Risk Factors for Developing Postthymectomy Myasthenia Gravis in Thymoma Patients Yoshito Yamada, MD, PhD, Shigetoshi Yoshida, MD, PhD, Takekazu Iwata, MD, PhD, Hidemi Suzuki, MD, PhD, Tetsuzo Tagawa, MD, PhD, Teruaki Mizobuchi, MD, PhD, Naoki Kawaguchi, MD, PhD, and Ichiro Yoshino, MD, PhD Department of General Thoracic Surgery, Chiba University Graduate School of Medicine, Chiba, Japan; and Douwa-kai Neurological Center, Chiba, Japan
MG, the AchR-Ab levels at the onset of MG were significantly higher than the preoperative levels (p [ 0.036). The analysis of the 33 antibody-positive patients showed a significant correlation between antibody levels and the onset of disease (p [ 0.003). The analysis of 12 patients with incomplete resection revealed that perioperative chemotherapy or radiotherapy reduced the risk of development of MG (p [ 0.009). Conclusions. A positive result for serum AchR-Ab, type B1/B2/B3 thymoma histology, and incomplete resection were considered risk factors for the development of postthymectomy MG. Preoperative levels of serum AchR-Ab should be examined in all thymoma patients. Perioperative chemotherapy or radiotherapy for the patients with incomplete resection might reduce the risk of MG development.
M
analysis of thymoma patients who had undergone thymectomy at our institute.
yasthenia gravis (MG) is an autoimmune disease caused by acetylcholine receptor antibodies (AchRAb) that block acetylcholine receptors at the postsynaptic neuromuscular junction. Thymoma occurs in 10% to 12% of patients with MG, and 15% of thymoma patients have MG [1–4]. Thymoma is a low-grade malignant tumor, and surgical resection is the gold standard for treating thymoma. In the preoperative evaluation of thymoma patients, the possibility of MG should be investigated because perioperative management of concomitant MG is crucial [5]. However, MG develops in some thymoma patients after a thymectomy (hereafter called postthymectomy MG) despite having no history of MG or symptoms [3, 6–11]. It is, therefore, important to clarify the clinical aspects and the mechanisms of postthymectomy MG, not only for perioperative management, but also for determination of the clinicopathology of thymoma and MG. This study aimed to identify the risk factors for the development of postthymectomy MG in a retrospective Accepted for publication Oct 31, 2014. Address correspondence to Dr Yamada’s current affiliation, Division of Thoracic Surgery, University Hospital Zurich, Raemistrasse 100, 8091 Zurich, Switzerland; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;99:1013–9) Ó 2015 by The Society of Thoracic Surgeons
Patients and Methods We carried out a retrospective review of the medical charts of patients who had been diagnosed with thymoma and had undergone a thymectomy between 1991 and 2011 at our institute. We extracted data for patients with no history of MG at the time of their surgery. Patients with thymic carcinoma were excluded. Clinical variables reviewed were age, sex, clinical history, spirometry results, surgical procedures (approaches and methods), serum AchR-Ab levels, postoperative complications, thymoma histologic type, postoperative thymoma outcome, use of perioperative chemotherapy or radiotherapy, and any evidence of postthymectomy MG. In clinical practice, serum AchR-Ab levels are measured in all cases, and patients with antibody seropositivity or any symptoms suggestive of MG are referred to neurologists. Neurologists diagnose patients with MG when (1) subjective or objective symptoms of MG are detected; (2) one or more positive results are confirmed in physical examinations, such as the edrophonium test, a 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.10.068
GENERAL THORACIC
Background. Thymoma patients with no prior history of myasthenia gravis (MG) occasionally have MG after thymectomy. This study aimed to identify risk factors for developing postthymectomy MG. Methods. We retrospectively reviewed the characteristics and clinical outcomes of thymoma patients without preoperative MG who underwent a thymectomy at our institute. Results. Of the 229 thymoma patients who underwent surgery from 1991 to 2011, 123 had no history of MG. The serum of all patients was analyzed for acetylcholine receptor antibodies (AchR-Ab), and 33 patients had detectable levels (greater than 0.2 nmol/L). Ten seropositive patients had MG develop postoperatively within 3 to 2,859 days, whereas 90 seronegative patients did not. Univariate analysis showed that a positive result for serum AchR-Ab (p < 0.001), type B1/B2/B3 thymoma histology (p [ 0.025), and incomplete resection (p [ 0.008) were risk factors for the development of MG. In patients with post-thymectomy
GENERAL THORACIC
1014
YAMADA ET AL POSTTHYMECTOMY MG IN THYMOMA PATIENTS
serum antibody test, single fiber electromyography, and the Harvey-Masland test; and (3) other diseases are ruled out. Treatments for MG are initiated only if MG is diagnosed. The basic procedure for thymoma was an extended thymectomy, which included the removal of all mediastinal fatty tissue between the two bilateral phrenic nerves from the top of the inferior horns of the thyroid to the diaphragm; in some cases, a limited resection was performed, for example, in a patient with a small, undiagnosed tumor and AchR-Ab seronegativity. The approach was sternotomy in most patients, but video-assisted thoracic surgery (VATS) was also performed in recent years if extended thymectomy was feasible. Regular postoperative follow-up, including a physical examination and a radiologic examination, was carried out every 6 to 12 months. A neurology check-up was undertaken by neurologists when any symptoms of MG were suggested. For cases before 2004, the pathologic diagnosis was reviewed again by pathologists according to the 2004 revised World Health Organization classification [12, 13]. For cases before 2000, the MG status was reviewed from clinical records and neurologic findings according to the Myasthenia Gravis Foundation of America (MGFA) classification [14]. Before this study, the Ethics Review Board of our institution (Institutional Review Board) examined and approved our research protocol, which adhered to the Declaration of Helsinki. Because of the retrospective nature of the study, individual patient consent was not obtained, but a summary of this study was posted for patients on the bulletin board in our outpatient ward. The results for each variable were expressed as mean and standard deviation. The AchR-Ab was evaluated based on both continuous and categorical variables, using the accepted cutoff value of 0.2 nmol/L. The effect of the continuous variables on the development of postthymectomy MG was assessed by the Cox regression test. The hazard ratio indicates an increase in the hazard for development of MG when the independent variable increases by 1 unit. It was reported with 95% confidence interval. For categoric variables, Kaplan-Meier curves for
Fig 1. Tree diagram of the thymoma patients in this study. The scheme shows patients categorized by preoperative (preop) and postoperative (postop) thymectomy myasthenia gravis (MG) and preoperative level of acetylcholine receptor antibody (AchR-Ab). (Minus symbol [] ¼ without MG; plus symbol [þ] ¼ with MG.)
Ann Thorac Surg 2015;99:1013–9
the development of postthymectomy MG over time were generated, and the 10-year freedom from postthymectomy MG was reported, including the standard error. The effect was assessed by log rank tests. The Wilcoxon signed rank test or paired t test was used to compare AchR-Ab levels at different time points according to the results of test of normality. A p value of 0.05 or less was considered significant. The statistical analyses were performed using IBM SPSS Statistics, version 22 (IBM Corp, Armonk, NY).
Results Of the 229 patients who underwent surgery for thymoma, 123 (53.7%) had no history of MG at the time of surgery, and 106 (46.3%) were diagnosed with MG (Fig 1). During the postoperative follow-up period (mean 2,672 days; range, 52 to 7,741), postthymectomy MG developed in 10 patients between 3 and 2,859 days after surgery. During our observation period, 1 patient died of thymoma and 4 died of other diseases. There were no MG-related deaths. Table 1 shows the characteristics of all patients who had no preoperative evidence of MG, and the univariate analysis for the development of postthymectomy MG. For the continuous variables, preoperative level of serum AchR-Ab (HR 1.15; 95% confidence interval: 1.08 to 1.22; p < 0.001) was related closely with the development of the disease. For the categoric variables, preoperative serum AchR-Ab positivity, type B1/B2/B3 thymoma histology, and incomplete resection were risk factors for the development of the disease. The 10-year freedom from MG for AchR-Ab positive patients and AchR-Ab negative patients was 61.5% 10.2% and 100%, respectively (p < 0.001; Fig 2A); for patients with type A/AB thymoma histology or type B1/B2/B3, it was 94.7% 4.0% and 81.1 6.3%, respectively (p ¼ 0.025; Fig 2B); and for patients with complete or incomplete resection, it was 91.2% 3.4% and 65.5 16.7%, respectively (p ¼ 0.008; Fig 2C). Surgical procedures for the patients with postthymectomy MG were sternotomy in 9 and VATS in 1; surgical procedures for the patients without the disease were sternotomy in 98 and VATS in 15. For factors
preop MG(-) 123
preop AchR-Ab (-) 90 preop AchR-Ab(+) 33
Thymectomy 229 preop MG(+) 106
postop MG(-) all
postop MG(-) 23 postop MG(+) 10
Ann Thorac Surg 2015;99:1013–9
YAMADA ET AL POSTTHYMECTOMY MG IN THYMOMA PATIENTS
1015
Table 1. Patient Characteristics and Univariate Analysis for the Development of Postthymectomy Myasthenia Gravis Characteristics
HRa (95% CI) or 10-Year Freedom From MG (% SE)
p Valueb
123 58.1 12.9
1.001 (0.95–1.05)
0.98
62 61 2.1 6.0
85.2 5.4 92.0 4.7 1.147 (1.08–1.22)
0.19
90 33
100 61.5 10.2
MG, the AchR-Ab levels at the onset of MG were significantly higher than the preoperative levels (p [ 0.036). The analysis of the 33 antibody-positive patients showed a significant correlation between antibody levels and the onset of disease (p [ 0.003). The analysis of 12 patients with incomplete resection revealed that perioperative chemotherapy or radiotherapy reduced the risk of development of MG (p [ 0.009). Conclusions. A positive result for serum AchR-Ab, type B1/B2/B3 thymoma histology, and incomplete resection were considered risk factors for the development of postthymectomy MG. Preoperative levels of serum AchR-Ab should be examined in all thymoma patients. Perioperative chemotherapy or radiotherapy for the patients with incomplete resection might reduce the risk of MG development.
M
analysis of thymoma patients who had undergone thymectomy at our institute.
yasthenia gravis (MG) is an autoimmune disease caused by acetylcholine receptor antibodies (AchRAb) that block acetylcholine receptors at the postsynaptic neuromuscular junction. Thymoma occurs in 10% to 12% of patients with MG, and 15% of thymoma patients have MG [1–4]. Thymoma is a low-grade malignant tumor, and surgical resection is the gold standard for treating thymoma. In the preoperative evaluation of thymoma patients, the possibility of MG should be investigated because perioperative management of concomitant MG is crucial [5]. However, MG develops in some thymoma patients after a thymectomy (hereafter called postthymectomy MG) despite having no history of MG or symptoms [3, 6–11]. It is, therefore, important to clarify the clinical aspects and the mechanisms of postthymectomy MG, not only for perioperative management, but also for determination of the clinicopathology of thymoma and MG. This study aimed to identify the risk factors for the development of postthymectomy MG in a retrospective Accepted for publication Oct 31, 2014. Address correspondence to Dr Yamada’s current affiliation, Division of Thoracic Surgery, University Hospital Zurich, Raemistrasse 100, 8091 Zurich, Switzerland; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;99:1013–9) Ó 2015 by The Society of Thoracic Surgeons
Patients and Methods We carried out a retrospective review of the medical charts of patients who had been diagnosed with thymoma and had undergone a thymectomy between 1991 and 2011 at our institute. We extracted data for patients with no history of MG at the time of their surgery. Patients with thymic carcinoma were excluded. Clinical variables reviewed were age, sex, clinical history, spirometry results, surgical procedures (approaches and methods), serum AchR-Ab levels, postoperative complications, thymoma histologic type, postoperative thymoma outcome, use of perioperative chemotherapy or radiotherapy, and any evidence of postthymectomy MG. In clinical practice, serum AchR-Ab levels are measured in all cases, and patients with antibody seropositivity or any symptoms suggestive of MG are referred to neurologists. Neurologists diagnose patients with MG when (1) subjective or objective symptoms of MG are detected; (2) one or more positive results are confirmed in physical examinations, such as the edrophonium test, a 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.10.068
GENERAL THORACIC
Background. Thymoma patients with no prior history of myasthenia gravis (MG) occasionally have MG after thymectomy. This study aimed to identify risk factors for developing postthymectomy MG. Methods. We retrospectively reviewed the characteristics and clinical outcomes of thymoma patients without preoperative MG who underwent a thymectomy at our institute. Results. Of the 229 thymoma patients who underwent surgery from 1991 to 2011, 123 had no history of MG. The serum of all patients was analyzed for acetylcholine receptor antibodies (AchR-Ab), and 33 patients had detectable levels (greater than 0.2 nmol/L). Ten seropositive patients had MG develop postoperatively within 3 to 2,859 days, whereas 90 seronegative patients did not. Univariate analysis showed that a positive result for serum AchR-Ab (p < 0.001), type B1/B2/B3 thymoma histology (p [ 0.025), and incomplete resection (p [ 0.008) were risk factors for the development of MG. In patients with post-thymectomy
GENERAL THORACIC
1014
YAMADA ET AL POSTTHYMECTOMY MG IN THYMOMA PATIENTS
serum antibody test, single fiber electromyography, and the Harvey-Masland test; and (3) other diseases are ruled out. Treatments for MG are initiated only if MG is diagnosed. The basic procedure for thymoma was an extended thymectomy, which included the removal of all mediastinal fatty tissue between the two bilateral phrenic nerves from the top of the inferior horns of the thyroid to the diaphragm; in some cases, a limited resection was performed, for example, in a patient with a small, undiagnosed tumor and AchR-Ab seronegativity. The approach was sternotomy in most patients, but video-assisted thoracic surgery (VATS) was also performed in recent years if extended thymectomy was feasible. Regular postoperative follow-up, including a physical examination and a radiologic examination, was carried out every 6 to 12 months. A neurology check-up was undertaken by neurologists when any symptoms of MG were suggested. For cases before 2004, the pathologic diagnosis was reviewed again by pathologists according to the 2004 revised World Health Organization classification [12, 13]. For cases before 2000, the MG status was reviewed from clinical records and neurologic findings according to the Myasthenia Gravis Foundation of America (MGFA) classification [14]. Before this study, the Ethics Review Board of our institution (Institutional Review Board) examined and approved our research protocol, which adhered to the Declaration of Helsinki. Because of the retrospective nature of the study, individual patient consent was not obtained, but a summary of this study was posted for patients on the bulletin board in our outpatient ward. The results for each variable were expressed as mean and standard deviation. The AchR-Ab was evaluated based on both continuous and categorical variables, using the accepted cutoff value of 0.2 nmol/L. The effect of the continuous variables on the development of postthymectomy MG was assessed by the Cox regression test. The hazard ratio indicates an increase in the hazard for development of MG when the independent variable increases by 1 unit. It was reported with 95% confidence interval. For categoric variables, Kaplan-Meier curves for
Fig 1. Tree diagram of the thymoma patients in this study. The scheme shows patients categorized by preoperative (preop) and postoperative (postop) thymectomy myasthenia gravis (MG) and preoperative level of acetylcholine receptor antibody (AchR-Ab). (Minus symbol [] ¼ without MG; plus symbol [þ] ¼ with MG.)
Ann Thorac Surg 2015;99:1013–9
the development of postthymectomy MG over time were generated, and the 10-year freedom from postthymectomy MG was reported, including the standard error. The effect was assessed by log rank tests. The Wilcoxon signed rank test or paired t test was used to compare AchR-Ab levels at different time points according to the results of test of normality. A p value of 0.05 or less was considered significant. The statistical analyses were performed using IBM SPSS Statistics, version 22 (IBM Corp, Armonk, NY).
Results Of the 229 patients who underwent surgery for thymoma, 123 (53.7%) had no history of MG at the time of surgery, and 106 (46.3%) were diagnosed with MG (Fig 1). During the postoperative follow-up period (mean 2,672 days; range, 52 to 7,741), postthymectomy MG developed in 10 patients between 3 and 2,859 days after surgery. During our observation period, 1 patient died of thymoma and 4 died of other diseases. There were no MG-related deaths. Table 1 shows the characteristics of all patients who had no preoperative evidence of MG, and the univariate analysis for the development of postthymectomy MG. For the continuous variables, preoperative level of serum AchR-Ab (HR 1.15; 95% confidence interval: 1.08 to 1.22; p < 0.001) was related closely with the development of the disease. For the categoric variables, preoperative serum AchR-Ab positivity, type B1/B2/B3 thymoma histology, and incomplete resection were risk factors for the development of the disease. The 10-year freedom from MG for AchR-Ab positive patients and AchR-Ab negative patients was 61.5% 10.2% and 100%, respectively (p < 0.001; Fig 2A); for patients with type A/AB thymoma histology or type B1/B2/B3, it was 94.7% 4.0% and 81.1 6.3%, respectively (p ¼ 0.025; Fig 2B); and for patients with complete or incomplete resection, it was 91.2% 3.4% and 65.5 16.7%, respectively (p ¼ 0.008; Fig 2C). Surgical procedures for the patients with postthymectomy MG were sternotomy in 9 and VATS in 1; surgical procedures for the patients without the disease were sternotomy in 98 and VATS in 15. For factors
preop MG(-) 123
preop AchR-Ab (-) 90 preop AchR-Ab(+) 33
Thymectomy 229 preop MG(+) 106
postop MG(-) all
postop MG(-) 23 postop MG(+) 10
Ann Thorac Surg 2015;99:1013–9
YAMADA ET AL POSTTHYMECTOMY MG IN THYMOMA PATIENTS
1015
Table 1. Patient Characteristics and Univariate Analysis for the Development of Postthymectomy Myasthenia Gravis Characteristics
HRa (95% CI) or 10-Year Freedom From MG (% SE)
p Valueb
123 58.1 12.9
1.001 (0.95–1.05)
0.98
62 61 2.1 6.0
85.2 5.4 92.0 4.7 1.147 (1.08–1.22)
0.19
90 33
100 61.5 10.2
