Original Thoracic
223
Is a Vessel-Sealing System Useful for Resection of Anterior Mediastinal Mass? Mario Santini1
Alfonso Fiorelli1
Giovanni Vicidomini1
1 Department of Cardio-Thoracic Disease, Second University of Naples,
Thoracic Surgery Unit, Naples, Italy 2 Department of Public Health, Second University of Naples, Section of Pathology, Naples, Italy
Gaetana Messina1
Marina Accardo2
Address for correspondence Mario Santini, MD, Department of Thoracic Surgery, Second University of Naples, P.zza L. Miraglia 2, Naples 80138, Italy (e-mail: [email protected]).
Abstract
Keywords
► LigaSure ► mediastinal resection ► mass
Objective To valuate if the LigaSure (Valleylab, Boulder, Colorado, United States) vessel-sealing system could reduce operative time, intraoperative blood loss, drainage duration, and hospital stay in patients with anterior mediastinal mass undergoing open resection. Methods Forty consecutive patients having resection of anterior mediastinal mass were randomized into two groups according to whether LigaSure was used (n ¼ 20) or not (n ¼ 20). Tumor size, operative time, intraoperative blood loss, chest tube output and duration, length of hospital stay, morbidity, and mortality were prospectively recorded, then intergroup differences were statistically analyzed. Results Both groups were well matched for age, tumor size, pathologic diagnosis, and incidence of complications. LigaSure significantly reduced operative duration (p < 0.0001) compared with the traditional technique but without leading to any significant reduction in intraoperative blood loss (p ¼ 0.2), chest tube output (p ¼ 0.2) and duration (p ¼ 0.2), and length of hospital stay (p ¼ 0.5). Conclusions The reduced operative time using LigaSure translates into less exposure to general anesthesia, which is particularly important for patients with myasthenia and potentially reducing cost.
Introduction Adequate hemostasis is an important step in surgical procedures involving any mediastinal organs to ensure a safe outcome. It may be achieved in a standard matter with surgical ties and/or vessel clips, but it is time-consuming.1 Monopolar or bipolar conventional electrocoagulation systems are faster than the ligation-and-tie technique, but they may damage the surrounding structures due to lateral heat dispersion with immediate or delayed complications such as esophageal injury, phrenic palsy, recurrent laryngeal injury, and other potentially life-threatening complications.2 LigaSure (Valleylab, Boulder, Colorado, United States) desiccates
received May 3, 2013 accepted after revision July 22, 2013 published online December 2, 2013
vascular tissues using a feedback-programmed amount of bipolar diathermy. This method of vessel sealing relies on the application of a precise amount of bipolar electrocoagulation and pressure to the tissue, leading to the denaturation of the collagen and elastin in vessel walls, resulting in a hemostatic seal. The technique is able to seal vessels up to 7 mm in diameter and the seal can withstand three times the normal systolic pressure.3 In addition, compared with conventional electrocoagulation instruments, this vessel-sealing mechanism has a significantly reduced thermal spread profile and therefore a decreased risk of injury to adjacent structures. Many reports showed LigaSure to be safe with reduced operative time in digestive,4 urologic,5 gynecologic,6 and
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1355227. ISSN 0171-6425.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thorac Cardiovasc Surg 2015;63:223–230.
Vessel-Sealing System for Resection of Anterior Mediastinal Mass thoracic procedures.7,8 The goal of the present study was to compare operative and postoperative data of patients undergoing resection of an anterior mediastinal mass using LigaSure versus the traditional hemostatic techniques to assess whether LigaSure could reduce the operative time (primary end point), drainage output and duration, and length of hospital stay (secondary end points) without worsening the postoperative outcome.
Materials and Methods Study Design This single-blinded unicenter randomized study enrolled all consecutive patients with anterior mediastinal mass scheduled for surgery from January 2006 to December 2011. The patients were divided into two groups: the LigaSure group (LG) included patients in whom resection was performed using LigaSure, and the traditional group (TG) included patients in whom hemostasis dissection was performed using traditional methods such as suture ligatures or mono- or bipolar electrocoagulation systems. Patients were unaware of allocation to either intraoperative use of the LigaSure or traditional technique. The technique to be used was assigned to the surgeon in the operating room from a randomized list engendered by statistical software (MedCalc statistical software version 12.4.0, MedCalc Software, Ostend, Belgium). Inclusion criteria were patients (1) with resectable anterior mediastinal mass, (1) fit for open surgery, (3) and who gave written informed consent for participating to the study. Exclusion criteria were (1) use of strategy other than sternotomy for resection, (2) any previous thoracic or mediastinal surgery that might bias operative times, and (3) use of other hemostatic devices in both treatment arms. The time of operation (total minutes from incision to skin closure) and intraoperative blood loss (milliliters) were the primary end points of the study. Secondary end points were chest tube output (milliliters), chest drain duration (days), length of hospital stay (days), morbidity, and mortality. The protocol of this study was approved by the Hospital Ethics Committee of the Second University of Naples.
Santini et al.
and in the CG (n ¼ 20). In addition to other variables, we also evaluated the data of patients with myasthenia gravis (MG) among two groups to avoid MG influencing the outcome results.
Operative Technique In this study, all operations were made by the same senior surgeon who had similar experience in using the LigaSure and traditional hemostatic methods. The same general anesthesia protocol was used in patients in both groups. All operations were performed through sternotomy with single-lumen endotracheal ventilation, without muscle relaxants in patients with MG. The handpiece was the LigaSure Precise unit. The seal width measured 1 to 3 mm. This device was fit with a central cutting blade between the two coagulative ligations (or sealing) and division of the mediastinal vessels or tissue. An insulated layer minimized heat transmission to the exterior jaw surfaces, with the thermal spread typically limited to less than 2 mm. Blood vessels, lymphatics, and tissues were likewise cut with LigaSure. Vessels larger than 7 mm in diameter were ligated by a thread on the patient side, then sealed and cut using the LigaSure. In the CG, the conventional method of suture ligation and standard bipolar electrosurgical dissection were used. In all patients, a chest tube was inserted into the anterior mediastinum through a separate substernal incision, and an additional chest tube was placed if the pleural cavity was opened. Postoperatively, patients were managed on the thoracic surgery ward or in the intensive care unit. Drainage tubes were removed when the output was less than 200 mL for a 24-hour period or for a period not longer than 4 days. Postoperative analgesia was managed with patient-controlled intravenous narcotics for the first 48 postoperative hours and then converted to oral pain medication.
Statistical Analysis A power calculation was made before initiation the study. We felt that a 20% reduction in operative time would be a clinically significant difference that might change the
Study Population In all patients, preoperative management included computed tomography scan of the chest, pulmonary function tests, and appropriate cardiac assessment. In each patient, the morphologic tumor size (millimeters) was measured three-dimensionally by the longest transaxial section. Other radiologic exams such as magnetic resonance imaging and/or positron emission tomography were performed if clinically indicated. Invasive diagnostic procedures to obtain a preoperative diagnosis were performed in cases of suspected hematopoietic malignancies such as lymphoma and Hodgkin disease. Fifty patients with resectable anterior mediastinal mass were assessed for eligibility. Of these, 10 patients were excluded as follows: in 7/10 (70%) cases the mass was resected via thoracotomy, 2/10 (20%) patients underwent previous thoracic procedure, and 1/10 (10%) refused to participate; thus, 40 patients were enrolled and randomized in the LG (n ¼ 20) Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
Fig. 1 The Consort diagram template of the present study.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
224
operative management of future patients. With a power of 0.80 and an α of 0.05, this required 17 patients in each group. Data are presented as mean standard deviation or percentage. The differences were assessed by chi-square (qualitative data) or Mann-Whitney test (quantitative data). Correlation between tumor size and operative time was evaluated using Spearman rank correlation test. A value of p < 0.05 was considered statistically significant. MedCalc statistical software Version 12.4.0 was used for analysis.
Results All 40 enrolled patients completed the study. The Consort diagram template is shown in ►Fig. 1.9 In the LG, indications for thymectomy (n ¼ 19) were an asymptomatic thymoma in 4 patients, thymoma with MG in 12, thymic carcinoma in 2, and thymic lipoma in 1. One patient was operated on for metastasis of second primary extragonadal germ cell tumor in the aortopulmonary window. In the TG, indications for thymectomy (n ¼ 18) were an asymptomatic thymoma in 4 patients, thymoma with MG in 12, a thymic carcinoma in 1, and a thymic cyst in 1. One patient was operated on with biopsy-proven diagnosis of lymphangioma; in another patient, the indication for surgery consisted of an unclear nonthymic anterior mediastinal mass, which was discovered to be a teratoma. In patients with MG, the preoperative regimen of anticholinesterase therapy with or without steroids was maintained
Santini et al.
until the time of operation. If the patient could not be stabilized with medication and/or in presence of preoperative risk factors for postoperative ventilation, preoperative plasmapheresis was performed.10 Each patient underwent serial neurologic examination by a referring neurologist in the early postoperative period. Reinstitution of anticholinesterase therapy was based upon the discretion of the neurologist. The surgical technique was similar in both groups. The pericardium was not opened, unless invaded. Complete resection included the tumor with the whole thymus gland and adjacent mediastinal celluloadipose tissue. The mediastinal pleura was not routinely opened, unless to facilitate surgical maneuvers or to excise extensive infiltration of the mediastinal fat. The phrenic nerve was always identified but the recurrent nerve was identified only if necessary. In three cases thymectomy was associated with resection of adjacent pulmonary parenchyma (n ¼ 1; patient in the LG) and of pericardium (n ¼ 2; one patient in the TG and another in the LG). Postoperatively, adjuvant chemo- and/or radiotherapy was applied if indicated. No intraoperative complications or deaths occurred in patients of both groups. Only one patient in the TG died of myasthenic crisis 30 days after the operation. The characteristics of the study population are summarized in ►Table 1. Both groups were well matched for age, tumor size, pathologic diagnosis, and incidence of complications. Among patients with MG, no significant difference was found regarding the preoperative and postoperative data (►Table 2).
Table 1 Characteristics of study population LigaSure groupa
Traditional group
pb
40
20 (50%)
20 (50%)
>0.9
47.9 15.5
47.4 16
51 19
0.3
Male
32
15 (46.8%)
17 (53%)
>0.9
Size (mm)
48.7 8.5
47.9 5.7
49 7.9
0.6
Teratoma
1
–
1 (5%)
>0.9
Thymic lipoma
1
1 (5%)
–
>0.9
Thymic cyst
1
–
1 (5%)
>0.9
Thymoma WHO type A
4
2 (10%)
2 (10%)
0.5
Thymoma WHO type AB
10
4 (20%)
6 (30%)
0.7
Thymoma WHO type B1
10
6 (30%)
4 (20%)
0.7
Thymoma WHO type B2
5
2 (10%)
3 (15%)
>0.9
Thymoma WHO type B3
3
2 (10%)
1 (5%)
>0.9
Thymic carcinoma
3
2 (10%)
1 (5%)
>0.9
Metastasis of second primary extragonadal germ cell tumor
1
1 (5%)
–
>0.9
Lymphangioma
1
–
1 (5%)
>0.9
–
1 (5%)
>0.9
Variables
Total
n (%) Age (y)
Pathologic diagnosis after resection
Complications Myasthenic crisis Abbreviation: WHO, World Health Organization. a Valleylab, Boulder, Colorado, United States. b Calculated using chi-square (qualitative data) or Mann-Whitney test (quantitative data). Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
225
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Santini et al.
Table 2 Characteristics of patients with myasthenia gravis (n ¼ 24) Variables
Total
LG
TG
pa
n (%)
24
12 (50%)
12 (50%)
>0.9
Male
16 (66%)
6 (25%)
10 (41%)
0.3
Age (y)
40 7.5
40 12
39 10
0.4
Duration of disease (mo)
25 4
26.3 4.9
25 4.7
0.3
Positive
23 (96%)
11 (47%)
12 (52%)
0.8
Negative
1
1
–
0.8
I
14 (58%)
8 (33%)
6 (25%)
0.6
IIA
6 (25%)
2 (8%)
4 (17%)
0.6
IIB
3 (12%)
2 (8%)
1 (4%)
0.7
IIIA
1 (4%)
–
1 (4%)
0.8
Antibodies
Osserman classification
Medical treatment Anticholinesterase
20 (83%)
9 (37%)
11 (46%)
0.5
Anticholinesterase with steroids
4 (16%)
3 (12%)
1 (4%)
0.5
Plasmapheresis
8 (33%)
5 (21%)
3 (12%)
0.6
ASA classification I
16 (66%)
7 (29%)
9 (37%)
0.6
II
7 (29%)
5 (21%)
2 (8%)
0.3
III
1
–
1 (4%)
0.8
I
7 (29%)
2 (8%)
5 (21%)
0.3
II
14 (58%)
8 (33%)
6 (26%)
0.6
Masaoka stage
3 (12%)
2 (8%)
1 (4%)
0.8
Postextubation in operative room
III
24 (100%)
12
12
>0.9
ICU stay (d)
1.8 0.5
1.9 0.5
1.8 3.5
0.7
Length hospital stay (d)
6.6 0.7
6.5 0.4
6.7 0.7
0.4
Myasthenic crisis
1
–
1 (4%)
0.8
Abbreviations: ASA, American Society of Anesthesiology; ICU, intensive care unit; LG, LigaSure (Valleylab, Boulder, Colorado, United States) group; TG, traditional group. a Calculated using chi-square (qualitative data) or Mann-Whitney test (quantitative data).
Primary Outcome
Secondary Outcome
The operative time in the LG was significantly shorter than in the TG. Mean operative time in the LG and TG was 104 18.5 vs. 132 13.5, respectively (difference ¼ 28; 95% confidence interval [CI]: 18.6–37.9; p < 0.0001; ►Fig. 2). When considering the entire study population, we found that the tumor size was significantly correlated with the operative time (p ¼ 0.006; r ¼ 0.42; 95% CI: 0.127–0.648; ►Fig. 3A). Among the LG, a significant correlation was found between tumor size and operative time (p ¼ 0.04; r ¼ 0.44, 95% CI: 0.005– 0.742; ►Fig. 3B) that was stronger in the TG (p ¼ 0.02; r ¼ 0.51; 95% CI 0.088–0.778; ►Fig. 3C). The intraoperative blood loss was similar between the two groups; it was 377 78.4 in the LG vs. 407 99.7 in the TG (95% CI: 23.3 to 83.3; p ¼ 0.2; ►Fig. 4).
The chest tube output of the LG was lower than the TG (181.5 77.1 vs. 206 83), but the difference was not significant (95% CI: 22.6 to 72.6; p ¼ 0.2). The LG and TG had similar chest drain duration (3 0.6 vs. 3.1 1.5, respectively; 95% CI: 0.2 to 0.4; p ¼ 0.2) and hospital stay (5.9 0.8 vs. 6.1 0.9; 95% CI: 0.36 to 0.7; p ¼ 0.5). The operative and postoperative outcomes are summarized in ►Table 3.
Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
Discussion LigaSure is an electrothermal bipolar method of vessel sealing introduced in clinical practice in 2004. Many favorable reports have been published since then, and LigaSure has
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
226
Santini et al.
Fig. 2 The operative time of the LigaSure (Valleylab, Boulder, Colorado, United States) group (LG) was significantly shorter than the traditional group (TG). Mean operative time in the LG and in TG was 104 18.5 min vs. 132 13.5 min, respectively (difference ¼ 28; 95% confidence interval: 18.6–37.9; p < 0.0001, Mann-Whitney test).
already gained wide acceptance among general surgeons, endocrine surgeons, urologists, and gynecologists performing open and laparoscopic procedures.4 We started using the LigaSure in 2006 to perform wedge resection, bullectomies, and fissure dissection in open and video-assisted procedures with good clinical results11; yet LigaSure facilitated surgical procedures in only some settings.12,13 In the light of our positive experience, we decided to use it for resecting anterior mediastinal tumors, which has not been reported previously. The rationale of this study was to demonstrate how LigaSure compared with hemostatic traditional strategy for surgical maneuvers in a particular anatomic district such as the anterior mediastinum without increasing the morbidity and mortality rates. Our data showed that LG significantly reduced mean operative time by 28% compared with the traditional hemostatic technique. As expected, a significant correlation was found between the tumor size and the operative time (p ¼ 0.006; r ¼ 0.42). Interestingly, we found that among the TG (p ¼ 0.02; r ¼ 0.51) the correlation between tumor size and operative time was stronger than that observed in the LG (p ¼ 0.04; r ¼ 0.44). In theory, the reduced operative duration associated with use of LigaSure could be due to faster intraoperative hemostatic control without needing to ligate and cut the vessels, and such a phenomenon was particularly evident in case of resection of large masses that required more ligatures. However, the interpretation of our data is complicated due to the difficulty of comparing the two groups in a very practical way. (1) Despite no significant differences regarding the nature of pathology, our study population included a variety of disease with or without MG. (2) The mean size of the tumor was similar between the two groups, but the LG presented three patients with tumor size < 35 mm whereas the TG had four patients with tumor > 60 mm. (3) A standard procedure was followed in all patients, but obviously patients with MG required a different surgical
227
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Fig. 3 The tumor size was significantly correlated with the operative time in (A) the entire study population (p ¼ 0.006; r ¼ 0.42, 95% confidence interval [CI] 0.127 to 0.648), (B) the LigaSure (Valleylab, Boulder, Colorado, United States) group (LG; (p ¼ 0.04; r ¼ 0.44, 95% CI: 0.005–0.742), and (C) the traditional group (TG; p ¼ 0.02; r ¼ 0.51, 95% CI 0.088 to 0.778, Spearman rank correlation test).
strategy compared with patients without MG; and also in patients with MG a different resection was required according to Masaoka stage. (4) To avoid any possible risk of rupture, in the LG an additional ligature was used in vessels > 7 mm in diameter that were closed by a thread on the patient side and then sealed and cut using the LigaSure. LigaSure is proven to Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Santini et al.
Fig. 4 The intraoperative blood loss was similar between the LigaSure (Valleylab, Boulder, Colorado, United States) and the traditional group (377 78.4 mL vs. 407 99.7 mL; p ¼ 0.2, Mann-Whitney test).
safely close vessel up to 7 mm in diameter,3–6,11 but it failed to seal larger vessels. Lacin et al reported that hemorrhage occurred in pulmonary arteries > 9 mm in diameters sealed with LigaSure.14 Lesser et al showed in an experimental setup that LigaSure did not result in complete fusion of the wall layers of pulmonary arteries and the burst pressure of 156 mm Hg in arteries > 5 mm in diameter was 6.4-fold less than after traditional ligation.15 However, these data on pulmonary arteries differ from data on systemic vessels. Harold et al found that the burst pressure of systemic arteries sealed with LigaSure was 4 to 5 mm (601 mm Hg) and 6 to 7 mm (442 mm Hg),16 twofold higher than the measurements of Lesser et al on pulmonary arteries of comparable size.15 Yet, a complete fusion of the vessel walls was seen on histologic findings, possibly because of the different anatomic structure of systemic arteries, which had a thicker muscle layer and more collagenous fibers than pulmonary vessels.16 Another important characteristic of LigaSure was that it made resection of the tumor easier and faster, especially close to the nerves and vascular structures, limiting the blunt dissection maneuvers. LigaSure permitted removal of all thymic tissue and in addition most of the mediastinal parathymic fat and lymphatic tissue, which could contain unrecognized thymic remnants. In some cases the thymus totally or in part descended into the mediastinum posterior rather than anterior to the crossing innominate vein; thus, in addition to the mediastinal dissection, the contiguous pretracheal fat and lymphatic tissues were also easily excised using LigaSure. After mobilization and dissection of the gland, it was used to cut the draining veins in to the innominate vein and the
Fig. 5 LigaSure (Valleylab, Boulder, Colorado, United States) facilitated the dissection of tumor, reducing the blunt maneuvers.
cervical horns of each lobe of the thymus. Finally, LigaSure safety cut the fibrotic remnant that attached the two upper poles of the thymus to the lower lobes of the thyroid gland into the lower neck. An example was reported in ►Fig. 5. However, the surgeon should evaluate the potential risk of injuring nerves and vessels close to tumor during the dissection. A pericardial or thymic cyst may be in close contact with the phrenic nerves; apical tumors may make contact with the stellate ganglion and brachial plexus. There is evidence that the use of every method of coagulation may lead to inadvertent damage to nearby structures through the lateral spread of thermal energy.17,18 Thus, judicious use of LigaSure is advised especially when close to vascular or nerve structures to avoid potential postoperative complications as Horner syndrome or diaphragm and vocal cord paralysis. Several studies showed that thermal injury of the surrounding tissue was much more evident after monopolar diathermy than after new devices such as LigaSure or
Table 3 The operative and postoperative outcomes of two groups
a
Variables
LigaSurea group
Traditional group
pb
Operative time (min)
104 18.5
132 13.5
223
Is a Vessel-Sealing System Useful for Resection of Anterior Mediastinal Mass? Mario Santini1
Alfonso Fiorelli1
Giovanni Vicidomini1
1 Department of Cardio-Thoracic Disease, Second University of Naples,
Thoracic Surgery Unit, Naples, Italy 2 Department of Public Health, Second University of Naples, Section of Pathology, Naples, Italy
Gaetana Messina1
Marina Accardo2
Address for correspondence Mario Santini, MD, Department of Thoracic Surgery, Second University of Naples, P.zza L. Miraglia 2, Naples 80138, Italy (e-mail: [email protected]).
Abstract
Keywords
► LigaSure ► mediastinal resection ► mass
Objective To valuate if the LigaSure (Valleylab, Boulder, Colorado, United States) vessel-sealing system could reduce operative time, intraoperative blood loss, drainage duration, and hospital stay in patients with anterior mediastinal mass undergoing open resection. Methods Forty consecutive patients having resection of anterior mediastinal mass were randomized into two groups according to whether LigaSure was used (n ¼ 20) or not (n ¼ 20). Tumor size, operative time, intraoperative blood loss, chest tube output and duration, length of hospital stay, morbidity, and mortality were prospectively recorded, then intergroup differences were statistically analyzed. Results Both groups were well matched for age, tumor size, pathologic diagnosis, and incidence of complications. LigaSure significantly reduced operative duration (p < 0.0001) compared with the traditional technique but without leading to any significant reduction in intraoperative blood loss (p ¼ 0.2), chest tube output (p ¼ 0.2) and duration (p ¼ 0.2), and length of hospital stay (p ¼ 0.5). Conclusions The reduced operative time using LigaSure translates into less exposure to general anesthesia, which is particularly important for patients with myasthenia and potentially reducing cost.
Introduction Adequate hemostasis is an important step in surgical procedures involving any mediastinal organs to ensure a safe outcome. It may be achieved in a standard matter with surgical ties and/or vessel clips, but it is time-consuming.1 Monopolar or bipolar conventional electrocoagulation systems are faster than the ligation-and-tie technique, but they may damage the surrounding structures due to lateral heat dispersion with immediate or delayed complications such as esophageal injury, phrenic palsy, recurrent laryngeal injury, and other potentially life-threatening complications.2 LigaSure (Valleylab, Boulder, Colorado, United States) desiccates
received May 3, 2013 accepted after revision July 22, 2013 published online December 2, 2013
vascular tissues using a feedback-programmed amount of bipolar diathermy. This method of vessel sealing relies on the application of a precise amount of bipolar electrocoagulation and pressure to the tissue, leading to the denaturation of the collagen and elastin in vessel walls, resulting in a hemostatic seal. The technique is able to seal vessels up to 7 mm in diameter and the seal can withstand three times the normal systolic pressure.3 In addition, compared with conventional electrocoagulation instruments, this vessel-sealing mechanism has a significantly reduced thermal spread profile and therefore a decreased risk of injury to adjacent structures. Many reports showed LigaSure to be safe with reduced operative time in digestive,4 urologic,5 gynecologic,6 and
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1355227. ISSN 0171-6425.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thorac Cardiovasc Surg 2015;63:223–230.
Vessel-Sealing System for Resection of Anterior Mediastinal Mass thoracic procedures.7,8 The goal of the present study was to compare operative and postoperative data of patients undergoing resection of an anterior mediastinal mass using LigaSure versus the traditional hemostatic techniques to assess whether LigaSure could reduce the operative time (primary end point), drainage output and duration, and length of hospital stay (secondary end points) without worsening the postoperative outcome.
Materials and Methods Study Design This single-blinded unicenter randomized study enrolled all consecutive patients with anterior mediastinal mass scheduled for surgery from January 2006 to December 2011. The patients were divided into two groups: the LigaSure group (LG) included patients in whom resection was performed using LigaSure, and the traditional group (TG) included patients in whom hemostasis dissection was performed using traditional methods such as suture ligatures or mono- or bipolar electrocoagulation systems. Patients were unaware of allocation to either intraoperative use of the LigaSure or traditional technique. The technique to be used was assigned to the surgeon in the operating room from a randomized list engendered by statistical software (MedCalc statistical software version 12.4.0, MedCalc Software, Ostend, Belgium). Inclusion criteria were patients (1) with resectable anterior mediastinal mass, (1) fit for open surgery, (3) and who gave written informed consent for participating to the study. Exclusion criteria were (1) use of strategy other than sternotomy for resection, (2) any previous thoracic or mediastinal surgery that might bias operative times, and (3) use of other hemostatic devices in both treatment arms. The time of operation (total minutes from incision to skin closure) and intraoperative blood loss (milliliters) were the primary end points of the study. Secondary end points were chest tube output (milliliters), chest drain duration (days), length of hospital stay (days), morbidity, and mortality. The protocol of this study was approved by the Hospital Ethics Committee of the Second University of Naples.
Santini et al.
and in the CG (n ¼ 20). In addition to other variables, we also evaluated the data of patients with myasthenia gravis (MG) among two groups to avoid MG influencing the outcome results.
Operative Technique In this study, all operations were made by the same senior surgeon who had similar experience in using the LigaSure and traditional hemostatic methods. The same general anesthesia protocol was used in patients in both groups. All operations were performed through sternotomy with single-lumen endotracheal ventilation, without muscle relaxants in patients with MG. The handpiece was the LigaSure Precise unit. The seal width measured 1 to 3 mm. This device was fit with a central cutting blade between the two coagulative ligations (or sealing) and division of the mediastinal vessels or tissue. An insulated layer minimized heat transmission to the exterior jaw surfaces, with the thermal spread typically limited to less than 2 mm. Blood vessels, lymphatics, and tissues were likewise cut with LigaSure. Vessels larger than 7 mm in diameter were ligated by a thread on the patient side, then sealed and cut using the LigaSure. In the CG, the conventional method of suture ligation and standard bipolar electrosurgical dissection were used. In all patients, a chest tube was inserted into the anterior mediastinum through a separate substernal incision, and an additional chest tube was placed if the pleural cavity was opened. Postoperatively, patients were managed on the thoracic surgery ward or in the intensive care unit. Drainage tubes were removed when the output was less than 200 mL for a 24-hour period or for a period not longer than 4 days. Postoperative analgesia was managed with patient-controlled intravenous narcotics for the first 48 postoperative hours and then converted to oral pain medication.
Statistical Analysis A power calculation was made before initiation the study. We felt that a 20% reduction in operative time would be a clinically significant difference that might change the
Study Population In all patients, preoperative management included computed tomography scan of the chest, pulmonary function tests, and appropriate cardiac assessment. In each patient, the morphologic tumor size (millimeters) was measured three-dimensionally by the longest transaxial section. Other radiologic exams such as magnetic resonance imaging and/or positron emission tomography were performed if clinically indicated. Invasive diagnostic procedures to obtain a preoperative diagnosis were performed in cases of suspected hematopoietic malignancies such as lymphoma and Hodgkin disease. Fifty patients with resectable anterior mediastinal mass were assessed for eligibility. Of these, 10 patients were excluded as follows: in 7/10 (70%) cases the mass was resected via thoracotomy, 2/10 (20%) patients underwent previous thoracic procedure, and 1/10 (10%) refused to participate; thus, 40 patients were enrolled and randomized in the LG (n ¼ 20) Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
Fig. 1 The Consort diagram template of the present study.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
224
operative management of future patients. With a power of 0.80 and an α of 0.05, this required 17 patients in each group. Data are presented as mean standard deviation or percentage. The differences were assessed by chi-square (qualitative data) or Mann-Whitney test (quantitative data). Correlation between tumor size and operative time was evaluated using Spearman rank correlation test. A value of p < 0.05 was considered statistically significant. MedCalc statistical software Version 12.4.0 was used for analysis.
Results All 40 enrolled patients completed the study. The Consort diagram template is shown in ►Fig. 1.9 In the LG, indications for thymectomy (n ¼ 19) were an asymptomatic thymoma in 4 patients, thymoma with MG in 12, thymic carcinoma in 2, and thymic lipoma in 1. One patient was operated on for metastasis of second primary extragonadal germ cell tumor in the aortopulmonary window. In the TG, indications for thymectomy (n ¼ 18) were an asymptomatic thymoma in 4 patients, thymoma with MG in 12, a thymic carcinoma in 1, and a thymic cyst in 1. One patient was operated on with biopsy-proven diagnosis of lymphangioma; in another patient, the indication for surgery consisted of an unclear nonthymic anterior mediastinal mass, which was discovered to be a teratoma. In patients with MG, the preoperative regimen of anticholinesterase therapy with or without steroids was maintained
Santini et al.
until the time of operation. If the patient could not be stabilized with medication and/or in presence of preoperative risk factors for postoperative ventilation, preoperative plasmapheresis was performed.10 Each patient underwent serial neurologic examination by a referring neurologist in the early postoperative period. Reinstitution of anticholinesterase therapy was based upon the discretion of the neurologist. The surgical technique was similar in both groups. The pericardium was not opened, unless invaded. Complete resection included the tumor with the whole thymus gland and adjacent mediastinal celluloadipose tissue. The mediastinal pleura was not routinely opened, unless to facilitate surgical maneuvers or to excise extensive infiltration of the mediastinal fat. The phrenic nerve was always identified but the recurrent nerve was identified only if necessary. In three cases thymectomy was associated with resection of adjacent pulmonary parenchyma (n ¼ 1; patient in the LG) and of pericardium (n ¼ 2; one patient in the TG and another in the LG). Postoperatively, adjuvant chemo- and/or radiotherapy was applied if indicated. No intraoperative complications or deaths occurred in patients of both groups. Only one patient in the TG died of myasthenic crisis 30 days after the operation. The characteristics of the study population are summarized in ►Table 1. Both groups were well matched for age, tumor size, pathologic diagnosis, and incidence of complications. Among patients with MG, no significant difference was found regarding the preoperative and postoperative data (►Table 2).
Table 1 Characteristics of study population LigaSure groupa
Traditional group
pb
40
20 (50%)
20 (50%)
>0.9
47.9 15.5
47.4 16
51 19
0.3
Male
32
15 (46.8%)
17 (53%)
>0.9
Size (mm)
48.7 8.5
47.9 5.7
49 7.9
0.6
Teratoma
1
–
1 (5%)
>0.9
Thymic lipoma
1
1 (5%)
–
>0.9
Thymic cyst
1
–
1 (5%)
>0.9
Thymoma WHO type A
4
2 (10%)
2 (10%)
0.5
Thymoma WHO type AB
10
4 (20%)
6 (30%)
0.7
Thymoma WHO type B1
10
6 (30%)
4 (20%)
0.7
Thymoma WHO type B2
5
2 (10%)
3 (15%)
>0.9
Thymoma WHO type B3
3
2 (10%)
1 (5%)
>0.9
Thymic carcinoma
3
2 (10%)
1 (5%)
>0.9
Metastasis of second primary extragonadal germ cell tumor
1
1 (5%)
–
>0.9
Lymphangioma
1
–
1 (5%)
>0.9
–
1 (5%)
>0.9
Variables
Total
n (%) Age (y)
Pathologic diagnosis after resection
Complications Myasthenic crisis Abbreviation: WHO, World Health Organization. a Valleylab, Boulder, Colorado, United States. b Calculated using chi-square (qualitative data) or Mann-Whitney test (quantitative data). Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
225
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Santini et al.
Table 2 Characteristics of patients with myasthenia gravis (n ¼ 24) Variables
Total
LG
TG
pa
n (%)
24
12 (50%)
12 (50%)
>0.9
Male
16 (66%)
6 (25%)
10 (41%)
0.3
Age (y)
40 7.5
40 12
39 10
0.4
Duration of disease (mo)
25 4
26.3 4.9
25 4.7
0.3
Positive
23 (96%)
11 (47%)
12 (52%)
0.8
Negative
1
1
–
0.8
I
14 (58%)
8 (33%)
6 (25%)
0.6
IIA
6 (25%)
2 (8%)
4 (17%)
0.6
IIB
3 (12%)
2 (8%)
1 (4%)
0.7
IIIA
1 (4%)
–
1 (4%)
0.8
Antibodies
Osserman classification
Medical treatment Anticholinesterase
20 (83%)
9 (37%)
11 (46%)
0.5
Anticholinesterase with steroids
4 (16%)
3 (12%)
1 (4%)
0.5
Plasmapheresis
8 (33%)
5 (21%)
3 (12%)
0.6
ASA classification I
16 (66%)
7 (29%)
9 (37%)
0.6
II
7 (29%)
5 (21%)
2 (8%)
0.3
III
1
–
1 (4%)
0.8
I
7 (29%)
2 (8%)
5 (21%)
0.3
II
14 (58%)
8 (33%)
6 (26%)
0.6
Masaoka stage
3 (12%)
2 (8%)
1 (4%)
0.8
Postextubation in operative room
III
24 (100%)
12
12
>0.9
ICU stay (d)
1.8 0.5
1.9 0.5
1.8 3.5
0.7
Length hospital stay (d)
6.6 0.7
6.5 0.4
6.7 0.7
0.4
Myasthenic crisis
1
–
1 (4%)
0.8
Abbreviations: ASA, American Society of Anesthesiology; ICU, intensive care unit; LG, LigaSure (Valleylab, Boulder, Colorado, United States) group; TG, traditional group. a Calculated using chi-square (qualitative data) or Mann-Whitney test (quantitative data).
Primary Outcome
Secondary Outcome
The operative time in the LG was significantly shorter than in the TG. Mean operative time in the LG and TG was 104 18.5 vs. 132 13.5, respectively (difference ¼ 28; 95% confidence interval [CI]: 18.6–37.9; p < 0.0001; ►Fig. 2). When considering the entire study population, we found that the tumor size was significantly correlated with the operative time (p ¼ 0.006; r ¼ 0.42; 95% CI: 0.127–0.648; ►Fig. 3A). Among the LG, a significant correlation was found between tumor size and operative time (p ¼ 0.04; r ¼ 0.44, 95% CI: 0.005– 0.742; ►Fig. 3B) that was stronger in the TG (p ¼ 0.02; r ¼ 0.51; 95% CI 0.088–0.778; ►Fig. 3C). The intraoperative blood loss was similar between the two groups; it was 377 78.4 in the LG vs. 407 99.7 in the TG (95% CI: 23.3 to 83.3; p ¼ 0.2; ►Fig. 4).
The chest tube output of the LG was lower than the TG (181.5 77.1 vs. 206 83), but the difference was not significant (95% CI: 22.6 to 72.6; p ¼ 0.2). The LG and TG had similar chest drain duration (3 0.6 vs. 3.1 1.5, respectively; 95% CI: 0.2 to 0.4; p ¼ 0.2) and hospital stay (5.9 0.8 vs. 6.1 0.9; 95% CI: 0.36 to 0.7; p ¼ 0.5). The operative and postoperative outcomes are summarized in ►Table 3.
Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
Discussion LigaSure is an electrothermal bipolar method of vessel sealing introduced in clinical practice in 2004. Many favorable reports have been published since then, and LigaSure has
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
226
Santini et al.
Fig. 2 The operative time of the LigaSure (Valleylab, Boulder, Colorado, United States) group (LG) was significantly shorter than the traditional group (TG). Mean operative time in the LG and in TG was 104 18.5 min vs. 132 13.5 min, respectively (difference ¼ 28; 95% confidence interval: 18.6–37.9; p < 0.0001, Mann-Whitney test).
already gained wide acceptance among general surgeons, endocrine surgeons, urologists, and gynecologists performing open and laparoscopic procedures.4 We started using the LigaSure in 2006 to perform wedge resection, bullectomies, and fissure dissection in open and video-assisted procedures with good clinical results11; yet LigaSure facilitated surgical procedures in only some settings.12,13 In the light of our positive experience, we decided to use it for resecting anterior mediastinal tumors, which has not been reported previously. The rationale of this study was to demonstrate how LigaSure compared with hemostatic traditional strategy for surgical maneuvers in a particular anatomic district such as the anterior mediastinum without increasing the morbidity and mortality rates. Our data showed that LG significantly reduced mean operative time by 28% compared with the traditional hemostatic technique. As expected, a significant correlation was found between the tumor size and the operative time (p ¼ 0.006; r ¼ 0.42). Interestingly, we found that among the TG (p ¼ 0.02; r ¼ 0.51) the correlation between tumor size and operative time was stronger than that observed in the LG (p ¼ 0.04; r ¼ 0.44). In theory, the reduced operative duration associated with use of LigaSure could be due to faster intraoperative hemostatic control without needing to ligate and cut the vessels, and such a phenomenon was particularly evident in case of resection of large masses that required more ligatures. However, the interpretation of our data is complicated due to the difficulty of comparing the two groups in a very practical way. (1) Despite no significant differences regarding the nature of pathology, our study population included a variety of disease with or without MG. (2) The mean size of the tumor was similar between the two groups, but the LG presented three patients with tumor size < 35 mm whereas the TG had four patients with tumor > 60 mm. (3) A standard procedure was followed in all patients, but obviously patients with MG required a different surgical
227
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Fig. 3 The tumor size was significantly correlated with the operative time in (A) the entire study population (p ¼ 0.006; r ¼ 0.42, 95% confidence interval [CI] 0.127 to 0.648), (B) the LigaSure (Valleylab, Boulder, Colorado, United States) group (LG; (p ¼ 0.04; r ¼ 0.44, 95% CI: 0.005–0.742), and (C) the traditional group (TG; p ¼ 0.02; r ¼ 0.51, 95% CI 0.088 to 0.778, Spearman rank correlation test).
strategy compared with patients without MG; and also in patients with MG a different resection was required according to Masaoka stage. (4) To avoid any possible risk of rupture, in the LG an additional ligature was used in vessels > 7 mm in diameter that were closed by a thread on the patient side and then sealed and cut using the LigaSure. LigaSure is proven to Thoracic and Cardiovascular Surgeon
Vol. 63
No. 3/2015
Vessel-Sealing System for Resection of Anterior Mediastinal Mass
Santini et al.
Fig. 4 The intraoperative blood loss was similar between the LigaSure (Valleylab, Boulder, Colorado, United States) and the traditional group (377 78.4 mL vs. 407 99.7 mL; p ¼ 0.2, Mann-Whitney test).
safely close vessel up to 7 mm in diameter,3–6,11 but it failed to seal larger vessels. Lacin et al reported that hemorrhage occurred in pulmonary arteries > 9 mm in diameters sealed with LigaSure.14 Lesser et al showed in an experimental setup that LigaSure did not result in complete fusion of the wall layers of pulmonary arteries and the burst pressure of 156 mm Hg in arteries > 5 mm in diameter was 6.4-fold less than after traditional ligation.15 However, these data on pulmonary arteries differ from data on systemic vessels. Harold et al found that the burst pressure of systemic arteries sealed with LigaSure was 4 to 5 mm (601 mm Hg) and 6 to 7 mm (442 mm Hg),16 twofold higher than the measurements of Lesser et al on pulmonary arteries of comparable size.15 Yet, a complete fusion of the vessel walls was seen on histologic findings, possibly because of the different anatomic structure of systemic arteries, which had a thicker muscle layer and more collagenous fibers than pulmonary vessels.16 Another important characteristic of LigaSure was that it made resection of the tumor easier and faster, especially close to the nerves and vascular structures, limiting the blunt dissection maneuvers. LigaSure permitted removal of all thymic tissue and in addition most of the mediastinal parathymic fat and lymphatic tissue, which could contain unrecognized thymic remnants. In some cases the thymus totally or in part descended into the mediastinum posterior rather than anterior to the crossing innominate vein; thus, in addition to the mediastinal dissection, the contiguous pretracheal fat and lymphatic tissues were also easily excised using LigaSure. After mobilization and dissection of the gland, it was used to cut the draining veins in to the innominate vein and the
Fig. 5 LigaSure (Valleylab, Boulder, Colorado, United States) facilitated the dissection of tumor, reducing the blunt maneuvers.
cervical horns of each lobe of the thymus. Finally, LigaSure safety cut the fibrotic remnant that attached the two upper poles of the thymus to the lower lobes of the thyroid gland into the lower neck. An example was reported in ►Fig. 5. However, the surgeon should evaluate the potential risk of injuring nerves and vessels close to tumor during the dissection. A pericardial or thymic cyst may be in close contact with the phrenic nerves; apical tumors may make contact with the stellate ganglion and brachial plexus. There is evidence that the use of every method of coagulation may lead to inadvertent damage to nearby structures through the lateral spread of thermal energy.17,18 Thus, judicious use of LigaSure is advised especially when close to vascular or nerve structures to avoid potential postoperative complications as Horner syndrome or diaphragm and vocal cord paralysis. Several studies showed that thermal injury of the surrounding tissue was much more evident after monopolar diathermy than after new devices such as LigaSure or
Table 3 The operative and postoperative outcomes of two groups
a
Variables
LigaSurea group
Traditional group
pb
Operative time (min)
104 18.5
132 13.5
