procedures performed at the University of California Medical Center, San Francisco, and compares the morbidity and clinical outcome of rib-resecting and rib-sparing operations, 617
Neurogenic thoracic outlet decompression: .S W K Cheng et al.
Patients and methods From 1983 to 1992, 168 primary operations were performed by a single surgeon on 146 patients for neurogenic thoracic outlet syndrome, using the supraclavicular approach. All patients with cervical ribs and those who had previous thoracic outlet decompressions were excluded from this review. There were 117 women (80%) and 29 men (20%); their mean age was 36 years (range 18-60) years. The procedure was carried out for the left extremity in 87 instances (52%); 22 patients had bilateral supraclavicular decompressions, while 124 were unilateral operations. In total, 125 procedures performed between 1983 and 1990 involved anterior and middle scalenectomy and first rib resection (rib-resection group); 43 procedures performed between 1986 and 1992 were supraclavicular anterior and middle scalenectomies with preservation of the first rib (rib-sparing group). The etiology, symptoms and physical findings of the two groups of patients were comparable with no statistically significant difference (Table 1). A history of trauma was reported in 115 cases (68%). Most of the patients had work-related injuries, with flexion-extension injuries of the neck sustained in traffic accidents preceding symptom occurrence in 14% of cases. The patients were usually treated by a period of physical therapy, including exercises and nerve blocks, before they were referred to the authors’ service. The mean duration of symptoms from onset to surgery was 3 years (range 2 months to 25 years) (Table 1). Pain involving the lateral neck and shoulder, and paresthesia radiating to the arms and hands remained the dominant symptoms and were almost universal. Unilateral headache and motor weakness were also common. The symptoms were usually severe and disrupted work (89%) or demanded constant analgesic medications (70%). Most patients had symptom distribution in the dermatomes of all five roots of the brachial plexus, and upper and lower plexus groups could not be clearly distinguished one from the other. The most consistent physical finding was supraclavicular tenderness, present in 88% of all patients. The elevated arm stress test (EAST) was positive in 86% of cases. The Adson test was not found to be reliable, being positive in about two-thirds of the cases. The diagnosis was usually made clinically, based on a characteristic history and classical symptoms and physical signs. Preoperative evaluation included radiography of the chest and cervical spine. A neurological consultation was made to rule out alternative diagnosis. Nerve conduction studies were only performed when the diagnosis of carpel tunnel syndrome was suspected and were abnormal only in 6% of the present cases. Computed tomography and magnetic resonance imaging scans of the neck have been used occasionally to exclude cervical spine pathology. Patients with coexisting vascular compression were excluded from this study.
618
Table 1 History and presentations of 168 cases of neurogenic thoracic outlet syndrome Scalenectomy and first rib resection
Scalenectomy (rib sparing)
All supraclavicular operations
125
43
168
History of trauma (%)
73
56
68
Whiplash neck injuries w Blunt neck trauma (96) Lifting (%) Others (%)
14
14
14
34 12 13
20 4 18
31 10 14
Mean duration of symptoms (months)
39
31
37
Pain (%) Paresthesia (%) Motor weakness (96) Unilateral headache cw
100 95 21
100 100
100 96
19 35
20
38
Work affected by symptoms (%) Cannot work (96) Constant analgesics required (%)
22
33
25
69
51 67
64
70
Supraclavicular tenderness (%) Surpaclavicular mass cw Reduced range of neck motion (%) Positive Adson’s test (W Positive elevated arm stress test (%) Motor deficit or atrophy (%)
87
88
88
2
7
4
31
19
28
66
70
67
84
91
86
9
5
8
Previous physical therapy failed (%) Previous nerve blocks failed (%) Previous neck surgery failed (96)
87
88
87
13
5
11
6
9
No. of operations
38
70
Operative technique and findings All 168 procedures were performed via the supraclavicular approach. The technique of supraclavicular thoracic outlet decompression has been described previously2. This involved anterior and middle scalenectomy followed by a thorough neurolysis of the brachial plexus. First rib resection was performed from 1983 to mid-1990. Afterwards the operation was modified by preserving the first rib after neurolysis of the brachial plexus. Occasionally the medial border of the neck of the first rib will displace the course of the first thoracic nerve root. A bone rongeur may be used to remove a small rim of the medial border of the rib, thus allowing an unrestricted space for the Tl root to emerge from the
CARDIOVASCULAR SURGERY DECEMBER 1995 VOL 3 NO 6
Neurogenicthoracic outlet decomptwsion.5. B! K. Chenget al.
Patient follow-up and assessment The patients underwent a supervised physical therapy program for 8 weeks after discharge. They were reassessedat regular intervals postoperatively. Those who were referred from outside the local area and who had achieved a stable result were also sent a detailed questionnaire for their progress. The results of opera tion were graded using a scoring system based on the percentage improvement of the preoperative symproms’. If all of the syrnptoms were consistently relieved, the patients were considered cured. If there was a > 75% improvement of all symptoms, the patients were classified as markedly improved. An improvement of 50-75% was graded improved and ~50% fair or worse. A <5O% improvement of qymptoms was considered a failure of the operation. Statistical methods b
C
Figure 1 a Illustrated detail showing removal of a rim of the medial border of the first rib using a bone rongeur. The middle scalene is transected in a line parallel to and inferior to the long thoracic nerve. b First rib after segmental osteotomy. c The Tl root runs an unrestricted course after segmental osteotomy of the first rib
intervertebral foramina and join the C8 root to form the lower trunk of the brachial plexus (Figure 1). A rib-sparing decompression has become the authors’ standard operation for neurogenic thoracic outlet syndrome. The operative findings are listed in Table 2. Most of the offending structures were soft tissue anomalies which could only be adequately exposed and removed via the supraclavicular approach. These included abnormally placed anterior and middle scalene insertions, muscular interdigitations, a scalenus minimus muscle, or fibrous bands. Bony anomalies were uncommon and mostly consisted of long transverse processesof the C7 vertebra. Table 2 Pathology encountered in 168 primary supraclavicular thoracic outlet syndrome operations* Pathology Bony anomalies Abnormal first rib Long C7 transverse process
No. of patients
The operative complications, length ot hospital stay, and results of the operation were compared between the rib-resection group and the rib-sparing group. Standard x2 analysis and Student’s t test were used to determine the statistical significance. For patients with bilateral operations, in order to eliminate in-sample dependence, only one side was chosen for statistical analysis (decided by random numbers generated from the computer). The results of thoracic outlet syndrome surgery tend to worsen over time. Early results were evaluated at the first follow-up visit after the patients had finished an adequate course of postoperative physical therapy and attained a stable state, usually from 3-4 months after surgery. Long-term results were reported using the life-table method, and the differences between the rib-resection group and rib-sparing group compared by the log-rank test and the Lee-Desu statistic”. A variety of patient factors were considered possible predictors for successafter the operation. These include the patient’s age, sex, history of trauma, nature of trauma, time interval of trauma to surgery and duration of symptoms. These were analysed using the stepwise logistic regression model, allowing for a dichotomous dependent variable (outcome), in order to identify the significant factor(s) which determines the operative results.
Results Complications
2 25 101 50 66 54
Operative complications are listed in T&e .3. The most frequent complication was injury to the pleura. Usually the pleural defect was small, and the rent could be easily repaired with direct suturing and by evacuating the air using a small catheter through the pleurotomy before wound closure. A significantly higher incidence of intraoperative pleural injury was observed with first rib resection (59%) than with the rib-sparing operation
CAf?DlOVASCULAR SURGERY DECEMBER IQQS VOL 3 NO 6
619
Soft tissue anomalies Anterior and middle scalene anomalies Scalenus minimus Fibrous bands Scarring *Patients with cervical ribs are not included in this study
Neurogenic thoracic outlet decompression: S. W K Cheng et al. Table 3 Complications of 166 primary supraclavicular thoracic outlet syndrome operations Scalenectomy and first ;i”z;e&on
n Pleural complications lntraoperative pleural injury Pneumothorax (postoperative) Pleural effusions/atelectasis
74(59) 14( 11) 14(i 1)
Scalenectomy (rib sparing) (n==43)
17(40) 3(7) 1(Z)
< 0.05 n.s. n.s.
5(4) 6(5) l(1) 7(6)
g(5)
n.s. n.s. n.s. n.s.
Lymphatic injuries Wound infection Hemorrhage Re-exploration
5(4) 263 4(3) 5(4)*
2(5) 1(Z) 0 l(2)’
n.s. n.s. ns. n.s.
Values in parentheses are percentages: ns., not statistically significant (P> 0.05); *three for lymphatic leaks in the neck, one for bleeding, one for recurrent scarring; ‘for lymphatic IeaWchylothorax
(40%, P < 0.05, x2 test). Postoperative pneumothorax was rare, and when it did occur, it was usually managed conservatively. Small asymptomatic collections in the pleural space were sometimes evident in these patients on postoperative chest radiography. There were no major vascular injuries or brachial plexus injury in all patients undergoing primary supraclavicular thoracic outlet decompressions. Paresis of the long thoracic nerve was a rare event. There was a small risk of lymphatic injury which may manifest as a chylous collection in the neck, chylothorax, or a lymphatic fistula. Four patients had to be re-explored for lymphatic leakage. One patient was reoperated upon for bleeding which originated from a small arterial branch. There was no operative mortality. Only one patient who had a supraclavicular first rib resection developed severe recurrent symptoms and had to be re-explored 17 months later. Dense scarring around the brachial plexus was discovered and a thorough neurolysis was performed. Overall, there was no statistically significant differences in these other complications between the two groups. Hospital stay and results 4 lists the early operative results for the two operations. The mean hospital stay was prolonged by 1 day in the rib-resection group (5.1 days) when compared with the rib-sparing group (3.9 days). This difference was statistically significant (P = 0.005, Student’s t test). There was rio significant difference in the clinical outcome of the two groups of patients (Table 4). The Table
620
Hospitalization and early results Scalenectomy and first rib resection (n = 125)
Scalenectomy (rib sparing) (n = 43)
P
5.1(2-19)
3.9(2-12)
0.005
13(1-20)
7( l-37)
P
Nerve injuries Phrenic nerve Long thoracic nerve Recurrent laryngeal nerve Horner’s syndrome
g(5)
Table 4
Mean (range) stay in hospital (days) Mean (range) follow-up time (months) Return to work (%) Early results Cured (%) Markedly improved (%) Improved (%) Fair (%) Worse (%) Overall success (%)*
52
63
18 38 27 16 1 83
27 35 29 6
n.s.
n.s.
n.s.. not statistically significant (P> 0.05): *success is defined as > 50% symptomatic improvement
overall early successwas 83 % in the rib-resection group and 91% in the rib-sparing group. There was again no significant difference between the two groups in the proportion of patients that can resume employment. The long-term results of the two operations as determined by life-table analysis is shown in Figure 2. The cumulative successrate at 2 years was 69% for the rib-resection group and 76% for the rib-sparing group. The difference was not statistically significant. Factors affecting outcome Of all the patient factors, only the duration of symptoms was found to be significant on stepwise logistic regression analysis (Table 5). The patients’ age, sex, and the history and nature of trauma did not affect the final outcome. Based on this finding, the patients have been classified into those who presented early (with symptom durations of 2 years or less) and those who presented
2
70 -
: ‘S m
50 -
:
E’
‘\ --em__.
*\\
60-
40 30
’ 3
-
Rib
spared
----
Rib
resected
’ 9
’ 12
’ 6
’ 15
Follow-up
’ 18
’ 21
’ 24
’ 27
’ 30
(months)
Figure 2 Results of 166 primary supraclavicular operations for thoracic outlet syndrome. There was no statistically significant difference (P> 0.05) between the rib resection (n = 125) and rib-sparing (n = 43) groups
CARDIOVASCULAR SURGERY DECEMBER 1995 VOL 3 NO 6
Neurogenic Table S
Factors affecting outcome: stepwise logistic regression
Factor
P
-----Sex Age History of trauma Nature of trauma Time of trauma to presentation Duration of symptoms
0.23 0.10 0.21 0.06 0.23 0.01*
*Statistically significant difference
Cl Success
do -
70
82 2u
50
f3 Failure
60 40 301
f /
24 101 OL
:.:.:.:.:.:.y::::: ............. j::::j::::j::::.,: :::::::::::::::::::: :::::::::::I$:::::: :::::::::::>.::::::: .:.:.:.:.:.>:.:.:.: .................... .................. . i:;:>;:i:;:~:;:::
i
-
Symptoms
S 2 yr
Symptoms
> 2 yr
J
Figure3 The effect of duration of symptoms on the outcome of thoracic outlet decompresslon. Patients with symptom durations of 2 years or less have a significantly better outcome than patients with symptoms of more than 2 years (PC 3.05!
late (more than 2 years). The overall success rate of the former was 83% which is significantly better than the 68% of the latter group (Figure 3, P < 0.05, x2 test).
Discussion Thoracic outlet syndrome has evolved from a vascular compression by bony structures in the thoracic outlet to one of neurological compression and irritation over the past 50 years. Clagett4 first proposed excision of the first rib in treating all these conditions, using a posterior approach. Later with the introduction of the transaxillary operation by Roosl, removing the first rib has been the method of choice for most surgeons dealing with this problem. It has never been established whether the rib or the soft tissue elements of the thoracic outlet are the principal factors causing mechanical irritation of the brachial plexus. In fact from most large series’ soft tissue anomalies were more often responsible for neurological compression and an anatomically abnormal first rib was the exception. Stallworth and colleagues6 performed transaxillary thoracic outlet syndrome decompressions by removing soft tissues alone without rib resection and claimed that the first rib
CARDIOVASCUIJIR SURGERY DECEMBER 1995 VOL 3 NO 6
thoracic outlet decompes9ofi:
2%VG:K Lkeng et al.
seldom caused irritation of the brachial plexus. As the complications associated with the transaxiltary rib resection became a matter of increasing concern’: a search was made by some surgeon*< for a lesser procedure that might be equally effective. Sanders and co-workers* compared supraclavicul,.lr \calenectomy with transaxillary rib excision and later with supraclavicular first rib resection” and found r~; diifcrrnces in results of the three operations. Originally thlese authors adopted rive transaxillnry approach to remove the first thoraclc r~.b, rather than precisely expose the brachial plexus itseit. A significant proportlon of tbese patients had persistem or rccwre~:t symptoms. Supraclavicular reoperation in rhese cases disclosed reattachment of the anterior :;wiene muscle to the scarred first rib bed, as well as other myofascial abnormalities. Their symptoms mvan5gbl> rmproved after scalenectomy. With this experience, anti as the role of the scalene muscles became evident, supraclavicuiar scalenectomy was combined with transaxillary first rib resectionl”. Ln 1983 the transaxilaary route was abandoned entirely but first rib resections ctrnritrued ttr be performed via the supraclavicuiar route, achieving improved results 2. The supraclavicular *rpproach pro,vides a view unparalleled by the transax&~ry or anai other route. It allows direct visualizarior1 of the exacr anatomical relationship between the i>c~~y and myofascial structures and rhe brachiaii plexus, This permits precise identification of offending pattmlog~ and J complete decompression of the pilexus from thcx intervertebral foramina to the clavicle. With increasing experience with the supraclavicuiar procedure and :I better understanding of the patholop offered by thr superior exposure, the necessity of routine first rib resection was first questioned. During most operations it was apparent that the first rib U’M IX\: irk diresr contact with the brachial plexus, ~iiici po:,ed ~I(J obstruction to rhe course of the ner~~e roots. The offending structures identified during supr;tclnvicular decompression were usually soft tissue &momalies. particularly muscular interdigitations between. the scalenes, myofascial bands, and perineum1 scarring. It 1s believed that the first rib has probably cml\ .I passivt role in thoracic outlet syndrome: it pro\.ides a matrix upon which myofascial elements adhere, and it is tbr contraction of these elements that primarily gust: nerve compression and elevation of the first rub. Based on this observation the first rib *as frrst spared during thoracic outlet decompression procedures in 1986. The operation can be modified if the medial neck of the first rib displaces the course of the Tl z:erve root by removing a small rim of the medial border 05 rhe rib with a bone rongeur, thus maintaining most of the integrity of the rib but at the same time allowing an unrestricted course of the Tl root. With this modification (sagittal osteotomy), a complete extraperiosteal excision of the first rib was nor deemed necessary. Supraclavicular scalenectomy and neur&s$< has be-
621
Neurogenicthoracicoutletdecompression: S lb!K Chenget al. come the authors’ procedure of choice for decompression of neurogenic thoracic outlet syndrome in patients with no gross bony anomalies. Resection of the first rib poses additional problems of increased postoperative pain and shoulder immobility which may prolong hospitalization and delay recovery of function. The undersurface of the rib is in close contact with the dome of the pleura, and a significantly higher incidence of pleural disruption has been observed during first rib resection. Although most of the pleural rents were minor and readily repaired, some did manifest as postoperative pneumothorax or pleural effusions which add to patient morbidity by compromising respiration. These patients also had more postoperative complaints of pain with the more extensive procedures and as a result their hospital stay was prolonged by 1 day. The other complications were not significantly increased, as they were unrelated to the removal of the rib. Furthermore, removal of the first rib creates a fibrous bed which is often a site of excessive scarring and may contribute to late recurrences. The rib-resection group did have a slightly inferior shortand long-term result than the rib-sparing group, although the differences were not statistically significant. So far, after up to 3 years of follow-up, none of the 43 patients who had rib-sparing operations required a second procedure for recurrent or persistent symptoms. One of the drawbacks of this study is that the patients were not randomized. Nevertheless the two groups of patients were comparable in sex and age distribution, and the history and physical findings were all similar. It would be justifiable to assume they came from the same patient population. All 168 procedures were performed by the same surgeon in one single hospital experienced in treating thoracic outlet syndrome, so the perioperative care would be identical. Unlike previous studies’ it was chosen not to include patients who had cervical ribs in this review. These patients constitute a separate entity where a definite bony abnormality can be shown to compress the brachial plexus and once the cervical rib is removed, first rib resection is usually not indicated. A better operative result has also been reported for this group’ which may constitute a bias when compared with rib-resection patients. Assessing the postoperative results of neurogenic thoracic outlet syndrome decompression is equally difficult because there is no objective scale of measuring success. The authors have tried to be as objective as possible by employing a scoring system based on the patient’s evaluation of percentage improvement of preoperative symptoms. Since many of the patients were involved in worker’s compensation olicies, and this has been shown to affect the outcomei !i , it was elected not to include the ability to return to work as a true reflection of operative results in the statistical analysis. The benefits of the thoracic outlet syndrome procedure tend to decrease with time as new scar formation occurs around the brachial plexus. Results have been
622
expressed both as early success rates and also by life-table methods across a longer period of follow-up. Resecting the first rib did not prove to be an advantage both on short-term and long-term assessments. Using this system a long-term failure rate of 20 to 30% at 2 years is reported. This may be high when compared with some series’> 9, but only those with more than 50% improvement as operative success have been included, and the ‘fair’ group has been classified as failures. The significance of the soft tissue anomalies is well demonstrated by the work of Roes’ who classified neurogenic thoracic outlet syndrome into upper and lower plexus groups and advocated supraclavicular scalenectomy for patients with upper plexus symptoms. In the authors’ experience, the patients usually presented with a mixture of symptoms and it has been impossible to distinguish them into clear upper and lower plexus groups. It is believed that this classification is unimportant as long as total decompression of the nerves of the brachial plexus is performed during the operation. This can be achieved by anterior and middle scalenectomy alone. The importance of the scalene muscles is further evident by recent reported observations of type I fiber predominance, type II atrophy and endomysial fibrosis 12,13. The exact significance of this histological pattern in pathogenesis is unknown and further control studies of scalene morphology will be necessary. The only significant factor that determined clinical outcomes was the duration of symptoms before surgery. Patients who were operated upon within 2 years of occurrence of symptoms achieved a significantly better outcome. This led the authors to believe that early diagnosis and decompression of neurogenic thoracic outlet syndrome are crucial for success. Because of the lack of physical signs and an objective diagnostic test, some had questioned the existence of neurogenic thoracic outlet syndrome, and advocated surgical decompression only when there is undisputed weakness and wasting of the hand muscles14. Unnecessarily prolonged conservative management or repeated referrals and neuroelectric studies lead only to a waste of resources and patient discomfort. Delay in treatment until neurological signs develop may result in permanent nerve damage and preclude the chance of a complete recovery. The actual neurological compression or irritation, be it bony or muscular, may not be as important as the degree of damage sustained by the nerve itself in determining the final outcome. The ideal decompressive operation for thoracic outlet syndrome should be aimed at restoring an optimal environment for the injured nerves to heal by a thorough neurolysis, and as early as possible, rather than focused on removing the skeletal boundaries of the thoracic outlet (first thoracic rib). It is concluded that in patients with neurogenic thoracic outlet syndrome, resecting the first rib adds no advantage in both short-term and long-term assessment CARDIOVASCULAR SURGERY DECEMBER 1995 VOL 3 NO 6
Neurogenicthorxic outlet decompresskz?: .s’ kti K Chenget al. of the patients’ outcome. In the authors’ experience it caused unnecessary morbidity and a longer hospital stay. Supraclavicular anterior and middle scalenectomy is recommended for patients with neurogenic thoracic outlet syndrome requiring operative intervention.
Acknowledgements These studies were supported Vascular Research Foundation.
in part by the Pacific
References 1. 2. 3. 4. 5. 6.
Koos DB. Transaxillary approach for first rib resection to relieve thoracic outlet syndrome. Ann Surg 1966; 163: 354-8. Reilly LM, Sroney RJ. Supraclavicular approach for thoracic outlet decompression. J Vusc Surg 1988; 8: 329-34. Lee E, Desu M. A computer program for comparing k samples with right censored data. Computer Progrums in Biomedicine 1972; 2: 315-21. Clagett OT Presidential address: research and prosearch. J Thorac Cardiouasc Surg 1962; 44: 153-66. Roos DB. The place for scalenectomy and first rib resection in thoracic oudet syndrome. Surgery 1982; 92: 1077-85. Stallworth JM, Quinn GJ, Aiken AF. Is rib resection necessary
CARDIOVASCULAR
SURGERY
DECEMBER
199s VOL 3 NO 6
for relief of thoracic outlet syndrome? Ilr,r~ 5tq !o’Y; l&S: 581-92. 7. Dale WA. Thoracic outlet compression synirom::. i rmque IX 1982. Arch Surg 1982; 117: 1437-45. 8. Sanders RJ, Monsour JW, Gerber WF, Adams WR, Thompson N. Scalenectomy vc~s’susfirst rib resection for treatment of the thoracic outlet syndrome. Surgery 1979; 85: ! 09-21. 9. Sanders RJ, Pearce WH. The treatment 11f rhoracic outlet [ tltisz .SUY~ syndrome: a comparison of different operatrons. 1989; 10: 626--34. 10. Qvarfordt PC, Ehrenfeld WK, Stoney KJ. Supraclavicular radial scalenectomy and transaxillary first rib resecrinn for the thoracic outlet syndrome: a combined approach. Am f Cq 1984; 148: 111-16. 11. Green RM, McNamara J, Ouriel K. Long term follow-up after thoracic outlet decompression: an analysis of factors determin ing outcome. J Vast Surg 1991; 14: 73%+h. 12. Machleder HI, Mall F, Verity A. The anterior ,caIene muscle in thoracic outlet compression syndrome. Wistochemical and morphometric studies: Arch SUY~ 1986; 121: I 141-4. 13. Sanders RJ, Jackson CGR, Banchero N, Pearce WH. Scalene muscle abnormalities in traumatic thoracic outlet syndrome. Am J Surg 1990; 1.59: 23176. 14. Wilbourn AJ. The thoracic outlet syndrom a~ ‘averdla,gxJlosed. Arch Nemo/ 1990; 47: 328-W.
Paper accepted 5 December 1994
VASCULAR PAPERS
Surgery, Vol. 3, No. b. pp. hll-623, 1095 Copyright @ 1995 Blsevier Science Ltd Printed in Great Britain. All rights reserved OYh7--2lIWY5 $lO.OO + 0.w
Neurogenicthoracic outlet decompression: rationale for sparing the first rib __-.._-.II .-..._-___ -.._. --S. W. K. Cheng*, L. M. Reilly, N. A. Nelken, W. V. Ellis and R. J. Stoney ReparTmentofSurgery, Divisionof VascularSurgery, M-488 Un~ve~i~/ofCalifornia,505Parn~-~(~u~ Avenue,SanFranckco, CA 94143 USA
A total of 168 primary supraclaviculardecompressionswere performed on 146 patients with neurogenic thoracic outlet syndrome. This report compares the results of rib resection (supraclavicular anterior and middle scalenectomy and first rib resection) with rib-sparing (supraclavicularanterior and middle scalenectomyalone) operations. All patients with cervical ribs were excluded.In total, 125 rib resections and 43 rib-sparing procedures were performed between 1983 and 1992 by a single surgeon. The patients were otherwise comparable in symptoms and physical signs. During surgery there was a significantly higher proportion of pleural injury associatedwith rib resection (59%) than with rib-sparing (40%) procedures, The mean hospital stay was also prolonged by 1 day in patients undergoing rib resection (P= 0.005). There was no significant difference in early successbetween the two groups (83% for rib resection, 91% for rib sparing) and no difference in those resuming employment (52% and 63% respectively).Life-table analysisshowed that the two groups have timPar long-term results (69% and 76% at 2 years). The only important factor determining clinical outcome in primary supraclavicular thoracic outlet syndrome decompression was the duration of symptoms before operation. Some 83% of patients with symptoms less than 2 years had a successful result compared with only 68% in those with symptoms longer than 2 years (PC 0.05). Spontaneous or post-traumatic neurogenic symptoms responded to operation identically.The theoretical benefit of first rib resectionto relievemechanicalcompressionof the brachial plexus is not evident from this review. Thorough removal of the scalene musculature and other myofascialanomalies,preferably through the supraclavicularapproach, leads to less patient morbidity, shortens hospitalization,and is recommended for patients with neurogenic thoracic outlet syndrome requiring operative intervention. -___
Keywords: thoracic outlet syndrome, supraclavicular. rib sparing _- ..__._.. -_
Mechanical compression of the brachial plexus is presumed to initiate the neurogenic symptoms of thoracic outlet syndrome. It has never been established whether the bony (first rib) or the muscular (scalene muscles)elementsof the thoracic outlet are the principal factors contributing to pathogenesis.Much controversy still exists concerning the ideal operation for this group of patients. Removing the first thoracic rib via the
Correspondence to: Dr R. J. Stoney *Present address: Department of Surgery, Queen Mary University of Hong Kong, Hong Kong.
CARDIOVASClJL4R SURGERY DECEMBER IQ95 VOL 3 NO 6
Hospital,
----
--.-..-1-.._ .._.-. -.____-__.___
transaxillary route’ is still the most popular operation. The authors have employed the supradavicular approach because of irs superior exposure, safety and consistently good results2. As experience was gained with the supraclavicular operation, it was observed that the first rib seldom caused direct brachial plexus irritation and its removal was therefore unnecessary. In an attempt to decreasepatient morbidity, the operation has since been modified such that the first thoracic rib is now preserved. This article reports the results of 168 consecutive primary supraclavicular rhoracic outlet decompression procedures performed at the University of California Medical Center, San Francisco, and compares the morbidity and clinical outcome of rib-resecting and rib-sparing operations, 617
Neurogenic thoracic outlet decompression: .S W K Cheng et al.
Patients and methods From 1983 to 1992, 168 primary operations were performed by a single surgeon on 146 patients for neurogenic thoracic outlet syndrome, using the supraclavicular approach. All patients with cervical ribs and those who had previous thoracic outlet decompressions were excluded from this review. There were 117 women (80%) and 29 men (20%); their mean age was 36 years (range 18-60) years. The procedure was carried out for the left extremity in 87 instances (52%); 22 patients had bilateral supraclavicular decompressions, while 124 were unilateral operations. In total, 125 procedures performed between 1983 and 1990 involved anterior and middle scalenectomy and first rib resection (rib-resection group); 43 procedures performed between 1986 and 1992 were supraclavicular anterior and middle scalenectomies with preservation of the first rib (rib-sparing group). The etiology, symptoms and physical findings of the two groups of patients were comparable with no statistically significant difference (Table 1). A history of trauma was reported in 115 cases (68%). Most of the patients had work-related injuries, with flexion-extension injuries of the neck sustained in traffic accidents preceding symptom occurrence in 14% of cases. The patients were usually treated by a period of physical therapy, including exercises and nerve blocks, before they were referred to the authors’ service. The mean duration of symptoms from onset to surgery was 3 years (range 2 months to 25 years) (Table 1). Pain involving the lateral neck and shoulder, and paresthesia radiating to the arms and hands remained the dominant symptoms and were almost universal. Unilateral headache and motor weakness were also common. The symptoms were usually severe and disrupted work (89%) or demanded constant analgesic medications (70%). Most patients had symptom distribution in the dermatomes of all five roots of the brachial plexus, and upper and lower plexus groups could not be clearly distinguished one from the other. The most consistent physical finding was supraclavicular tenderness, present in 88% of all patients. The elevated arm stress test (EAST) was positive in 86% of cases. The Adson test was not found to be reliable, being positive in about two-thirds of the cases. The diagnosis was usually made clinically, based on a characteristic history and classical symptoms and physical signs. Preoperative evaluation included radiography of the chest and cervical spine. A neurological consultation was made to rule out alternative diagnosis. Nerve conduction studies were only performed when the diagnosis of carpel tunnel syndrome was suspected and were abnormal only in 6% of the present cases. Computed tomography and magnetic resonance imaging scans of the neck have been used occasionally to exclude cervical spine pathology. Patients with coexisting vascular compression were excluded from this study.
618
Table 1 History and presentations of 168 cases of neurogenic thoracic outlet syndrome Scalenectomy and first rib resection
Scalenectomy (rib sparing)
All supraclavicular operations
125
43
168
History of trauma (%)
73
56
68
Whiplash neck injuries w Blunt neck trauma (96) Lifting (%) Others (%)
14
14
14
34 12 13
20 4 18
31 10 14
Mean duration of symptoms (months)
39
31
37
Pain (%) Paresthesia (%) Motor weakness (96) Unilateral headache cw
100 95 21
100 100
100 96
19 35
20
38
Work affected by symptoms (%) Cannot work (96) Constant analgesics required (%)
22
33
25
69
51 67
64
70
Supraclavicular tenderness (%) Surpaclavicular mass cw Reduced range of neck motion (%) Positive Adson’s test (W Positive elevated arm stress test (%) Motor deficit or atrophy (%)
87
88
88
2
7
4
31
19
28
66
70
67
84
91
86
9
5
8
Previous physical therapy failed (%) Previous nerve blocks failed (%) Previous neck surgery failed (96)
87
88
87
13
5
11
6
9
No. of operations
38
70
Operative technique and findings All 168 procedures were performed via the supraclavicular approach. The technique of supraclavicular thoracic outlet decompression has been described previously2. This involved anterior and middle scalenectomy followed by a thorough neurolysis of the brachial plexus. First rib resection was performed from 1983 to mid-1990. Afterwards the operation was modified by preserving the first rib after neurolysis of the brachial plexus. Occasionally the medial border of the neck of the first rib will displace the course of the first thoracic nerve root. A bone rongeur may be used to remove a small rim of the medial border of the rib, thus allowing an unrestricted space for the Tl root to emerge from the
CARDIOVASCULAR SURGERY DECEMBER 1995 VOL 3 NO 6
Neurogenicthoracic outlet decomptwsion.5. B! K. Chenget al.
Patient follow-up and assessment The patients underwent a supervised physical therapy program for 8 weeks after discharge. They were reassessedat regular intervals postoperatively. Those who were referred from outside the local area and who had achieved a stable result were also sent a detailed questionnaire for their progress. The results of opera tion were graded using a scoring system based on the percentage improvement of the preoperative symproms’. If all of the syrnptoms were consistently relieved, the patients were considered cured. If there was a > 75% improvement of all symptoms, the patients were classified as markedly improved. An improvement of 50-75% was graded improved and ~50% fair or worse. A <5O% improvement of qymptoms was considered a failure of the operation. Statistical methods b
C
Figure 1 a Illustrated detail showing removal of a rim of the medial border of the first rib using a bone rongeur. The middle scalene is transected in a line parallel to and inferior to the long thoracic nerve. b First rib after segmental osteotomy. c The Tl root runs an unrestricted course after segmental osteotomy of the first rib
intervertebral foramina and join the C8 root to form the lower trunk of the brachial plexus (Figure 1). A rib-sparing decompression has become the authors’ standard operation for neurogenic thoracic outlet syndrome. The operative findings are listed in Table 2. Most of the offending structures were soft tissue anomalies which could only be adequately exposed and removed via the supraclavicular approach. These included abnormally placed anterior and middle scalene insertions, muscular interdigitations, a scalenus minimus muscle, or fibrous bands. Bony anomalies were uncommon and mostly consisted of long transverse processesof the C7 vertebra. Table 2 Pathology encountered in 168 primary supraclavicular thoracic outlet syndrome operations* Pathology Bony anomalies Abnormal first rib Long C7 transverse process
No. of patients
The operative complications, length ot hospital stay, and results of the operation were compared between the rib-resection group and the rib-sparing group. Standard x2 analysis and Student’s t test were used to determine the statistical significance. For patients with bilateral operations, in order to eliminate in-sample dependence, only one side was chosen for statistical analysis (decided by random numbers generated from the computer). The results of thoracic outlet syndrome surgery tend to worsen over time. Early results were evaluated at the first follow-up visit after the patients had finished an adequate course of postoperative physical therapy and attained a stable state, usually from 3-4 months after surgery. Long-term results were reported using the life-table method, and the differences between the rib-resection group and rib-sparing group compared by the log-rank test and the Lee-Desu statistic”. A variety of patient factors were considered possible predictors for successafter the operation. These include the patient’s age, sex, history of trauma, nature of trauma, time interval of trauma to surgery and duration of symptoms. These were analysed using the stepwise logistic regression model, allowing for a dichotomous dependent variable (outcome), in order to identify the significant factor(s) which determines the operative results.
Results Complications
2 25 101 50 66 54
Operative complications are listed in T&e .3. The most frequent complication was injury to the pleura. Usually the pleural defect was small, and the rent could be easily repaired with direct suturing and by evacuating the air using a small catheter through the pleurotomy before wound closure. A significantly higher incidence of intraoperative pleural injury was observed with first rib resection (59%) than with the rib-sparing operation
CAf?DlOVASCULAR SURGERY DECEMBER IQQS VOL 3 NO 6
619
Soft tissue anomalies Anterior and middle scalene anomalies Scalenus minimus Fibrous bands Scarring *Patients with cervical ribs are not included in this study
Neurogenic thoracic outlet decompression: S. W K Cheng et al. Table 3 Complications of 166 primary supraclavicular thoracic outlet syndrome operations Scalenectomy and first ;i”z;e&on
n Pleural complications lntraoperative pleural injury Pneumothorax (postoperative) Pleural effusions/atelectasis
74(59) 14( 11) 14(i 1)
Scalenectomy (rib sparing) (n==43)
17(40) 3(7) 1(Z)
< 0.05 n.s. n.s.
5(4) 6(5) l(1) 7(6)
g(5)
n.s. n.s. n.s. n.s.
Lymphatic injuries Wound infection Hemorrhage Re-exploration
5(4) 263 4(3) 5(4)*
2(5) 1(Z) 0 l(2)’
n.s. n.s. ns. n.s.
Values in parentheses are percentages: ns., not statistically significant (P> 0.05); *three for lymphatic leaks in the neck, one for bleeding, one for recurrent scarring; ‘for lymphatic IeaWchylothorax
(40%, P < 0.05, x2 test). Postoperative pneumothorax was rare, and when it did occur, it was usually managed conservatively. Small asymptomatic collections in the pleural space were sometimes evident in these patients on postoperative chest radiography. There were no major vascular injuries or brachial plexus injury in all patients undergoing primary supraclavicular thoracic outlet decompressions. Paresis of the long thoracic nerve was a rare event. There was a small risk of lymphatic injury which may manifest as a chylous collection in the neck, chylothorax, or a lymphatic fistula. Four patients had to be re-explored for lymphatic leakage. One patient was reoperated upon for bleeding which originated from a small arterial branch. There was no operative mortality. Only one patient who had a supraclavicular first rib resection developed severe recurrent symptoms and had to be re-explored 17 months later. Dense scarring around the brachial plexus was discovered and a thorough neurolysis was performed. Overall, there was no statistically significant differences in these other complications between the two groups. Hospital stay and results 4 lists the early operative results for the two operations. The mean hospital stay was prolonged by 1 day in the rib-resection group (5.1 days) when compared with the rib-sparing group (3.9 days). This difference was statistically significant (P = 0.005, Student’s t test). There was rio significant difference in the clinical outcome of the two groups of patients (Table 4). The Table
620
Hospitalization and early results Scalenectomy and first rib resection (n = 125)
Scalenectomy (rib sparing) (n = 43)
P
5.1(2-19)
3.9(2-12)
0.005
13(1-20)
7( l-37)
P
Nerve injuries Phrenic nerve Long thoracic nerve Recurrent laryngeal nerve Horner’s syndrome
g(5)
Table 4
Mean (range) stay in hospital (days) Mean (range) follow-up time (months) Return to work (%) Early results Cured (%) Markedly improved (%) Improved (%) Fair (%) Worse (%) Overall success (%)*
52
63
18 38 27 16 1 83
27 35 29 6
n.s.
n.s.
n.s.. not statistically significant (P> 0.05): *success is defined as > 50% symptomatic improvement
overall early successwas 83 % in the rib-resection group and 91% in the rib-sparing group. There was again no significant difference between the two groups in the proportion of patients that can resume employment. The long-term results of the two operations as determined by life-table analysis is shown in Figure 2. The cumulative successrate at 2 years was 69% for the rib-resection group and 76% for the rib-sparing group. The difference was not statistically significant. Factors affecting outcome Of all the patient factors, only the duration of symptoms was found to be significant on stepwise logistic regression analysis (Table 5). The patients’ age, sex, and the history and nature of trauma did not affect the final outcome. Based on this finding, the patients have been classified into those who presented early (with symptom durations of 2 years or less) and those who presented
2
70 -
: ‘S m
50 -
:
E’
‘\ --em__.
*\\
60-
40 30
’ 3
-
Rib
spared
----
Rib
resected
’ 9
’ 12
’ 6
’ 15
Follow-up
’ 18
’ 21
’ 24
’ 27
’ 30
(months)
Figure 2 Results of 166 primary supraclavicular operations for thoracic outlet syndrome. There was no statistically significant difference (P> 0.05) between the rib resection (n = 125) and rib-sparing (n = 43) groups
CARDIOVASCULAR SURGERY DECEMBER 1995 VOL 3 NO 6
Neurogenic Table S
Factors affecting outcome: stepwise logistic regression
Factor
P
-----Sex Age History of trauma Nature of trauma Time of trauma to presentation Duration of symptoms
0.23 0.10 0.21 0.06 0.23 0.01*
*Statistically significant difference
Cl Success
do -
70
82 2u
50
f3 Failure
60 40 301
f /
24 101 OL
:.:.:.:.:.:.y::::: ............. j::::j::::j::::.,: :::::::::::::::::::: :::::::::::I$:::::: :::::::::::>.::::::: .:.:.:.:.:.>:.:.:.: .................... .................. . i:;:>;:i:;:~:;:::
i
-
Symptoms
S 2 yr
Symptoms
> 2 yr
J
Figure3 The effect of duration of symptoms on the outcome of thoracic outlet decompresslon. Patients with symptom durations of 2 years or less have a significantly better outcome than patients with symptoms of more than 2 years (PC 3.05!
late (more than 2 years). The overall success rate of the former was 83% which is significantly better than the 68% of the latter group (Figure 3, P < 0.05, x2 test).
Discussion Thoracic outlet syndrome has evolved from a vascular compression by bony structures in the thoracic outlet to one of neurological compression and irritation over the past 50 years. Clagett4 first proposed excision of the first rib in treating all these conditions, using a posterior approach. Later with the introduction of the transaxillary operation by Roosl, removing the first rib has been the method of choice for most surgeons dealing with this problem. It has never been established whether the rib or the soft tissue elements of the thoracic outlet are the principal factors causing mechanical irritation of the brachial plexus. In fact from most large series’ soft tissue anomalies were more often responsible for neurological compression and an anatomically abnormal first rib was the exception. Stallworth and colleagues6 performed transaxillary thoracic outlet syndrome decompressions by removing soft tissues alone without rib resection and claimed that the first rib
CARDIOVASCUIJIR SURGERY DECEMBER 1995 VOL 3 NO 6
thoracic outlet decompes9ofi:
2%VG:K Lkeng et al.
seldom caused irritation of the brachial plexus. As the complications associated with the transaxiltary rib resection became a matter of increasing concern’: a search was made by some surgeon*< for a lesser procedure that might be equally effective. Sanders and co-workers* compared supraclavicul,.lr \calenectomy with transaxillary rib excision and later with supraclavicular first rib resection” and found r~; diifcrrnces in results of the three operations. Originally thlese authors adopted rive transaxillnry approach to remove the first thoraclc r~.b, rather than precisely expose the brachial plexus itseit. A significant proportlon of tbese patients had persistem or rccwre~:t symptoms. Supraclavicular reoperation in rhese cases disclosed reattachment of the anterior :;wiene muscle to the scarred first rib bed, as well as other myofascial abnormalities. Their symptoms mvan5gbl> rmproved after scalenectomy. With this experience, anti as the role of the scalene muscles became evident, supraclavicuiar scalenectomy was combined with transaxillary first rib resectionl”. Ln 1983 the transaxilaary route was abandoned entirely but first rib resections ctrnritrued ttr be performed via the supraclavicuiar route, achieving improved results 2. The supraclavicular *rpproach pro,vides a view unparalleled by the transax&~ry or anai other route. It allows direct visualizarior1 of the exacr anatomical relationship between the i>c~~y and myofascial structures and rhe brachiaii plexus, This permits precise identification of offending pattmlog~ and J complete decompression of the pilexus from thcx intervertebral foramina to the clavicle. With increasing experience with the supraclavicuiar procedure and :I better understanding of the patholop offered by thr superior exposure, the necessity of routine first rib resection was first questioned. During most operations it was apparent that the first rib U’M IX\: irk diresr contact with the brachial plexus, ~iiici po:,ed ~I(J obstruction to rhe course of the ner~~e roots. The offending structures identified during supr;tclnvicular decompression were usually soft tissue &momalies. particularly muscular interdigitations between. the scalenes, myofascial bands, and perineum1 scarring. It 1s believed that the first rib has probably cml\ .I passivt role in thoracic outlet syndrome: it pro\.ides a matrix upon which myofascial elements adhere, and it is tbr contraction of these elements that primarily gust: nerve compression and elevation of the first rub. Based on this observation the first rib *as frrst spared during thoracic outlet decompression procedures in 1986. The operation can be modified if the medial neck of the first rib displaces the course of the Tl z:erve root by removing a small rim of the medial border 05 rhe rib with a bone rongeur, thus maintaining most of the integrity of the rib but at the same time allowing an unrestricted course of the Tl root. With this modification (sagittal osteotomy), a complete extraperiosteal excision of the first rib was nor deemed necessary. Supraclavicular scalenectomy and neur&s$< has be-
621
Neurogenicthoracicoutletdecompression: S lb!K Chenget al. come the authors’ procedure of choice for decompression of neurogenic thoracic outlet syndrome in patients with no gross bony anomalies. Resection of the first rib poses additional problems of increased postoperative pain and shoulder immobility which may prolong hospitalization and delay recovery of function. The undersurface of the rib is in close contact with the dome of the pleura, and a significantly higher incidence of pleural disruption has been observed during first rib resection. Although most of the pleural rents were minor and readily repaired, some did manifest as postoperative pneumothorax or pleural effusions which add to patient morbidity by compromising respiration. These patients also had more postoperative complaints of pain with the more extensive procedures and as a result their hospital stay was prolonged by 1 day. The other complications were not significantly increased, as they were unrelated to the removal of the rib. Furthermore, removal of the first rib creates a fibrous bed which is often a site of excessive scarring and may contribute to late recurrences. The rib-resection group did have a slightly inferior shortand long-term result than the rib-sparing group, although the differences were not statistically significant. So far, after up to 3 years of follow-up, none of the 43 patients who had rib-sparing operations required a second procedure for recurrent or persistent symptoms. One of the drawbacks of this study is that the patients were not randomized. Nevertheless the two groups of patients were comparable in sex and age distribution, and the history and physical findings were all similar. It would be justifiable to assume they came from the same patient population. All 168 procedures were performed by the same surgeon in one single hospital experienced in treating thoracic outlet syndrome, so the perioperative care would be identical. Unlike previous studies’ it was chosen not to include patients who had cervical ribs in this review. These patients constitute a separate entity where a definite bony abnormality can be shown to compress the brachial plexus and once the cervical rib is removed, first rib resection is usually not indicated. A better operative result has also been reported for this group’ which may constitute a bias when compared with rib-resection patients. Assessing the postoperative results of neurogenic thoracic outlet syndrome decompression is equally difficult because there is no objective scale of measuring success. The authors have tried to be as objective as possible by employing a scoring system based on the patient’s evaluation of percentage improvement of preoperative symptoms. Since many of the patients were involved in worker’s compensation olicies, and this has been shown to affect the outcomei !i , it was elected not to include the ability to return to work as a true reflection of operative results in the statistical analysis. The benefits of the thoracic outlet syndrome procedure tend to decrease with time as new scar formation occurs around the brachial plexus. Results have been
622
expressed both as early success rates and also by life-table methods across a longer period of follow-up. Resecting the first rib did not prove to be an advantage both on short-term and long-term assessments. Using this system a long-term failure rate of 20 to 30% at 2 years is reported. This may be high when compared with some series’> 9, but only those with more than 50% improvement as operative success have been included, and the ‘fair’ group has been classified as failures. The significance of the soft tissue anomalies is well demonstrated by the work of Roes’ who classified neurogenic thoracic outlet syndrome into upper and lower plexus groups and advocated supraclavicular scalenectomy for patients with upper plexus symptoms. In the authors’ experience, the patients usually presented with a mixture of symptoms and it has been impossible to distinguish them into clear upper and lower plexus groups. It is believed that this classification is unimportant as long as total decompression of the nerves of the brachial plexus is performed during the operation. This can be achieved by anterior and middle scalenectomy alone. The importance of the scalene muscles is further evident by recent reported observations of type I fiber predominance, type II atrophy and endomysial fibrosis 12,13. The exact significance of this histological pattern in pathogenesis is unknown and further control studies of scalene morphology will be necessary. The only significant factor that determined clinical outcomes was the duration of symptoms before surgery. Patients who were operated upon within 2 years of occurrence of symptoms achieved a significantly better outcome. This led the authors to believe that early diagnosis and decompression of neurogenic thoracic outlet syndrome are crucial for success. Because of the lack of physical signs and an objective diagnostic test, some had questioned the existence of neurogenic thoracic outlet syndrome, and advocated surgical decompression only when there is undisputed weakness and wasting of the hand muscles14. Unnecessarily prolonged conservative management or repeated referrals and neuroelectric studies lead only to a waste of resources and patient discomfort. Delay in treatment until neurological signs develop may result in permanent nerve damage and preclude the chance of a complete recovery. The actual neurological compression or irritation, be it bony or muscular, may not be as important as the degree of damage sustained by the nerve itself in determining the final outcome. The ideal decompressive operation for thoracic outlet syndrome should be aimed at restoring an optimal environment for the injured nerves to heal by a thorough neurolysis, and as early as possible, rather than focused on removing the skeletal boundaries of the thoracic outlet (first thoracic rib). It is concluded that in patients with neurogenic thoracic outlet syndrome, resecting the first rib adds no advantage in both short-term and long-term assessment CARDIOVASCULAR SURGERY DECEMBER 1995 VOL 3 NO 6
Neurogenicthorxic outlet decompresskz?: .s’ kti K Chenget al. of the patients’ outcome. In the authors’ experience it caused unnecessary morbidity and a longer hospital stay. Supraclavicular anterior and middle scalenectomy is recommended for patients with neurogenic thoracic outlet syndrome requiring operative intervention.
Acknowledgements These studies were supported Vascular Research Foundation.
in part by the Pacific
References 1. 2. 3. 4. 5. 6.
Koos DB. Transaxillary approach for first rib resection to relieve thoracic outlet syndrome. Ann Surg 1966; 163: 354-8. Reilly LM, Sroney RJ. Supraclavicular approach for thoracic outlet decompression. J Vusc Surg 1988; 8: 329-34. Lee E, Desu M. A computer program for comparing k samples with right censored data. Computer Progrums in Biomedicine 1972; 2: 315-21. Clagett OT Presidential address: research and prosearch. J Thorac Cardiouasc Surg 1962; 44: 153-66. Roos DB. The place for scalenectomy and first rib resection in thoracic oudet syndrome. Surgery 1982; 92: 1077-85. Stallworth JM, Quinn GJ, Aiken AF. Is rib resection necessary
CARDIOVASCULAR
SURGERY
DECEMBER
199s VOL 3 NO 6
for relief of thoracic outlet syndrome? Ilr,r~ 5tq !o’Y; l&S: 581-92. 7. Dale WA. Thoracic outlet compression synirom::. i rmque IX 1982. Arch Surg 1982; 117: 1437-45. 8. Sanders RJ, Monsour JW, Gerber WF, Adams WR, Thompson N. Scalenectomy vc~s’susfirst rib resection for treatment of the thoracic outlet syndrome. Surgery 1979; 85: ! 09-21. 9. Sanders RJ, Pearce WH. The treatment 11f rhoracic outlet [ tltisz .SUY~ syndrome: a comparison of different operatrons. 1989; 10: 626--34. 10. Qvarfordt PC, Ehrenfeld WK, Stoney KJ. Supraclavicular radial scalenectomy and transaxillary first rib resecrinn for the thoracic outlet syndrome: a combined approach. Am f Cq 1984; 148: 111-16. 11. Green RM, McNamara J, Ouriel K. Long term follow-up after thoracic outlet decompression: an analysis of factors determin ing outcome. J Vast Surg 1991; 14: 73%+h. 12. Machleder HI, Mall F, Verity A. The anterior ,caIene muscle in thoracic outlet compression syndrome. Wistochemical and morphometric studies: Arch SUY~ 1986; 121: I 141-4. 13. Sanders RJ, Jackson CGR, Banchero N, Pearce WH. Scalene muscle abnormalities in traumatic thoracic outlet syndrome. Am J Surg 1990; 1.59: 23176. 14. Wilbourn AJ. The thoracic outlet syndrom a~ ‘averdla,gxJlosed. Arch Nemo/ 1990; 47: 328-W.
Paper accepted 5 December 1994