Role of thymectomy in patients with myasthenia gravis

INTRODUCTION — Myasthenia gravis (MG) is an autoimmune disorder of the postsynaptic neuromuscular junction characterized by fluctuating weakness involving variable combinations of ocular, bulbar, limb, and respiratory muscles. MG has a unique relationship with the thymus gland, which has therapeutic implications in two distinct patient populations.

In a minority of patients, MG is a paraneoplastic manifestation of a thymic neoplasm, most commonly thymoma, and surgical resection of the thymus is indicated both for definitive tumor management and for MG therapy to reduce long-term exposure to pharmacotherapy and to improve outcomes. In the majority of patients without a thymoma (nonthymomatous MG), resection of the nondiseased thymus gland (thymectomy) also has become an established therapeutic modality in selected patients as it also has the potential to reduce long-term exposure to pharmacotherapy and to improve outcomes.

This topic will discuss surgical management for MG in patients with and without a thymoma. The general approach to the treatment of MG is discussed separately. (See "Overview of the treatment of myasthenia gravis".)

PATIENTS WITH A THYMIC MASS

Association between thymoma and myasthenia — MG is a well-recognized paraneoplastic manifestation of thymoma, affecting up to one-half of patients with thymoma. Thymomas are rare neoplasms arising from the thymus in the anterior mediastinum (figure 1). Most patients present between 40 and 60 years of age. Clinical evaluation for MG, including assessment of acetylcholine receptor (AChR) antibody status, is indicated for all patients with confirmed or suspected thymoma. Other autoimmune and paraneoplastic syndromes are also associated with thymoma (table 1) [1]. (See "Clinical presentation and management of thymoma and thymic carcinoma".)

Conversely, approximately 10 percent of patients with MG have an associated thymoma or, more rarely, thymic carcinoma. Mediastinal and chest imaging with computed tomography (CT) scan or magnetic resonance imaging (MRI) is therefore part of the diagnostic evaluation of all patients with newly diagnosed MG to look for a thymoma (algorithm 1). (See "Diagnosis of myasthenia gravis", section on 'Thymomas and other thymic masses'.)

Evaluation and surgical management of thymoma — Patients who have an anterior mediastinal mass identified at the time of MG diagnosis should be referred to a thoracic surgeon. Timely diagnosis and treatment of thymoma is important but should not supersede symptomatic treatment of MG, as patients with MG are generally high risk for surgical procedures due to risk of respiratory compromise. Risk factors for complications and indications for preoperative bridge therapy with intravenous immune globulin (IVIG) or plasmapheresis are discussed below. (See 'Perioperative management' below.)

A definitive diagnosis generally requires a tissue sample, which can be obtained either by biopsy prior to treatment or as part of a planned resection of the entire mass. If thymoma is suspected based on clinical and radiographic features, preoperative biopsy is typically not necessary, and a tissue diagnosis is obtained at the time of resection. The approach to evaluation and tissue diagnosis of a mediastinal mass is reviewed in detail separately. (See "Approach to the adult patient with a mediastinal mass", section on 'Thymic mass'.)

Thymomas are slow-growing tumors with the potential for local invasion and distant metastasis. Definitive treatment of thymoma is indicated in all patients with MG, regardless of antibody status and MG type (eg, generalized, bulbar, ocular). Treatment of thymoma in patients with MG consists of complete tumor excision, including complete resection of the thymus. Multimodality approaches are used for patients with unresectable disease as well as those with thymic carcinoma. Surgical management of thymoma and thymic carcinoma is discussed separately. (See "Clinical presentation and management of thymoma and thymic carcinoma".)

Post-thymectomy myasthenia outcomes — Although data are limited, patients with MG and thymoma tend to be more severely affected at MG diagnosis and do not respond to thymectomy as well as those without thymoma [2,3]. A retrospective study of 207 patients with MG who underwent thymectomy for thymoma found that less than 10 percent of patients were in remission after a mean follow-up of 10 years, and approximately 80 percent were still requiring immunotherapy [2].

PATIENTS WITH NONTHYMOMATOUS MYASTHENIA GRAVIS

Immunopathogenic role of the thymus — The thymus gland is essential for T cell differentiation and the establishment of central tolerance, and removal of the thymus has been recognized as a therapeutic modality in patients with MG since the 1940s. Nevertheless, the role of the thymus in the origin of autoimmunity in patients with MG is not completely understood. An antigenic role is suggested by observations that small populations of "myoid" cells within the thymus express acetylcholine receptor (AChR) on the cell surface [2,4,5]. In addition, pathologic specimens from patients with nonthymomatous AChR-positive MG often demonstrate hyperplasia. These concepts are reviewed separately. (See "Pathogenesis of myasthenia gravis", section on 'The thymus and the origin of autoimmunity'.)

Indications and evidence for thymectomy — For patients with newly diagnosed MG who do not have a thymoma (ie, nonthymomatous MG), the role of thymectomy as a therapeutic intervention for MG depends on AChR antibody status, MG type (generalized versus ocular), and patient age (algorithm 1).

Evaluation for thymectomy is an important aspect of treatment for all patients with generalized MG. A longitudinal survey of adult patients with MG who did not undergo thymectomy reported the most frequent reason for not undergoing thymectomy was failure to discuss the option (35 percent). Most patients (57 percent) reported they would have considered thymectomy if their physician spent time discussing it as an effective therapeutic intervention [6]. (See "Overview of the treatment of myasthenia gravis", section on 'Role and timing of thymectomy'.)

Generalized AChR-positive myasthenia — The evidence supporting thymectomy as a therapeutic modality for nonthymomatous MG is strongest in patients with generalized, AchR-associated MG. For most of these patients, thymectomy is recommended to improve disease control and reduce immunotherapeutic requirements compared with pharmacotherapy alone [4,7-11]. The benefit of thymectomy is not immediate, and thymectomy is not a cure.

We evaluate all patients aged 18 to 50 for thymectomy who have nonthymomatous generalized MG and positive AchR antibodies, in agreement with international consensus guidelines [11]. Thymectomy in these patients may improve clinical outcomes, minimize immunotherapy requirements, and reduce exacerbations and hospitalizations.

The benefits of thymectomy are less certain for some adults older than 50 years old. (See 'Older adults (>50 years)' below.)

Evidence to support a long-term benefit of thymectomy in this patient population consists of meta-analyses of observational studies as well as the results of a single multicenter assessor-blinded randomized trial [12-14]. The MGTX trial screened nearly 7000 patients to enroll 126 patients (ages 16 to 65 years; median age approximately 33 years) with generalized AchR antibody-associated MG and disease duration of less than five years. The most common reasons for ineligibility among screened patients were disease duration >5 years and age outside of the defined range. Patients were randomly assigned to extended transsternal thymectomy plus alternate-day prednisone or alternate-day prednisone alone. The following outcomes were noted [12]:

●The time-weighted average Quantitative Myasthenia Gravis score (with higher scores indicating more severe weakness) over a three-year period was lower for the thymectomy group compared with the prednisone-alone group (6.15 versus 8.99, estimated difference 2.85, 95% CI 0.47-5.22).

●The average requirement for alternate-day prednisone over three years was lower for the thymectomy group (32 versus 54 mg, estimated difference 22 mg, 95% CI 12-32). The proportion of patients requiring immunosuppression with azathioprine was lower for the thymectomy group (17 versus 48 percent, estimated difference 31 percent, 95% CI 16-47).

●The proportion of patients hospitalized for MG exacerbations was lower for the thymectomy group (9 versus 37 percent, estimated difference 28 percent, 95% CI 14-42).

●The proportion of patients who achieved minimal manifestation status (ie, no symptoms or functional limitations from MG but may have some weakness on examination of some muscles) was greater in the thymectomy group at 12 months (67 versus 37 percent) and at 36 months (67 versus 47 percent).

●Outcomes were similar for all prespecified subgroup comparisons, including sex, age <40 years versus ≥40 years, and prior glucocorticoid use.

●The frequency of adverse events over the entire study period was higher in the nonsurgical group. There were no deaths in the thymectomy group and there was one reported complication of thymectomy (2 percent).

●Among 50 patients followed out to five years (45 percent of trial participants), gains in symptom scores, prednisone dosing, and disease status persisted in the thymectomy group [15], although confidence in these data is limited by loss to follow-up and the ongoing glucocorticoid requirement in most patients.

Thus, in this highly selected group, thymectomy was safe, reduced the need for immunotherapy, and increased the likelihood of achieving minimal disease manifestations over a three-year period [12]. Limitations of the trial include the inability to blind patients to the intervention and the relatively small and highly selected nature of the trial participants, which raises some concerns about generalizability.

Meta-analyses published before and after the MGTX trial have also suggested that thymectomy is associated with MG remission and improvement, although confidence in the pooled data is limited by the observational nature of the majority of the studies, heterogenous methodology, and lack of adjustment for confounding [13,14,16-18]. In a 2016 systematic review that identified 22 retrospective studies of patients with nonthymomatous MG, the odds ratio for achieving remission with thymectomy compared with conservative treatment (ie, medication) was 2.44 (95% CI 1.91-3.12); the pooled proportion of patients achieving remission with thymectomy was 31 percent, and the corresponding proportion for conservative treatment was 15 percent [14]. A subsequent meta-analysis that included the MGTX trial, an earlier small single-center trial, and 17 retrospective studies came to similar conclusions [18].

Additional studies are needed to understand whether other factors, beyond antibody status, disease distribution, and age, may help predict benefit from thymectomy. Outcomes after thymectomy appear to be roughly comparable in mild, moderate, and severe MG [13], although the findings of one systematic review of 18 retrospective studies suggested that patients with mild MG preoperatively were more likely to achieve remission after thymectomy than those with more severe disease [19].

The presence of thymic hyperplasia on pathology specimens has been variably associated with benefit from thymectomy, but conventional imaging techniques (computed tomography [CT] and magnetic resonance imaging [MRI]) do not reliably distinguish hyperplastic tissue from normal or atrophic tissue. Further studies are needed to define whether diffusion-weighted imaging (DWI) or other MRI techniques can be used to predict response to thymectomy [20,21].

Seronegative myasthenia — We offer thymectomy to patients with generalized MG who are seronegative (AChR, muscle-specific tyrosine kinase [MuSK] antibody, low-density lipoprotein 4 [LRP4] negative) who otherwise meet the criteria for thymectomy similar to patients with AChR-associated MG (ie, generalized MG, age, and fitness for surgery) [11]. (See 'Generalized AChR-positive myasthenia' above and 'Older adults (>50 years)' below.)

Thymectomy may improve outcomes and reduce intolerable side effects from immunotherapy. However, direct evidence to support a benefit of thymectomy for patients with seronegative MG is more limited, as these patients represent a minority of all patients with MG (approximately 6 to 12 percent), and they were not eligible for the MGTX trial. A retrospective cohort study of these two groups (AChR antibody positive and negative) found a similar response rate after thymectomy with a minimum three-year follow-up period [22].

Older adults (>50 years) — There is limited experience with thymectomy in older adults. While the MGTX trial was open to patients up to age 65 years, the median age in the thymectomy arm was 32 years (range 18 to 63 years), and there were few patients >50 years of age in either arm [12]. In a retrospective single-center study, clinical benefit was reported in all patients who underwent thymectomy for MG either associated with thymoma (28 patients) or with no thymoma (45 early-onset and 28 late-onset MG patients), but complete stable remission was seen more in the early-onset MG cohort (median age 31.1 versus 59.8 years old) [23]. (See 'Generalized AChR-positive myasthenia' above.)

We stratify treatment decisions regarding thymectomy for older adults based on age thresholds, in agreement with international consensus guidelines (algorithm 1) [11]:

●For patients aged 51 to 65 years with nonthymomatous generalized MG and positive AChR antibodies, thymectomy may be offered to selected patients based on symptom severity and operative risks. Patients should be counseled that the benefit of thymectomy in this group is uncertain.

●For patients age >65 years with nonthymomatous generalized MG and positive AChR antibodies, thymectomy is generally not indicated.

In general, older adults are more likely to have atrophic changes in the thymus, which may indicate less of a pathogenic role in the disease process, and surgical risks increase with advancing age. While no specific age limit for thymectomy has been established, we generally advise against thymectomy for most patients age 60 years and older, based on concern that risks of thymectomy outweigh potential benefits in this age range. However, others feel that advanced age is not strictly exclusionary and suggest an individualized approach based upon the individual patient's risk and benefit assessment [8].

Observational evidence includes a single-center retrospective study that identified 43 patients older than 60 years at MG onset who underwent robotic-assisted thymectomy for nonthymomatous MG between 2003 and 2017 [24]. The median length of hospital stay was five days, and pathology most commonly showed thymic involution (86 percent). There was one perioperative death due to aortic dissection. With a median follow-up of 60 months, 20 percent of patients had a good outcome (minimal disease manifestations or better). On multivariable analysis, lack of comorbidities and mild disease before thymectomy were independently associated with good outcome after thymectomy.

MuSK and LRP4 antibody-positive myasthenia — The available evidence, although limited to observational series with significant limitations, does not support a role for thymectomy either for patients with MuSK antibody-associated MG [7,11,25-27] or those with low-density lipoprotein 4 (LRP4) antibody-associated MG [11]. In a series of 110 patients with MuSK-positive MG, 40 patients who underwent thymectomy had a postintervention status comparable to 70 patients who did not undergo surgery [25]. This observation does not exclude the possibility that thymectomy may be beneficial, but many centers, including our own, do not recommend thymectomy in patients without thymoma who have MuSK-positive or LRP4-positive MG.

The role of thymectomy in seronegative MG is discussed separately. (See 'Seronegative myasthenia' above.)

Ocular myasthenia — The role of thymectomy in ocular MG is controversial. While it may be as beneficial as in those with generalized disease, some centers do not advise thymectomy for patients with ocular MG [28]. Other centers, particularly those where the less invasive extended transsternal procedure is performed (including the author's center), offer this as a therapeutic option. Thymectomy for ocular MG is discussed in greater detail separately. (See "Ocular myasthenia gravis", section on 'Thymectomy'.)

Surgical considerations

Operative approach — The goal of thymectomy in patients with nonthymomatous MG is an extended resection to remove the thymus itself as well as surrounding mediastinal and cervical fat, which contains variable amounts of ectopic thymic tissue [29]. The surgical approach should maximize resection of this ectopic thymic tissue while avoiding damage to the recurrent laryngeal, left vagus, and phrenic nerves.

The traditional and most invasive approach is an extended transsternal thymectomy via a median sternotomy, which allows full exploration to completely remove all thymic tissue and associated fat. This was the approach used in the MGTX trial, which is the basis for the recommendation for thymectomy in patients with nonthymomatous MG [12]. (See 'Generalized AChR-positive myasthenia' above.)

Practice since the MGTX trial has shifted toward less invasive approaches to thymectomy, which we now prefer over traditional open thymectomy in most cases. These approaches appear as effective as open thymectomy and have lower morbidity and faster recovery times [11,30]. However, they may provide less access to extracapsular thymic tissue in surrounding adipose tissue. Patients should be counseled that the effectiveness of less invasive approaches is less well established than the effectiveness of extended transsternal thymectomy [13].

The less invasive approaches include:

●Video-assisted or robotic-assisted thymectomy via a transthoracic approach (eg, video-assisted thoracoscopic surgery [VATS]) [31-33]. (See "Thymectomy", section on 'Minimally invasive thymectomy'.)

●Extended transcervical thymectomy through a low horizontal neck incision [34-37]. Although this technique may not adequately expose the entire thymus, thereby increasing risk of residual thymus being left behind, proponents argue that with video assistance, surgical exposure is adequate. (See "Thymectomy", section on 'Transcervical thymectomy'.)

There are only limited data comparing transsternal thymectomy with less invasive approaches in patients with nonthymomatous MG, making it impossible to say with confidence that they are equally effective [10]. In retrospective studies, however, MG outcomes appear similar across different techniques [37-39]. There is no doubt that the less invasive approaches have a lower morbidity and shorter length of hospitalization than traditional transsternal thymectomy [40,41]. In addition, approaches that avoid splitting the sternum avoid potential sternotomy complications. (See "Thymectomy", section on 'Open versus minimally invasive thymectomy' and "Surgical management of sternal wound complications".)

Timing of surgery — Thymectomy in patients with MG is an elective procedure and should not be performed in the setting of myasthenic crisis or otherwise poorly controlled symptoms. All patients require a complete preoperative evaluation by a multidisciplinary team involving the surgeon, anesthesiologist, and neurologist. Symptoms of MG should be under good control if possible, with minimal bulbar and respiratory symptoms. It is probably also helpful to taper the glucocorticoids as low as the clinical status allows in order to reduce postoperative infections and problems with wound healing. (See "Anesthesia for the patient with myasthenia gravis", section on 'Preoperative evaluation'.)

Some have suggested that the response to thymectomy is better if thymectomy is performed early in the course of the disease. This is based on remission rates that are greater if performed early after the onset of disease than if performed later [42]. However, the rate that patients go into remission is higher early in disease. Thus, any intervention performed earlier would appear to do better than if it were performed later in the disease course [13,28]. The MGTX trial excluded patients with disease duration greater than five years [12]. Despite the lack of a proven benefit of early thymectomy, thymectomy is usually recommended by most experts within the first three years of the diagnosis [7,9].

Perioperative management — Thymectomy in patients with MG is an elective procedure and should not be performed in the setting of myasthenic crisis or otherwise poorly controlled symptoms. All patients require a complete preoperative evaluation by a multidisciplinary team involving the surgeon, anesthesiologist, and neurologist.

Patients are at increased risk for surgical complications due to the disease itself, medications used for anesthesia, and the potential for interruptions in maintenance MG therapies. Reported rates of myasthenic crisis after thymectomy range from 2 to 25 percent in various series [43-45]. The most consistently identified risk factors for postoperative myasthenic crisis or need for prolonged ventilation are preoperative respiratory and bulbar dysfunction [46-49]. The preoperative evaluation of patients with MG is discussed further separately. (See "Anesthesia for the patient with myasthenia gravis", section on 'History'.)

For patients with well-controlled MG and minimal disease manifestations, pyridostigmine and immunotherapy, including maintenance intravenous immune globulin (IVIG) if applicable, should be continued through the perioperative period. Patients with significant respiratory impairment or dysphagia should not undergo surgery until MG is under better control.

In the occasional patient with persistent mild residual respiratory impairment or dysphagia despite treatment with pyridostigmine and immunotherapy (eg, glucocorticoids), we administer a course of IVIG (2 grams/kg over two to five days) or plasmapheresis (three to five exchanges over 7 to 14 days) preoperatively. This practice is empiric, with a goal of reducing the risk of a postoperative flare evolving quickly into myasthenic crisis. Treatment should be timed to end the week prior to surgery so that the effects of the rapid therapy peak and persist through the perioperative period. In the case of plasmapheresis, this also allows time for coagulation factors removed by the exchange to recover. (See "Overview of the treatment of myasthenia gravis", section on 'Plasma exchange and IVIG as rescue or bridge therapies'.)

Extra vigilance is warranted in the postoperative period for patients with risk factors for myasthenic crisis. (See "Myasthenic crisis" and "Anesthesia for the patient with myasthenia gravis".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Myasthenia gravis" and "Society guideline links: Thymomas and thymic carcinomas".)

SUMMARY AND RECOMMENDATIONS

●Relationship between thymus and myasthenia gravis – In approximately 10 percent of patients, myasthenia gravis (MG) is a paraneoplastic manifestation of a thymic neoplasm, most commonly thymoma. All patients with newly diagnosed MG should have chest imaging to screen for thymoma. (See 'Association between thymoma and myasthenia' above.)

●Thymectomy for patients with thymoma – Definitive treatment of thymoma is indicated in all patients with MG, regardless of antibody status and MG type (eg, generalized, bulbar, ocular). (See 'Evaluation and surgical management of thymoma' above.)

●The role of thymectomy for patients without a thymoma – For patients with newly diagnosed MG who do not have a thymoma (ie, nonthymomatous MG), the role of thymectomy as a therapeutic intervention for MG depends on acetylcholine receptor (AChR) antibody status, MG type (generalized versus ocular), and patient age (algorithm 1). The benefit of thymectomy is not immediate, and thymectomy is not a cure.

•For patients 18 to 50 years of age with nonthymomatous generalized AChR antibody-associated MG, we recommend thymectomy (Grade 1B). (See 'Generalized AChR-positive myasthenia' above.)

•For patients age 18 to 50 years with nonthymomatous generalized seronegative MG (ie, no detectable AChR, muscle-specific tyrosine kinase [MuSK], or low-density lipoprotein 4 [LRP4] antibodies), we suggest thymectomy (Grade 2C). (See 'Seronegative myasthenia' above.)

•For patients 51 to 65 years of age with nonthymomatous generalized AChR antibody-associated MG or seronegative MG, we suggest thymectomy to selected patients based on symptom severity and operative risks (Grade 2C). Patients should be counseled that the benefit of thymectomy in this group is uncertain. (See 'Older adults (>50 years)' above.)

•For patients >65 years of age with nonthymomatous generalized AChR antibody-associated MG or seronegative MG and for all patients with MuSK or LRP4 antibody-associated MG, we suggest not performing thymectomy (Grade 2C). (See 'MuSK and LRP4 antibody-positive myasthenia' above and 'Older adults (>50 years)' above.)

•The role of thymectomy for some patients with ocular MG is controversial. (See "Ocular myasthenia gravis", section on 'Thymectomy'.)

●Surgical approaches – Surgical approaches to thymectomy include open transsternal (midline sternotomy) and less invasive methods with a transthoracic (thoracoscopic) or extended transcervical approach. Less invasive approaches are associated with lower morbidity and faster recovery but, in some cases, may provide less access to extracapsular thymic tissue. (See 'Surgical considerations' above.)

•For patients with MG and a thymoma, treatment consists of complete tumor excision, including complete resection of the thymus. Multimodality approaches are used for patients with unresectable disease as well as those with thymic carcinoma.

•For patients with nonthymomatous MG who are selected for thymectomy, we suggest a minimally invasive approach rather than traditional median sternotomy (Grade 2C).

●Surgical timing – Thymectomy in patients with MG is an elective procedure and should not be performed in the setting of myasthenic crisis or poorly controlled symptoms. For patients with mild residual respiratory or bulbar symptoms despite optimized symptomatic and glucocorticoid therapy, we suggest preoperative treatment with a rapid immunotherapy (ie, plasmapheresis or intravenous immune globulin [IVIG]) (Grade 2C). (See 'Surgical considerations' above.)