- Split View
-
Views
-
Cite
Cite
Chenlu Yang, Deping Zhao, Xiao Zhou, Jiaan Ding, Gening Jiang, A comparative study of video-assisted thoracoscopic resection versus thoracotomy for neurogenic tumours arising at the thoracic apex, Interactive CardioVascular and Thoracic Surgery, Volume 20, Issue 1, January 2015, Pages 35–39, https://doi.org/10.1093/icvts/ivu328
- Share Icon Share
The surgical outcome of neurogenic tumours arising at the thoracic apex remains largely undefined. In this retrospective study, we compared the efficacy and safety of thoracoscopic surgery and thoracotomy for neurogenic tumours at the thoracic apex in 63 patients who received surgical treatment between 1992 and 2012 at our medical centre.
Forty-four (69.8%) patients received thoracotomy (Group A) and 19 (30.2%) patients underwent video-assisted thoracoscopic surgery (Group B). Operative time, estimated blood loss (EBL), postoperative length of hospital stay and nervous system complications were recorded.
The two groups of patients were comparable in demographic and baseline characteristics except that Group A patients had a significantly larger tumour size (mean, 4.9 ± 1.0 cm) than Group B patients (mean, 4.1 ± 1.2 cm; P = 0.01). The mean operative time was markedly greater for Group A (120.2 ± 41.2 min) than Group B (93.2 ± 34.5 min; P = 0.009). Group A had significantly greater EBL (245.23 ± 197.78 ml) than Group B (117.4 ± 138.2 ml; P < 0.001). Total tumour resection was achieved in all patients and all neurogenic tumours were benign. The mean length of postoperative hospital stay was markedly longer in Group A (7.0 ± 2.1 days) than Group B (4.8 ± 2.0; P < 0.001). Postoperatively, brachial plexus injury was found in 1 patient (2.3%) in Group A and 4 patients (21.1%) in Group B (P = 0.026).
Though thoracoscopic surgery is associated with diminished blood loss, reduced operative time and shortened hospital stay, it has a markedly increased incidence of brachial plexus injury.
INTRODUCTION
Thoracic neurogenic tumours typically originate from the neural crest during the development of peripheral nerve systems such as those of the autonomic or paraganglionic nerves and are mostly concentrated in the posterior mediastinum [1]. These tumours are most commonly seen in children but are less frequently found and mostly benign in adults [2]. Tumours originated from peripheral nervous systems include neurilemmoma and neurofibromatosis and those originated from the autonomic ganglions include ganglioneuroma and some tumours are paraganglioneuromas [3]. More than 90% of posterior mediastinum tumours are neurogenic tumours, and they are concentrated in the paravertebral sulcus region and account for 12–21% of primary mediastinum tumours [1]. Approximately 10% of neurogenic tumours may invade the spine and grow into dumbbell-shaped tumours [4].
Thoracoscopic surgery has increasingly become the standard approach for resection of neurogenic tumours [5, 6]. However, neurogenic tumours at the thoracic apex mostly originate from T1 to T3 spinal nerve roots or the sympathetic trunk and are relatively high in position and adjacent to such important structures as the stellate ganglion, brachial plexus and subclavian artery. These features render thoracoscopic operation difficult. Inadequate angle for manoeuverability may cause uncontrollable bleeding or brachial plexus injury associated with eletrocoagulation [5, 7]; however, there has been no systemic study on the use of thoracoscopic surgery for neurogenic tumours at the thoracic apex. In the current retrospective study, we compared the efficacy and safety of thoracoscopic surgery and thoracotomy for neurogenic tumours at the thoracic apex in 63 patients who received surgical treatment between 1992 and 2012 at our medical centre.
MATERIALS AND METHODS
Patients
We reviewed the medical and surgical records of patients with pathologically confirmed neurogenic tumours who received surgical treatment at Shanghai Thoracic Diseases Hospital between 1 January 1992 and 31 December 2012. The study protocol was approved by the local institutional review board at the authors’ affiliated hospital and patient consent was not required because of the retrospective nature of the study. A patient was included if computed tomography (CT) and/or magnetic resonance imaging (MRI) showed that the tumour was located in the thoracic apex and the superior border of the tumour was higher than the superior edge of the first rib. Patients were excluded (i) if their tumour size exceeded 6 cm in diameter; (ii) if the tumour was not confined to the thorax; (iii) if the tumour boundary was not clearly marked and (iv) if they had diabetes, coronary heart disease, hypertension or other chronic diseases. Clinical records were examined for demographics, tumour characteristics, adjuvant therapies, surgical approach, surgical outcome and complications, local recurrence of disease and overall survival. All patients had complete history, underwent physical examination and received electrocardiography and routine blood chemistries.
Surgical techniques
The patient was placed in the lateral position with the operative side facing upward. The non-operated lung was ventilated after the double-lumen endotracheal tube was placed in the trachea. The chest was entered via the intercostal space after the posterolateral incision was made for thoracotomy. In 2 cases, the tumour was exposed after median sternotomy with an assisted transverse incision in the intercostal space and elevation of the anterior chest wall. For thoracoscopic surgery, the thoracoscope was entered via a 1.5 cm midaxillary incision in the sixth intercostal space and then based on tumour location, an assisted 3–4 cm anterior axillary incision (utility incision) was made in the third intercostal space. The remaining procedure was identical in patients receiving thoracotomy (Group A) and those undergoing video-assisted thoracoscopic surgery (VATS) (Group B). The parietal pleura were excised along and 5 mm from the raised edge of the tumour border, which was then removed from the tumour bed using suction instrument in VATS or manually in thoracotomy. Tumour bed bleeding was managed by electrocoagulation and tumour pedicle was ligated in thoracotomy and severed by electrocoagulation in VATS. Intercostal and paravertebral feeding vessels were electrocoagulated and ligated or clamped, if necessary. The tumour specimen was sent for histological examination. In certain patients undergoing thoracoscopic surgery, the utility incision was extended to allow retrieval of the tumour. Two chest drainage tubes were placed in the patient postoperatively. Operations were performed by a team of similarly experienced surgeons.
Patient evaluation
Operative time for tumour excision, estimated blood loss (EBL), postoperative length of hospital stay and complications were recorded. Tumour size was determined as the greatest diameter of the resected tumour specimen. Postoperative nervous system complications were considered present if patients newly developed Horner’s syndrome and brachial plexus injury after surgery or postoperative aggravation of pre-existing nervous system symptoms. Horner’s syndrome was established if blepharoptosis and hemifacial anhydrosis were present and brachial plexus injury was diagnosed if thumb opposition and hand grip were weakened and fingertip numbness was present.
Statistical analysis
Descriptive statistics was used to describe patient demographic and baseline characteristics and postoperative nervous system complications. Normally distributed data were expressed as mean±standard deviation and analysed using the Statistical Product and Service Solutions (SPSS) software versions 15.0 (SPSS, Inc., Chicago, IL, USA). Non-normally distributed data were expressed as median. Student’s t-test was used to compare operative time and EBL between the two groups and Mann–Whitney U-test was used to compare postoperative length of hospital stay. The χ2 test was used to examine differences with categorical variables. P < 0.05 was considered statistically significant.
RESULTS
Patient demographic and baseline characteristics
Seventy-four patients received surgical treatment at our hospital during the review period. Eight patients were excluded because their tumour diameter exceeded 6 cm, 1 patient was excluded because the tumour extended into the neck and 2 patients were excluded because the relationship was not clearly defined between the tumour and its adjacent structures. Finally, 63 patients were included in this retrospective analysis. The demographic and baseline characteristics of the patients are given in Table 1. The median age of the patients was 42.6 (range, 7–69) years and there were slightly more female patients (53.6%) than male patients (46.4%). The majority of the patients (63.5%) had left thoracic inlet tumours. A total of 18 (28.6%) patients had preoperative nervous system symptoms including Horner’s syndrome in 4 patients and upper limb numbness in 3 patients. Approximately three quarters of the tumours (74.6%) were neurilemmoma, followed by ganglioneuroma (17.5%) and neurofibromatosis (7.9%). The mean size of the tumour was 4.7 ± 1.1 (range, 1.7–6) cm. In addition, 36.5% of the tumours were pedunculated. All patients (n = 45) underwent CT scan while only slightly more than half of the patients (n = 28) underwent MRI for diagnosis of tumours at the apex. Furthermore, no patients received any preoperative neo-adjuvant therapy.
Variables . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Male (%) | 28 (44.4) | 20 (45.5) | 8 (42.1) | 0.806 |
Age (years) | 41.97 ± 13.74 | 40.5 ± 14.35 | 45.37 ± 11.89 | 0.170 |
Tumour size (cm) | 4.66 ± 1.14 | 4.91 ± 1.02 | 4.08 ± 1.21 | 0.010 |
Tumour at thoracic location (%) | 0.094 | |||
Left | 23 (36.5) | 19 (43.2) | 4 (21.1) | |
Right | 40 (63.5) | 25 (56.8) | 15 (78.9) | |
Pathological type (%) | 0.051 | |||
Neurilemmoma | 47 (74.6) | 29 (65.9) | 18 (94.7) | |
Neurofibromatosis | 5 (7.9) | 5 (11.4) | 0 (0) | |
Ganglioneuroma | 11 (17.5) | 10 (22.7) | 1 (5.3) | |
Pedunculated tumour (%) | 23 (36.5) | 19 (43.2) | 4 (21.1) | 0.094 |
Preoperative nervous system symptoms (%) | 18 (28.6) | 14 (31.8) | 4 (21.1) | 0.385 |
Variables . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Male (%) | 28 (44.4) | 20 (45.5) | 8 (42.1) | 0.806 |
Age (years) | 41.97 ± 13.74 | 40.5 ± 14.35 | 45.37 ± 11.89 | 0.170 |
Tumour size (cm) | 4.66 ± 1.14 | 4.91 ± 1.02 | 4.08 ± 1.21 | 0.010 |
Tumour at thoracic location (%) | 0.094 | |||
Left | 23 (36.5) | 19 (43.2) | 4 (21.1) | |
Right | 40 (63.5) | 25 (56.8) | 15 (78.9) | |
Pathological type (%) | 0.051 | |||
Neurilemmoma | 47 (74.6) | 29 (65.9) | 18 (94.7) | |
Neurofibromatosis | 5 (7.9) | 5 (11.4) | 0 (0) | |
Ganglioneuroma | 11 (17.5) | 10 (22.7) | 1 (5.3) | |
Pedunculated tumour (%) | 23 (36.5) | 19 (43.2) | 4 (21.1) | 0.094 |
Preoperative nervous system symptoms (%) | 18 (28.6) | 14 (31.8) | 4 (21.1) | 0.385 |
Variables . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Male (%) | 28 (44.4) | 20 (45.5) | 8 (42.1) | 0.806 |
Age (years) | 41.97 ± 13.74 | 40.5 ± 14.35 | 45.37 ± 11.89 | 0.170 |
Tumour size (cm) | 4.66 ± 1.14 | 4.91 ± 1.02 | 4.08 ± 1.21 | 0.010 |
Tumour at thoracic location (%) | 0.094 | |||
Left | 23 (36.5) | 19 (43.2) | 4 (21.1) | |
Right | 40 (63.5) | 25 (56.8) | 15 (78.9) | |
Pathological type (%) | 0.051 | |||
Neurilemmoma | 47 (74.6) | 29 (65.9) | 18 (94.7) | |
Neurofibromatosis | 5 (7.9) | 5 (11.4) | 0 (0) | |
Ganglioneuroma | 11 (17.5) | 10 (22.7) | 1 (5.3) | |
Pedunculated tumour (%) | 23 (36.5) | 19 (43.2) | 4 (21.1) | 0.094 |
Preoperative nervous system symptoms (%) | 18 (28.6) | 14 (31.8) | 4 (21.1) | 0.385 |
Variables . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Male (%) | 28 (44.4) | 20 (45.5) | 8 (42.1) | 0.806 |
Age (years) | 41.97 ± 13.74 | 40.5 ± 14.35 | 45.37 ± 11.89 | 0.170 |
Tumour size (cm) | 4.66 ± 1.14 | 4.91 ± 1.02 | 4.08 ± 1.21 | 0.010 |
Tumour at thoracic location (%) | 0.094 | |||
Left | 23 (36.5) | 19 (43.2) | 4 (21.1) | |
Right | 40 (63.5) | 25 (56.8) | 15 (78.9) | |
Pathological type (%) | 0.051 | |||
Neurilemmoma | 47 (74.6) | 29 (65.9) | 18 (94.7) | |
Neurofibromatosis | 5 (7.9) | 5 (11.4) | 0 (0) | |
Ganglioneuroma | 11 (17.5) | 10 (22.7) | 1 (5.3) | |
Pedunculated tumour (%) | 23 (36.5) | 19 (43.2) | 4 (21.1) | 0.094 |
Preoperative nervous system symptoms (%) | 18 (28.6) | 14 (31.8) | 4 (21.1) | 0.385 |
Forty-four (69.8%) patients received thoracotomy (Group A) and 19 (30.2%) patients underwent VATS (Group B) (Table 1). The two groups of patients were comparable in age and sex. Though a higher proportion of patients in Group A (31.8%) had nervous system symptoms prior to surgery compared with Group B (21.1%), there was no statistical difference (P = 0.094). Neurilemmoma was the predominant tumour type in both Group A (65.9%) and Group B (94.7%). Furthermore, a higher percentage of tumours in Group A (43.2%) was pedunculated than that of Group B (21.1%); however, there was no statistical difference between the two groups (P = 0.094). Group A patients had a significantly larger tumour size (mean, 4.9 ± 1.0 cm; range, 3–6 cm) than Group B patients (mean, 4.1 ± 1.2 cm; range, 1.7–6 cm) (P = 0.01). Patients in the two groups were comparable in demographic and baseline characteristics except tumour size.
Surgical outcomes
The surgical characteristics of the patients are given in Table 2. The mean operative time was 112.1 ± 41.0 (range, 50–240) min for all patients and was markedly greater for patients in Group A (120.2 ± 41.2 min) than Group B (93.2 ± 34.5 min) (P = 0.009). The mean EBL was 206.7 ± 190.2 (range, 10–1600) ml. Patients in Group A had significantly greater EBL (245.2 ± 197.8 ml) than patients in Group B (117.4 ± 138.2 ml) (P < 0.001). Total tumour resection was achieved in all patients. No death occurred during the surgery. Pathological examination indicated that all neurogenic tumours were benign. The tumour was multiple in origin in 1 patient receiving thoracotomy who also had Von-Recklinghausen disease. The mean length of postoperative hospital stay was 6.3 ± 2.3 days for all patients and was markedly longer in patients in Group A (7.0 ± 2.1 days) than Group B (4.8 ± 2.0 days) (P < 0.001).
Variables . | All, n = 63 . | Group, A n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Assisted midline L-shaped transverse incision (%) | 2 (3.2) | 2 (4.5) | 0 (0.0) | 1.000 |
Assisted small incision (%) | 6 (9.5) | 0 (0.0) | 6 (31.6) | 0.001 |
Operative time (min) | 112.06 ± 41.00 | 120.23 ± 41.23 | 93.16 ± 34.49 | 0.009 |
Estimated blood loss (ml) | 206.67 ± 190.20 | 245.23 ± 197.78 | 117.37 ± 138.2 | <0.001 |
Postoperative length of hospital stay (days) | 6.32 ± 2.27 | 6.98 ± 2.06 | 4.79 ± 2.02 | <0.001 |
Postoperative improvement of nervous system symptom (%) | 15/18 (83.33) | 13/14 (92.85) | 2/4 (50) | 0.108 |
Variables . | All, n = 63 . | Group, A n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Assisted midline L-shaped transverse incision (%) | 2 (3.2) | 2 (4.5) | 0 (0.0) | 1.000 |
Assisted small incision (%) | 6 (9.5) | 0 (0.0) | 6 (31.6) | 0.001 |
Operative time (min) | 112.06 ± 41.00 | 120.23 ± 41.23 | 93.16 ± 34.49 | 0.009 |
Estimated blood loss (ml) | 206.67 ± 190.20 | 245.23 ± 197.78 | 117.37 ± 138.2 | <0.001 |
Postoperative length of hospital stay (days) | 6.32 ± 2.27 | 6.98 ± 2.06 | 4.79 ± 2.02 | <0.001 |
Postoperative improvement of nervous system symptom (%) | 15/18 (83.33) | 13/14 (92.85) | 2/4 (50) | 0.108 |
Variables . | All, n = 63 . | Group, A n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Assisted midline L-shaped transverse incision (%) | 2 (3.2) | 2 (4.5) | 0 (0.0) | 1.000 |
Assisted small incision (%) | 6 (9.5) | 0 (0.0) | 6 (31.6) | 0.001 |
Operative time (min) | 112.06 ± 41.00 | 120.23 ± 41.23 | 93.16 ± 34.49 | 0.009 |
Estimated blood loss (ml) | 206.67 ± 190.20 | 245.23 ± 197.78 | 117.37 ± 138.2 | <0.001 |
Postoperative length of hospital stay (days) | 6.32 ± 2.27 | 6.98 ± 2.06 | 4.79 ± 2.02 | <0.001 |
Postoperative improvement of nervous system symptom (%) | 15/18 (83.33) | 13/14 (92.85) | 2/4 (50) | 0.108 |
Variables . | All, n = 63 . | Group, A n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Assisted midline L-shaped transverse incision (%) | 2 (3.2) | 2 (4.5) | 0 (0.0) | 1.000 |
Assisted small incision (%) | 6 (9.5) | 0 (0.0) | 6 (31.6) | 0.001 |
Operative time (min) | 112.06 ± 41.00 | 120.23 ± 41.23 | 93.16 ± 34.49 | 0.009 |
Estimated blood loss (ml) | 206.67 ± 190.20 | 245.23 ± 197.78 | 117.37 ± 138.2 | <0.001 |
Postoperative length of hospital stay (days) | 6.32 ± 2.27 | 6.98 ± 2.06 | 4.79 ± 2.02 | <0.001 |
Postoperative improvement of nervous system symptom (%) | 15/18 (83.33) | 13/14 (92.85) | 2/4 (50) | 0.108 |
The nervous system symptoms improved in 83.3% (15/18) of the patients who had preoperative nervous system symptoms (Group A, 92.9%, 13/14 vs Group B, 50.0%, 2/4), with no difference between the two groups (P = 0.108). Nervous system symptoms were alleviated in 2 patients postoperatively but aggravated in 8 patients.
The patients were followed up for a mean duration of 81 (range, 12–216 months). No tumour recurrence was reported. At the final follow-up visit, there was no significant improvement in nervous system symptoms except in 1 patient in Group A who showed apparent improvement in fingertip numbness 2 years after surgery.
Postoperative adverse events
Horner’s syndrome was present in 3 patients (6.8%) in Group A and 1 patient (7.1%) in Group B (P > 0.05) (Table 3). Brachial plexus injury was present in 1 patient (2.3%) in Group A and 4 patients (21.1%) in Group B and was first noticed postsurgery in all patients. No surgical adhesion was reported in both groups. One patient in Group A had bleeding of the subclavian artery with an EBL of 1600 ml. The patient also developed postoperative respiratory failure.
Postoperative adverse events . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Horner’s syndrome (%) | 4*(6.9) | 3*(6.8) | 1*(7.1) | 1.000 |
Blepharoptosis | 4 | 3 | 1 | |
Facial anhydrosis | 2 | 1 | 1 | |
Total events | 6 | 4 | 2 | |
Brachial plexus injury (%) | 5*(7.9) | 1*(2.3) | 4*(21.1) | 0.026 |
Finger numbness | 5 | 1 | 4 | |
Weakened thumb opposition | 2 | 0 | 2 | |
Hand grip | 2 | 0 | 2 | |
Total events | 9 | 1 | 8 |
Postoperative adverse events . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Horner’s syndrome (%) | 4*(6.9) | 3*(6.8) | 1*(7.1) | 1.000 |
Blepharoptosis | 4 | 3 | 1 | |
Facial anhydrosis | 2 | 1 | 1 | |
Total events | 6 | 4 | 2 | |
Brachial plexus injury (%) | 5*(7.9) | 1*(2.3) | 4*(21.1) | 0.026 |
Finger numbness | 5 | 1 | 4 | |
Weakened thumb opposition | 2 | 0 | 2 | |
Hand grip | 2 | 0 | 2 | |
Total events | 9 | 1 | 8 |
Postoperative adverse events . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Horner’s syndrome (%) | 4*(6.9) | 3*(6.8) | 1*(7.1) | 1.000 |
Blepharoptosis | 4 | 3 | 1 | |
Facial anhydrosis | 2 | 1 | 1 | |
Total events | 6 | 4 | 2 | |
Brachial plexus injury (%) | 5*(7.9) | 1*(2.3) | 4*(21.1) | 0.026 |
Finger numbness | 5 | 1 | 4 | |
Weakened thumb opposition | 2 | 0 | 2 | |
Hand grip | 2 | 0 | 2 | |
Total events | 9 | 1 | 8 |
Postoperative adverse events . | All, n = 63 . | Group A, n = 44 . | Group B, n = 19 . | P-value . |
---|---|---|---|---|
Horner’s syndrome (%) | 4*(6.9) | 3*(6.8) | 1*(7.1) | 1.000 |
Blepharoptosis | 4 | 3 | 1 | |
Facial anhydrosis | 2 | 1 | 1 | |
Total events | 6 | 4 | 2 | |
Brachial plexus injury (%) | 5*(7.9) | 1*(2.3) | 4*(21.1) | 0.026 |
Finger numbness | 5 | 1 | 4 | |
Weakened thumb opposition | 2 | 0 | 2 | |
Hand grip | 2 | 0 | 2 | |
Total events | 9 | 1 | 8 |
DISCUSSION
More than 80% of neurogenic tumours are benign and are originated from the neural crest, and 95% of the posterior mediastinum neurogenic tumours develop from the spinal nerve roots, sympathetic chain and branches of the intercostal nerves, which account for 12–21% of mediastinum tumours [1, 8, 9]. Approximately half of neurogenic tumour patients are asymptomatic. Since Landreneau et al. [10] first reported the use of thoracoscope for resection of posterior mediastinum neurogenic tumours in 1992, surgeons have increasingly shown that thoracoscopic surgery is the preferred method for this type of benign tumours. Compared with thoracotomy, thoracoscopic surgery shortens the operative time, results in a faster postoperative recovery, lessens postoperative complications and reduces postoperative pain [5–9, 11].
Riquet et al. [7] and Yamaguchi et al. [11] recommended against thoracoscopic surgery for tumours >6 cm in size, malignant tumours, tumours extended into the spinal cord, tumours located at the costophrenic angle and tumours at the thoracic apex. The current study focused on neurogenic tumours at the thoracic apex smaller than 6 cm. We also excluded neurogenic tumours that had extended into the spinal cord and become dumbbell-shaped, malignant neurogenic tumours and tumours with apparent invasion of large blood vessels and nerves. There has been no consensus for the best approach for surgical resection of these tumours. Currently, there is no literature available on surgical experience of neurogenic tumours at the thoracic apex. Here, we reviewed our experience with 63 patients with neurogenic tumours at the thoracic apex over a 20-year span. Patients receiving thoracotomy and thoracoscopic surgery were comparable in demographic characteristics. Except that patients receiving thoracotomy had a significantly larger tumour size, no statistically significant difference in demographic and other baseline characteristics was observed between the two groups. We achieved total tumour resection in all patients. We found that thoracoscopic surgery had a markedly shortened operative time, significantly smaller blood loss and noticeably reduced postoperative length of hospital stay. However, there was no difference in long-term outcome as the final follow-up visit showed virtually no apparent improvement in nervous system symptom in both groups of patients.
One patient receiving thoracotomy (2.3%) had brachial plexus injury while 4 patients receiving thoracoscopic surgery (21.1%) had brachial plexus injury, with a statistical significance. Possible causes of the significantly higher incidence of brachial plexus injury in patients receiving thoracoscopic surgery included (i) inadequate exposure of the superior border of the tumour; (ii) the relation between the tumour and adjacent structures cannot be fully explored manually; (iii) increased eletrocoagulation. The inferior trunk of the brachial plexus is mainly composed of C8+T1 spinal nerves and is located posterior to the subclavian artery and anterior to the transverse process of C7 and is close to the visceral pleura of the lung apex. In both thoracotomy and thoracoscopic surgery, the superior portion of the tumour is observed in a posterosuperior direction and the anatomic relation between the superior portion of the tumour and its adjacent structures may not be visualized adequately. Thoracotomy can isolate the tumour by blunt dissection and is not limited in manoeuverability as in thoracoscopic surgery, which is limited by the small size of the utility port and reduced manoeuvering angle. The tumour is dissociated by suction and eletrocoagulation in thoracoscopic surgery, which may predispose the patient to brachial plexus injury.
Tumours located at the thoracic apex are considered contraindicated for thoracoscopic surgery [8, 11]. Recently, some surgeons have recommended a combination approach for such tumours with thoracoscopic surgery and neurosurgery [12–15]. Sakuraba et al. [16] consider that tumours located no lower than T3 can be directly resected via the cervical incision instead of entering the visceral pleura. We resected such tumours in 2 patients using the approach and achieved a satisfactory outcome.
The robotic platform has been increasingly used in thoracic surgery [17]. The platform may offer a meticulous approach for surgery and greater manoeuverability because of expanded operating angles. But it has no distinct advantage in tumour exploration with the upper pole of the tumour being a blind spot. Tumours at the thoracic inlet should be inspected in a systemic approach in the superoinferior direction for the exploration of the brachial plexus in order to minimize brachial injury. The use of a robotic platform for tumours at the thoracic inlet awaits a definite proof of superiority in surgical outcomes.
In conclusion, though thoracoscopic surgery for neurogenic tumours located at the thoracic apex is associated with diminished blood loss, reduced operative time and shortened hospital stay, it is associated with an increased incidence of brachial plexus injury. Alternative approaches such as thoracoscopic surgery and neurosurgery may be explored for better efficacy and safety.
Acknowledgements
We thank all the people who gave us help in the research.
Conflict of interest: none declared.
REFERENCES
Author notes
Contributed equally.
Comments
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved
We thank to Yang et al. for their study about the resection of the posterior mediastinal lesions [1]. We want to add a comment on the approach for the posterior mediastinal tumors.
We agree that thoracoscopic resection of posterior mediastinal tumours can be performed successfully with decreased operation time, hospital stay and patient discomfort. But we think that those that are located to the apex of the hemithorax should not be resected thoracoscopically. Authors wrote that thoracoscopic surgery was associated with reduced operation time, blood loss and hospital stay. These were good results but there was one complication (brachial plexus lesion) that was higher in the video-assisted thoracoscopic surgery (VATS) group. We think that this complication is much more important than the benefits mentioned above.
Most posterior mediastinal tumours have a benign character [2]. No surgeon wants such a complication after performing a surgery for a benign lesion. Also, the thoracoscopic approaches near to the great vessels carries the risk of massive haemorrhage because of the limited manoeuvrability. At our institution, we used muscle sparing thoracotomy on a patient who had a lesion at the apex of the right hemithorax. More than half of the operation was performed with blunt finger dissection. We did not spend too much time for the exposure, as is so in VATS procedures, and for the dissection. Closure of the hemitorax was also not time-consuming because there was not a lot of muscle tissue to be cut through.
Authors declared that they used two chest tubes after the operation. We think that placing two chest tubes for these kind of surgeries adds to the discomfort of the patient. Chest tubes can be the only annoying thing after thoracic procedures for the patients. In general, we use single chest tubes for all procedures performed, other than lung surgery. We place the chest tube to the apex of the hemithorax and open a hole on the lower part of it to drain any effusions. Also, cutting the specimen before taking it out of the thorax is a better approach than to enlarge the utility thoracotomy [3].
References
[1] Yang C, Zhao D, Zhou X, Ding J, Jiang G. A comparative study of video-assisted thoracoscopic resection vs thoracotomy for neurogenic tumours arising at the thoracic apex. Interact CardioVasc Thorac Surg 2014;doi: 10.1093/icvts/ivu328.
[2] Davidson KG, Walbaum PR, McCormack RJ. Intrathoracic neural tumors. Thorax 1978;33:359-367.
[3] Cardillo G, Carleo F, Khalil MW, Carbone L, Treggiari S, Salvadori L, et al. Surgical treatment of benign neurogenic tumours of the mediastinum: a single institution report. Eur J Cardiothorac Surg 2008;34:1210-4.
Conflict of interest:
none declared
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved
The Shangai group [1] presented a cohort of 63 surgically treated patients who were operated on during a time period of 20 years. Not so long ago, similar results were presented by the Osaka group [2]. Both group of authors agree that it is an accepted opinion today that benign intrathoracic neurogenic tumors smaller then 6 cm in size are good candidates for video-assisted thoracoscopic surgery resection, which can be achieved safely with rapid recovery [3, 4]. When the tumour exceeds 6 cm in size and is located at a costophrenic angle or at the thoracic apex, the real dilemma of choosing the most appropriate surgical approach arises.
In the last 5 years at our department, we have surgically treated two male patients in their early thirties, with palpable masses with an ulnar distribution in the left supraclavicular region, paraesthesia and loss of strength in the left hand. Computed tomography scans revealed a smooth, round tumour of more than 9 cm in diameter. In both cases, a left supraclavicular approach was undertaken. A supraclavicular incision permitted an accurate and safe exposure of the cervical portion of the tumour and its mobilization from the superior and median trunk of the brachial plexus and from the subclavian artery. Intraoperative electrophysiological monitoring helped to spare nerve function after tumour resection. In both cases, a round, firm mass with a size of more than 9 X 6 cm was seen occupying the apical region of the left chest. The mass appeared to be connected to the lower trunk of the brachial plexus close to the upper border of the first rib. The tumour was enucleated and shelled out of the lower trunk, avoiding any injury to the trunk itself. Pleural cover and remnants of the capsule at the root of neck were easily cleared. Histology of the tumours established the diagnosis of Schwannoma. Twelve months after surgery, there were no residual numbness, paraesthesia and loss of strength of the left arm in both patients.
In conclusion, considering that neurogenic tumours arising at the thoracic apex are generally benign, all efforts should be directed towards their surgical resection with minimally-invasive approaches, even when they arise as multiple simultaneous lesions or in unusual locations. For those lesions extending in the cervical region that do not involve the spinal canal, a multidisciplinary team including thoracic and vascular surgeons allows their resection in a safe single-stage procedure.
References
[1] Yang C, Zhao D, Zhou X, Ding J, Jiang G. A comparative study of video-assisted thoracoscopic resection vs thoracotomy for neurogenic tumours arising at the thoracic apex. Interact CardioVasc Thorac Surg 2014;doi: 10.1093/icvts/ivu328.
[2] Takeda S, Miyoshi S, Minami M, Matsuda H. Intrathoracic neurogenic tumors--50 years'experience in a Japanese institution. Eur J Cardiothorac Surg 2004;26:807-812.
[3] Riquet M, Mouroux J, Pons F, Debrosse D, Dujon A, Dahan M, Jancovici R. Videothoracic excision of thoracic neurogenic tumors. Ann Thorac Surg 1995;60:943-6.
[4] Yamaguchi M, Yoshino I, Fukuyama S, Osoegawa A, Kameyama T, Tagawa T, Maehara Y. Surgical treatment of neurogenic tumors of the chest. Ann Thorac Cardiovasc Surg 2006;12:194-196.
Conflict of interest:
none declared.
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved
We read with great interest the article by Yang et al. [1], reporting on an a series of 63 patients undergoing surgery for neurogenic tumours arising at the thoracic apex. Surgical resection is considered the mainstay of treatment for such tumours [2]. Generally, postero-lateral thoracotomy has been the traditional surgical approach, in order to obtain a radical resection. On the other hand, in the early 90s, Landreneau et al. reported the first video-assisted thoracoscopic (VATS) procedure for treating posterior mediastinal tumours [3] and, nowadays, VATS has largely been documented as a safe and effective treatment.
As a rule, the incidence of mediastinal tumours is low, and tumours from posterior region occur in about 20% of all mediastinal ones. In addition, tumours at the apex of the chest cavity close to the great vessels and the stellate ganglion may lead to an increased risk of vascular and nervous injury postoperatively. Thus, it is still unclear as to which treatment is most effective for such rare neoplasms.
In this setting, the Authors [1] performed a comparative analysis on 44 patients undergoing thoracotomy (Group A) and 19 receiving a VATS resection (Group B) and reported a diminished intraoperative blood loss, reduced operative time and shortened hospital stay in the VATS group. Though both groups were comparable in terms of clinical (age, gender, side) and pathological (histology) features, they differed statistically in pathological tumour size. In our opinion, this difference translates in a confounding selection bias: in fact, as reported by others [4], bigger solid tumours are inherently and strongly associated with a more difficult resection, due to hard exposure of the tumour pedicle (usually shadowed) and more vascular supplies. For this reason, we suggest a propensity score matching method to perform an accurate comparison between both groups.
Moreover, the Authors demonstrated an increased incidence of brachial plexus injuries in Group B. In our opinion, rather than a postoperative neurologic examination, a quality-of-life evaluation in larger cohort studies could be more sensitive to identify the best surgical procedure in such "benign" neoplasms.
We would really appreciate Authors' reflections and reactions on the issues raised.
References
[1] Yang C, Zhao D, Zhou X, Ding J, Jiang G. A comparative study of video-assisted thoracoscopic resection vs thoracotomy for neurogenic tumours arising at the thoracic apex. Interact CardioVasc Thorac Surg 2014;doi: 10.1093/icvts/ivu328.
[2] Strollo DC, Rosado-de-Christenson ML, Jett JR. Primary mediastinal tumors: part II. Tumors of the middle and posterior mediastinum. Chest 1997;112: 1344-57
[3] Landreneau RJ, Dowling RD, Ferson PF. Thoracoscopic resection of a posterior mediastinal neurogenic tumor. Chest 1992;102:1288- 90.
[4] Li Y, Wang J. Experience of video-assisted thoracoscopic resection for posterior mediastinal neurogenic tumours: a retrospective analysis of 58 patients. ANZ J Surg 2013 Sep;83:664-8.
Conflict of interest:
none declared