Determinants of Suboptimal Outcome Following Thymectomy in Myasthenia Gravis
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.314565
Source of Support: None, Conflict of Interest: None
Keywords: Determinants, myasthenia gravis, outcome, remission, thymectomyKey Message: Nearly a third of myasthenia gravis patients had suboptimal outcome following thymectomy. Complete stable remission was attained in 14.8% after a mean follow up of 4.3 years. Thymoma was an independent predictor of suboptimal outcome after thymectomy.
Acquired myasthenia gravis (MG) is a chronic autoimmune disorder of neuromuscular junction transmission characterized by fluctuating weakness and fatigability. Despite being one of the most prolifically researched autoimmune diseases, various aspects of the disease course and therapeutic response remain undefined.
The course and severity of MG is highly variable. Natural history studies have shown that the maximum disease severity occurs in nearly 85% of the patients within seven years of disease onset. While 10%–20% report spontaneous remissions, at the other end of the spectrum, nearly 20% develop life-threatening myasthenic crisis. Remissions are often incomplete with 58%–66% patients requiring immunomodulatory therapy beyond eight years of symptom onset.
First performed and propagated by Alfred Blalock in 1940s, various modifications in the removal of thymus gland are being practiced worldwide, namely, cervical approach, video-assisted thoracoscopic thymectomy and maximal thymectomy. The role of thymus in MG pathogenesis was suggested by the detection of pathological changes in thymus gland in nearly 80% of MG patients, especially in acetylcholine receptor antibody (AChR-Ab) positive cases. Immunohistological studies in the thymus in MG demonstrate the expression of autoantibodies in the thymic epithelial cells and myoid cells, infiltration with autoreactive T and B cells and reduction of regulatory T cells. Gene expressions favorable to the development of an autoimmune reaction are also noted in the thymic tissue. Thymic tissue in myasthenic patients shows abundant CD4+ T cells specific for AChR., The evidence for pathogenic role of thymus in MG with antibodies other than those against AChR is more tenuous.
The benefits of thymectomy are not uniform across all subgroups of MG. Prior retrospective studies on thymectomized patients identified multiple factors influencing the likelihood of clinical remission. Poor attainment of remission was reported with thymomatous MG,,,,,, thymic carcinoma and later age of disease onset,,,,, in some studies whereas others failed to reveal any significant association of these factors with the outcome.,,,,,,,,,,,,,, The recognition of these factors is essential for the prognostication and drug modification following thymectomy in MG.
We reviewed the records of patients with MG. They underwent maximal thymectomy in our center over the past 15 years to identify the frequency and determinants of suboptimal outcome with combined medical and surgical therapy in them.
We retrospectively screened the records of all patients with MG who underwent maximal thymectomy at our center from January 2000 to December 2014 for eligibility for inclusion in the study. Thymectomized patients aged 18 and above whose pathology slides were available for review and had a minimum follow-up period of six months postoperatively were included in the study. The study was approved by the Institutional Ethics Committee in January 2014.
Diagnosis of MG was based on the presence of typical clinical features of fluctuating and fatigable weakness with classical distribution and one or both of decrement of CMAP amplitude of 10% or more in repetitive nerve stimulation and AChR-Ab positivity. The demographic and clinical profile, details of surgery, medical therapy, and follow-up data were collected using a structured pro forma. The preoperative maximum clinical severity was graded with the Myasthenia Gravis Foundation of America (MGFA) clinical classification. Age of onset ≤40 years was designated as early onset MG and above 40 years as of late onset MG.
A uniform protocol for medical management was followed in the center. Symptomatic patients with ocular or generalized MG were initiated on choline-esterase inhibitory medication (pyridostigmine) with those with more severe illness or suboptimal response being started on concomitant prednisolone therapy. In patients whom the duration of steroid therapy was expected to be prolonged or with severe steroid-related side effects, one of the immunosuppressant medications, azathioprine or mycophenolate mofetil, was added. Plasma exchange or intravenous immunoglobulin was used in patients with myasthenic crisis or impending crisis.
Maximal thymectomy was performed for all patients to ensure complete removal of regular and ectopic thymic tissue in the mediastinum and neck. The surgical strategy adopted was uniform and all the patients were considered for surgery only after achievement of at least minimal manifestation status on medications.
Meticulous preoperative preparation, operative technique, and postoperative care were employed for all the patients. On the day of surgery, the patient was premedicated with 5 mg of diazepam along with all antimyasthenic drugs. General anesthesia was induced with midazolam and fentanyl and maintained on 1%–2% isoflurane and 50% oxygen and nitrous oxide.
Access to mediastinum was obtained with a median sternotomy. Thymic lobes were separated in the superior mediastinum and left innominate vein delineated and protected after ligating thymic vein (s). In the neck, upper pole of each thymic lobe was cleared from under the strap muscles followed by intentional entry into the right pleural cavity. The right lateral thymic lobe including tumor, if any, and parathymic prepericardial fat from diaphragm were excised in continuity with appropriate care taken to keep a safe distance from phrenic nerve. Similar approach was continued on the left side with extreme caution not to use electrocautery to safeguard phrenic nerve, particularly when dissection proceeded cephalad toward thoracic inlet. After obtaining hemostasis and placing basal pleural drainage tubes, sternotomy was repaired in usual fashion.
The status of the phrenic nerves was affirmed by noting the position of the diaphragm in a chest radiogram after transiently disconnecting ventilator. In general, uncomplicated patients could be discharged on the eighth postoperative day on the same doses of all the medications the patient was on preoperatively.
The MGFA clinical class of the patients and the doses of pyridostigmine, prednisolone, and other immunosuppressant medicines were collected one month before thymectomy, six months after the surgery, annually for the next five years, and at the final follow-up visit. The occurrence and precipitating factors for crisis before and after thymectomy were also noted.
Surgical outcomes were assigned according to the criteria of the Task Force of the Medical Scientific Advisory Board of the MGFA. Complete stable remission (CSR) was defined as the persistent disappearance of all myasthenic symptoms without any ongoing treatment for at least one year. Pharmacologic remission (PR) was defined as the disappearance of all MG symptoms with either ongoing low-dose immunosuppressive and/or pyridostigmine treatments and MM as the presence of weakness without functional limitation with or without therapy. Patients were classified into “improvement group” if steroid dose was reduced by 30% of initial dose or patient had significant improvement in clinical symptoms with the same or reduced dose of steroid, “unchanged” if there was no significant change in clinical status and dose of medicines, and “worsened” if the patient was initiated on new immunotherapy or more than 30% increase in baseline medication was required postsurgery to control the symptoms. For analysis, we grouped patients with CSR, PR, MM, and improvement status together as good outcome group and unchanged and worsened status as poor outcome group.
Serological and pathological data
The data on the antibody status of the patients were also recorded wherever available. Patients were grouped into seropositive or seronegative based on the presence or absence, respectively, of AChR-Ab. Thymic pathologies were reviewed by our pathologist and classified into follicular hyperplasia, thymoma, involuted, or normal.
Statistical analyses were performed using IBM SPSS Statistics for Windows, version 21.0, Armonk, NY: IBM Corp. USA. Descriptive analysis was performed to express the continuous variables as mean ± standard deviation and categorical variables as numbers and percentages. Statistical comparisons between different subgroups were by student's t-test for means and by Chi-square tests or Fisher's exact tests for proportions. A P value of less than 0.05 was considered significant.
During the 15-year period, 187 patients with MG underwent thymectomy, of whom 59 with inadequate follow-up were excluded from the study. The 128 patients included in the analysis had a male–female ratio of 1.06:1 and mean age of onset of 38.98 (±12.29) years. MG was early onset in 68 (53.13%) patients.
The demographic and clinical details of the patients are illustrated in [Table 1]. Nearly half (49.22%) had moderate weakness (MGFA III) by MGFA clinical classification preoperatively. There were no patients with pure ocular myasthenia. Comorbidities were noted in 26 patients with other autoimmune disorders occurring in 6 (three autoimmune thyroiditis, two nephrotic syndrome, and one rheumatoid arthritis). None of the patients had extrathymic malignancies on follow-up.
The AChR-Ab status was available in 82 (64.06%) of the 128 patients of whom 69 (84.15%) were AChR-Ab positive. Muscle specific kinase antibody (MuSK-Ab) status was available only in 20 patients and all were negative for this.
Following maximal thymectomy, there was no mortality and none of the patients had immediate postoperative crisis. The pathological analysis of the specimens showed thymomas in 66 (51.56%) patients. Postthymectomy patients with thymomatous MG were significantly older than nonthymomatous patients (P = 0.006). The comparison between thymomatous and nonthymomatous groups is detailed in [Table 2].
Following thymectomy, both groups required significantly lower doses of prednisolone. At final follow-up (mean follow-up 51.68 ± 33.21 months), the thymomatous group were on a mean daily prednisolone dose of 17.77 ± 17.70 mg compared to 9.78 ± 15.97 mg in the nonthymomatous group (P = 0.009). The dose of prednisolone was significantly more in the thymoma group and the difference was more pronounced between the second and fourth years of therapy [Figure 1]. Of the 84 patients who were on prednisolone prior to thymectomy, 30 patients (35.7%) could successfully discontinue the drug at last follow-up. Withdrawal of steroid was more often possible in nonthymomatous group (48.8% versus 23.3%, P = 0.015). Thirty-six (28.13%) patients received steroid-sparing immunosuppressant medications following surgery—33 azathioprine and 3 mycophenolate mofetil.
CSR was attained by 19 (14.84%) patients. Eighty-eight (68.75%) patients had good outcome, whereas 40 (31.25%) had poor outcome till the final follow-up. The number of patients attaining CSR increased with every year of follow-up with more than half achieving the maximum improvement by five years.
Suboptimal response was noted in 40 (31.25%) patients at final follow-up. Eleven (8.6%) patients had myasthenic crisis after thymectomy, four within the first six months. Among the patients who developed crisis after 6 months, five were related to drug tapering or drug default, one due to acute high-dose steroid treatment and one related to respiratory infection. Presence of crisis preoperatively was not a predictor for crisis after surgery.
On univariate analysis, the presence of thymoma was associated with suboptimal remission or improvement compared to nonthymomatous patients (P = 0.001). The sex, age of onset, preoperative disease severity, seropositive status, and presence of comorbidities were not found to have any significant relation to the postthymectomy outcome [Table 3]. The use of steroids prior to surgery was found to be associated with a better outcome (P = 0.035), although there was no difference in the baseline MG severity between the two groups [Table 3]. A higher preoperative dose of prednisolone was also associated with good outcome (P = 0.035).
Ethnic influences on the course and outcome of MG has been reported previously in several studies.,,, In this study, we examined the determinants of suboptimal outcome patients following maximal thymectomy from a single tertiary care center from South India with specific focus on the influence of thymic pathology and drug titration. Although thymectomy produced benefit in both thymomatous and nonthymomatous MG patients, the former group had a relatively less favorable outcome.
The clinical benefit of thymectomy is thought to be derived by eliminating the role of thymus in the sensitization to AChR and the subsequent self-sustained autoimmune reaction. It has been noticed that for the initial one year postthymectomy, clinical remission may not be apparent and in a few patients, the maximum effect may manifest five years or more after the surgery. This delay in achievement of remission is theoretically attributable to the persistence of sensitized B and T cells in circulation after the removal of thymus.
Till lately, the role of thymectomy in nonthymomatous MG was ambiguous. Multiple nonrandomized data showed conflicting results and major methodological flaws existed in the individual studies., With the results of the Thymectomy Trial in Nonthymomatous Myasthenia Gravis Patients Receiving Prednisone Therapy (MGTX) published recently, thymectomy for nonthymomatous MG with elevated AChR-Ab levels is now firmly established.
Comparing thymectomy outcomes between different studies has limitations because of the significant variability in the protocols for selection of cases, techniques of surgery, classification of patient outcome, and follow-up between the studies. A set of criteria validated by the MGFA to overcome these constraints was employed in this study. CSR is the most commonly used index to measure postthymectomy outcome as it is the least amenable to interobserver variation.
In this study, we used both CSR and overall clinical improvement, i.e., good outcome, as the main postthymectomy outcome measure to compare with other studies. The subjects in our study had mean follow-up of 51.68 months after surgery. While the overall improvement rate of 68.75% was comparable to other large studies,,,,, the CSR rate was slightly lower than those reported in other studies (21% to 57.8% in 5 years).,,, The lower proportion of CSR in this study is partly attributable to the preference of many patients to retain a small dose of choline-esterase inhibitor medications even after achieving complete clinical improvement. The higher number of thymoma patients in this group may also have translated to less frequent CSR. Some studies showed that more patients attained CSR with increasing duration of follow-up., However, the course of MG is highly unpredictable and relapses occur after periods of inactivity and especially when immunotherapy is discontinued as was demonstrated in a study from western India. This study also demonstrated the highest yearly remission rate in the first year of the disease which then declined.
The decision on titration and withdrawal of cholinergic and immunosuppressant therapy in MG patients is often physician-guided and based on local or national protocols. In our institute, a slow taper of drugs is practiced following achievement of symptom suppression. We meticulously examined the pattern and response to drug titration following thymectomy in this study. We noted a steep decline in the pyridostigmine and prednisolone doses in the first two years following surgery for both groups following which a slower titration pattern was followed. In the thymoma group, in the fourth and fifth years of follow-up, many required a hike of medications, a pattern which was not observed in the nonthymomatous group. This underlines the higher likelihood of thymoma patients to relapse on drug withdrawal. In this study, the nonthymomatous MG group had significantly higher rates of CSR (25.80% versus 4.54%) and overall clinical improvement than the thymoma group (83.7% versus 54.54%, P < 0.001). This agrees with the previous observations reporting a poorer outcome in thymoma patients.,,,,,,
In the earlier studies, other consistent factors found to be associated with a good outcome were young age of onset (less than 45 years),,,,,,, female sex,, shorter duration from onset to thymectomy,, and preoperatively ocular or mild MG.,,, However, none of these factors attained statistical significance in this study. This may partly be due to the higher preoperative disease severity and higher proportion of thymoma patients in our study. AChR-Ab positive patients did not significantly differ from the seronegative patients in outcome. The presence of ectopic thymus that can occur in up to 48% patients who undergo thymectomy was reported as a very significant negative predictor for CSR. An interesting observation in our study was that preoperative steroid use predicted a better outcome. A similar conclusion was noted with preoperative high-dose steroid therapy in a retrospective Japanese cohort where the patients who received preoperative steroids had significantly higher probability of achieving pharmacological remission. The effects of thymectomy are experimentally shown to be synergistic rather than independent of pharmacological immune modulation. The impact of the timing of the immune modulation on these effects needs to specifically be addressed in future studies.
Myasthenic crisis after thymectomy was heretofore reported most commonly within the first six months of thymectomy and hence in part attributed to the effects of surgery itself.,, In this study, there were no immediate perioperative myasthenic crises and most of the crises occurred during the period of drug withdrawal suggesting ongoing disease activity as the etiology. Contrary to the previous reports, we did not notice any impact of the premorbid disease severity or preoperative crisis after thymectomy. The consistent achievement of an optimal disease control prior to surgery, may explain our observations.
Our study had some notable limitations. The study was retrospective and hence had an inherent selection bias. The follow-up period and the medication use were variable between the patients. AChR-Ab status was not available in 42 patients and MuSK-Ab status was not available in majority of the patients prior to thymectomy.
In conclusion, following maximal thymectomy, nearly one-third patient may fail to show satisfactory clinical improvement by five years of follow-up. The major determinant for a suboptimal outcome was the presence of thymoma, although both thymomatous and nonthymomatous patients benefited from the procedure. Following a strict institute protocol for MG control preoperatively and during surgery can help minimize postoperative complications. The study raises a few stimulating questions on the ideal synergistic medical management protocol in the preoperative and postoperative periods in MG patients that require well-planned prospective studies to resolve.
Financial support and sponsorship
This research was funded by an internal student grant from Sree Chitra Tirunal Institute for Medical Sciences and Technology.
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3]