Minimally invasive thymectomy for patients with thymoma: surgical, oncological and neurological outcomes

Abstract

Background

Traditionally, sternotomy has been the gold standard approach for the treatment of thymomas. However, interest in minimally invasive techniques such as video-assisted and robot-assisted thymectomy is gaining momentum. Concerns have been raised over the possibility of en-bloc resection using minimal access techniques due to limited working space and increased tumour manipulation leading to tumour breach and recurrence.

Methods

An observational cohort study was conducted at a tertiary referral centre from 1 April 2012 to 31 December 2020 and followed up until 31 March 2023. Consecutive patients of thymoma were evaluated for demography, symptoms and imaging preoperatively and included for the study. Post minimally invasive thymectomy, surgical, neurological and oncological outcomes were evaluated through out-patient and telephonic follow-up.

Results

Fifty-two patients underwent minimally invasive thymectomy. Video-assisted thoracoscopic surgery for thymectomy was used in 49 patients (94.2%) with 3 (5.8%) patients undergoing robot-assisted thoracoscopic surgery. Median tumour size was 4 cm (interquartile range (IQR) 3–5) with a median operative duration of 150 min (IQR 120–180), blood loss of 135 ml (IQR 42.5–250), post-operative stay of 3 days (IQR 2–4) and no 30-day surgery related morbidity and mortality. Annual contrast-enhanced computed tomography imaging, available in 46 (88.5%) patients, showed no recurrence at a median follow-up of 43 (IQR, 21–75) months. No symptoms suggesting recurrence was noted at a median clinical follow-up of 57 (IQR 44–95.5) months and 88.5% patients were expected to survive a period of 10 years.

Conclusion

Minimally invasive thymectomy is technically feasible with minimal morbidity and acceptable intermediate-term oncological outcomes in patients suffering with thymoma.

Trial registration

Institute ethical committee approval: Ref no. IECPG-551/14.11.2018.

Clinical Trial Registry of India: Ref no. CTRI/2019/04/018784.

Introduction

Thymomas are relatively rare tumours arising from the thymus gland with an incidence of approximately 0.13 cases/100000 person years [1]. They are slow growing tumours associated with a variety of paraneoplastic and autoimmune conditions, most commonly myasthenia gravis (MG), in up to 25–40% patients [2, 3].

Surgical excision is the mainstay treatment. This involves en-bloc removal of the gland with the tumour, all perithymic tissue in the pre-pericardial space from inferior border of thyroid to diaphragm in between both phrenic nerves and the removal of any adjacent structures involved by the tumour when necessary. R0 resection is the most important prognostic factor for patients undergoing thymectomy for thymoma [4–8]. Extended thymectomy via sternotomy has been considered the standard approach in the management of thymoma [9, 10].

Application of minimally invasive techniques in thoracic surgery offered advantages of less pain, smaller scars and early return to work. This led to an emerging interest among the surgeons to utilise minimally invasive techniques to perform thymectomy. However, the adoption of minimally invasive thymectomy (MIT) in managing thymomas has been slow and cautious due to oncological concerns [11]. It has been a common belief that MIT limits the operative space in the anterior mediastinum which may lead to increased manipulation of the tumour with potential for capsular breach, tumour seeding and thus recurrence [12]. The issue of oncologic safety and neurologic outcomes in patients with thymoma and MG undergoing MIT is not settled and the surgeon’s expertise and discretion has generally determined the use of this technique [13].

We have been performing minimally invasive thymectomy for MG and have established the safety and efficacy of the technique [14]. The present study was undertaken to evaluate the role of MIT with respect to surgical, oncological and neurological outcomes in patients presenting with thymoma with or without MG.

Materials and methods

Methodology

This is an observational cohort study conducted at a tertiary care hospital in India. All patients of thymoma with or without MG who were deemed suitable for MIT were offered surgery via video-assisted thoracoscopic surgery (VATS) or robot-assisted thoracoscopic surgery (RATS) approach. Consenting patients were included in the study. Patients with large thymomas not amenable to use of minimal access techniques, invasive thymomas, thymic carcinomas, uncontrolled MG and those who had received neo-adjuvant chemotherapy and/or radiotherapy were excluded from the study.

A total of 54 patients were considered for MIT from 1 April 2012 to 31 December 2020. Two patients underwent conversion to sternotomy. Finally, 52 patients were followed up until 31 March 2023 and analysed for the outcomes. Ethical clearance was obtained from the Institute Ethics Committee (Ref No: IECPG-551/14.11.2018) and registered with the Clinical Trials Registry of India (Ref No: CTRI/2019/04/018784).

Data on demography, history, physical findings and preoperative workup was recorded in a proforma. Patients with MG were evaluated and classified according to the Myasthenia Gravis Foundation of America (MGFA) clinical classification, and their medicine requirements were recorded. A contrast-enhanced computed tomogram (CECT) of the thorax was routinely performed pre-operatively for diagnosis and characterization of lesion. After adequate pre-operative optimization, patients were operated using MIT. Intraoperative details recorded included the approach, side of entry, operative time, blood loss, extent of resection, tumour size and weight, need for conversion to open technique and intraoperative difficulty if any. Post-operative details in terms of intercostal tube drain (ICD) indwelling time, ICD output, post-operative crisis in myasthenic patients, complications (if any), length of hospital stay and perioperative mortality were noted. Patients were followed up in the out-patient department and through telephone for 30-day outcomes and subsequently at 6 months, 1 year and then annually for tumour recurrence using a CECT scan. Patients with MG were additionally evaluated for status of MG symptoms based on Myasthenia Gravis Foundation of America–Post Intervention Status (MGFA-PIS) and change in status scales. During these visits, history and physical examination were also recorded for evaluation of myasthenic symptoms and symptoms of any tumour recurrence such as voice changes, cough, breathing difficulty, features of major airway, vascular, cardiac compression or metastasis. Drug dose modifications during this period were also recorded for patients with myasthenia gravis.

Definition of successful outcomes

Surgical outcome

Absence of conversion to open and absence of significant perioperative morbidity (Clavien-Dindo ≥ 2) or mortality.

Oncological outcome

Absence of clinical and/or radiological evidence of tumour recurrence.

Neurological outcome

Complete stable remission or pharmacological remission for at least 1 year and improvement in change of status as defined by the MGFA-PIS score.

Statistics

Data was collected and tabulated on Microsoft® Excel Version 16.43. Analysis of the data was performed on IBM® SPSS® version 25. Collected data on categorical scale were presented as number and percentage (%) while data on continuous scale were presented as median (interquartile range (IQR)). Pre- and postoperative dosage of drugs were compared using the Wilcoxon’s signed rank test. A p-value of < 0.05 was considered significant. Kaplan–Meier analysis was used to evaluate predicted survival.

Results

Of the 54 patients who underwent thymectomy during the study period, 52 patients were included for analysis. Two (3.7%) patients required conversion to sternotomy and were excluded from the analysis. One conversion was because of dense adhesions, while bleeding led to conversion in the other. Both patients did not have any perioperative morbidity and had a stable post-operative period. Baseline demographics, clinical evaluation and tumour characteristics are given in Table 1. Fifty-one (98%) patients had MG with ptosis (45, 86.5%) and limb weakness (41, 78.8%) being the most common symptoms at presentation. Thirteen (25%) patients had at least one episode of crisis during the course of their disease before surgery and all of them received plasmapheresis while 53.84% patients received intravenous immunoglobulin (IVIG) therapy in addition.

Table 1.

Baseline demographics

Parameter (n = 52)	Number (percentage) Median (IQR)
Age	44 (37.5–50)
Gender	M(56%)/F(44%)
Symptoms
- Ptosis   - Limb weakness   - Dysphagia   - Diplopia   - Chewing difficulty   - Dysphonia   - Nasal regurgitation	- 86.5% - 78.8% - 61.5% - 50.0% - 42.3% - 32.7% - 28.8%
MGFA clinical classification
- 1   - 2A   - 2B   - 3A   - 3B   - 4   - 5   - Non-myasthenic	- 3.8% - 21.2% - 25% - 15.4% - 17.3% - 9.7% - 5.7% - 1.9%
Masaoka stage
• 1  • 2A  • 2B  • 3	33 (63.5%) 13(25%) 4(7.7%) 2 (3.8%)
WHO stage
• A  • AB  • B1  • B2  • B3	6 (11.5%) 12 (23.1%) 3 (5.8%) 21 (40.4%) 10 (19.2%)
Tumour size (cm)	4 (3–5)

Surgical outcomes

VATS was the most commonly used technique with 49 (94.2%) patients undergoing the procedure and 3 (5.8%) undergoing RATS thymectomy. There was no 30-day mortality. Post-operative myasthenic crisis was seen in 5 (9.8%) patients. Two of these patients needed prolonged intubation and intensive care unit (ICU) stay of 1 and 10 days respectively. There were no other major surgical complications. Other surgical outcomes are outlined in Table 2.

Table 2.

Surgical outcomes

Parameter (n = 52)	Number (percentage) Median (IQR)
Procedure
• VATS   • RATS	49 (94.2%) 3 (5.8%)
Side approach
• U/L   • B/L	47 (88.5%) 5 (9.6%)
Operative time (minutes)	150 (120–180)
Blood loss (ml)	135 (42.5–250)
ICD indwelling time (days)	2 (1–3)
ICD Amount (ml)	165 (70–357.5)
Post-operative complications	9.8% (crisis)
LOHS (days)	3 (2–4)

Oncological outcomes

No patient had any clinical evidence of recurrence at a median follow-up of 57 (IQR 44–95.5) months. Annual CECT imaging was available for 46 (92%) patients. Two patients died within a year of surgery, one due to progressive myasthenia and the other due to stroke. Two (4%) patients were lost to follow-up and 2 (4%) patients refused follow-up imaging since they were symptom free (Table 3). Among the 46 patients who had annual CECT scan, none had any evidence of recurrence at a median follow-up of 43 (IQR 21–75) months.

Table 3.

Oncological outcomes

Parameter	Number (percentage) Median (IQR)
Clinical follow-up (months)	57 (44–95.5)
Clinical recurrence (n = 52)	0
Imaging on follow-up (n = 50)
• Available   • Pending   • Lost to F/U	46 (92%) 2 (4%) 2 (4%)
Recurrence on imaging (n = 46)	0
Imaging follow-up (months)	43 (21–75)

Neurological outcomes

Successful outcome in terms of improved MGFA change in status was seen in 66.7% patients while complete stable remission (CSR)/pharmacological remission (PR) was seen in 23.5% cases. Worsening of symptoms was seen in 2 patients while 6 patients died during follow-up due to exacerbation of myasthenia or other reasons. Further details on neurological outcomes are available in Table 4.

Table 4.

Neurological outcomes

Parameters (n = 51)	Number (percentage)
MGFA—post intervention status
• CSR   • PR   • MM   • Died from Myasthenia Gravis   • Died from other causes	5 (9.8%) 7 (13.7%) 33 (64.7%) 3 (5.9%) 3 (5.9%)
MGFA change in status
• Improved   • Unchanged   • Worsened   • Died	34 (66.7%) 9 (17.7%) 2(3.9%) 6 (11.7%)

Reduction in the need of medications post-operatively was seen with 53.6% patients going off steroids and 25.4% patients going off pyridostigmine (Fig. 1). Patients who continued to take anti-myasthenic medications had a significant reduction in the dosage of pyridostigmine (240 (IQR 180–270) mg to 120 (IQR 0–180) mg (p < 0.0001)) and prednisolone (20 (3.5–37.5) mg to 0 (0–5) mg (p < 0.0001)). However, there was no significant change in the azathioprine dosage (87.5 mg (0–125) to 87.5 mg (0–100), p = 0.86) (Fig. 2). Of 34 patients requiring 3 drugs pre-operatively for myasthenia gravis, 20 (59%) patients needed 2 or less medications with 6 (17.6%) needing no medication post thymectomy. Thus, only 14 (41%) patients required 3 medications in the follow-up to control MG symptoms (Fig. 3).

Fig. 1.

Drugs stopped post-thymectomy

Fig. 2.

Change in drug dose requirement

Fig. 3.

Change in number of medicines post thymectomy

At a median follow-up of 57 months, 6 (11.53%) patients died. The cause of death being myasthenic crisis in 3 and cardiovascular event in the other 3. Time-dependent survival analysis using the Kaplan–Meier analysis showed an 88.5% probability of survival at 10 years (Fig. 4).

Fig. 4.

Post-operative predicted survival

Discussion

Our study evaluated the surgical, oncological and neurological outcomes of 54 patients who underwent MIT for thymoma in the presence or absence of MG.

The median age of our patients is 44 years which is earlier than the 6th–7th decade reported with isolated thymoma [3]. All but one patient in our study had thymoma with myasthenia gravis (98.1%) due to referral from neurology and may explain an earlier age as it has been reported that thymoma associated with MG presents in 4th/5th decade [3].

Fifty-two patients underwent minimal access thymectomy with a success rate of 96.3%. Two (3.7%) out of the 54 patients initially recruited for the study had to be converted to open technique following bleeding in one and adhesions in the other. Conversion rates have been reported to range from 1.3 to 13.11%, with bleeding, adhesions and advanced disease with adjacent organ involvement being the most common reasons for conversion [12, 15, 16].

An en-bloc extended thymectomy was performed in all our patients. We resected the ipsilateral phrenic nerve in 1 (1.9%), pericardium in 7 (13.46%) and adjacent lung tissue in 1 (1.9%) in order to achieve R0 resection. Others too have reported need for en-bloc resection of phrenic nerve in up to 11.8%, pericardium in 2–9% and adjacent lung tissue in 2–6% patients undergoing MIT for thymoma [16–18]. The median operative time was 150 min, with median blood loss of 135 ml and ICD indwelling time of 2 days. These findings corroborate with other studies having operative times ranging from 65 to 207 min, blood loss of 20–200 ml and ICD requirement of 1.3–4.13 days respectively [12, 15, 16, 19].

We had no major 30-day surgical morbidity or mortality in our patients. Complication rates have been reported to range from 0% to 22.7% for patients undergoing thymectomy via minimal access approach [16] with a metanalysis by Friedant et al. reporting a pooled mean complication rate of 2.4% [12]. Although most studies have reported no mortality [15–17, 19], a study by Liu et al. reported mortality rate of 2.6% amongst 76 VATS operated patients of thymoma with or without myasthenia in their retrospective series of 120 patients comparing open with VATS thymectomy [20].

Post-thymectomy crisis has been reported in nearly 3–30% patients [17, 21] with the presence of thymoma being an independent risk factor for the development of post thymectomy crisis in two recent metanalysis. In the same studies, minimal access techniques were seen to have favourable results with respect to post thymectomy crisis [21, 22]. Five (9.8%) of our patients developed myasthenic crisis in the post-operative period. One patient required prolonged intubation and ICU stay for a period of 10 days while others were managed with drug dose escalation, intravenous immunoglobulin or plasmapheresis. All recovered and were safely discharged. Mean post-operative hospital stay in this group of patients was of 4.8 days when compared to a median of 3 days overall.

The median length of hospital stay in our study was 3 days and it reflects on the true benefit of minimal access surgery and its role in thymectomy. Mean length of hospital stay has ranged from 1 to 10.6 days [16] with a recent study from the national cancer database of reporting a median stay of 3 days similar to our study [23].

The tumour size on pathological evaluation ranged from 2 to 10 cm with a median of 4 cm. Ninety-six percent had Masaoka Stage I/II tumours. In a metanalysis of the International Thymic Malignancy Interest Group (ITMIG) database comparing open thymectomy (OT) and MIT for thymoma without MG, a median tumour size of 5.5 cm ranging from 0.4 to 15.5 cm was noted. On propensity matching, they found equivalent R0 resection rates by either approach and did not find tumour size to be a deterrent to use of minimal access approach [24]. A cutoff size of 5 cm was considered safe for the MIT approach in thymoma by Kimura et al. This cut off was based on three (6.7%) recurrences of tumours larger than 5 cm. Two of these recurrences were attributed to intraoperative tumour capsule breach while manipulation [25]. A retrospective review by the Japanese Association for the Research of the Thymus (JART) group including 2793 patients from 1991 to 2010 found significant increase in recurrence for tumours larger than 5 cm and increased tumour-related deaths for tumours larger than 8 cm [26]. However, another recent study from the US national cancer database found no difference in R0 resection or overall survival amongst patients with tumours larger than 4 cm in patients undergoing minimal invasive thymectomy for stage I–III tumours [23]. The definition of a safe tumour size, thus, continues to be controversial and further prospective studies with longer outcomes are necessary to clearly identify the oncological safety of minimal access surgical techniques with respect to tumour size.

None of our patients evaluated clinically or using a CECT thorax had a recurrence of thymoma at a median follow-up of 56.5 months and 43 months respectively. Recurrence data is sparse in literature as thymoma are slow-growing tumours and most studies have reported short- to intermediate-term outcomes. Various studies have reported recurrence rates of 2.6–5.9% at follow-up ranging from 12 to 78 months [24]. Recurrence rates in a metanalysis by Hess et al. was 0.83% at a follow-up of under 1.8–6 years, [16] while the ChART group reported a recurrence of was 2.9% at a median follow-up of 33.5 months [27]. A single-centre retrospective review following VATS thymectomy reported a recurrence rates of 3.5% at 5 years with a 89.3% follow-up rate [28]. These findings indicate that minimal-access thymectomy may be a safe alternative to the standard open thmyectomy technique as far as oncological outcomes are concerned.

Overall projected 10-year survival in our study is 88%. This included all causes of mortality, including the ones not specific to disease. Two studies reporting outcomes of MIT showed a survival rate of 84 aand 90% at 5 years [29, 30]. Another study by Hwang et al. published in 2022 compared the outcomes of OT (784) and MIT (455) in patients with early or locally advanced thymomas from the data collected from database of Korean Association for Research on Thymus (KART) showed no significant differecne between the two groups with respect to the overall survival at 10 years which was 87.7% and 85.5% in OT and MIT respectively [31]. This corroborates with the predicted outcomes as calculated in our study.

The role of thymectomy in non-thymomatous MG has been established following the recent MGTX groups trial published in the New England Journal of Medicine [32]. While some studies have shown that the presence of thymoma portends a poor outcome compared to those with a normal or hyperplastic thymus post-operatively with respect to resolution of myasthenic symptoms [5, 33, 34], others have not found thymoma to be predictor of poor outcome in myasthenia gravis [14, 35–38]. Improvement in myasthenic status according to the MGFA-PIS change in status category was noted in 34 (66.7%) of our patients with 41.1% reduction in the number of patients taking three drugs preoperatively. Fifty-four percent of patients went off steroids while 25.4% of patients went off acetylcholinesterase inhibitors with a significant reduction of drug dosage. A study by Li et al. evaluated the outcome of VATS thymectomy. The study included 150 patients of thymoma of which 44 had MG. They noted neurological improvement in 19 (52.8%) patients with deterioration in 8 (25%) out of the 36 patients who were followed up [28]. In another study from China, comparing neurological outcomes of VATS thymectomy amongst thymomatous MG and MG with thymic hyperplasia, an 8% drop in steroid use was seen postoperatively amongst thymomatous myasthenia patients at a mean follow-up of 25.3 months. They had CSR rates of 26%; however, severe MG (MGFA grade IV) were excluded in the study and most patients had isolated ocular myasthenia (MGFA Grade I) [38]. A recent multi-institution study by the Korean Association of Research on Thymus (KART) group evaluated long-term neurological outcomes in 253 patients with thymomatous MG. Their crude CSR and PR rates were 25.9% at 50.4 months which was similar to our finding of 23% at 63.88 months [39]. Thus, it is evident that thymectomy in patients of thymomatous myasthenia gravis leads to improvement in neurological parameters.

Our study presents a reasonable number of patients who underwent minimal access thymectomy for thymoma with surgical, neurological and oncological outcome analysis at a median follow-up of almost 5 years and projected survival analysis of 10 years. Our study demonstrates that minimally invasive thymectomy is a safe modality for carefully selected patients with thymoma with no evidence of recurrence at follow-up of over 5 years. Patients with myasthenia gravis in addition to thymoma are seen to have clinical improvement in their neurological status in nearly two-thirds with a large proportion of patients benefitting from the decrease in the number of medications required to control neurological symptoms.

Limitations

Being an ambispective study, there are certain weaknesses in this study. Firstly, it has an inherent potential for bias and also does not have a control arm to compare the results with. The other weakness was that the recruitment in the prospective arm and regular imaging follow-up was hampered due to the corona virus disease (COVID) pandemic during the study period.

Conclusion

Minimally invasive thymectomy is technically feasible with minimal morbidity and acceptable intermediate-term oncological outcomes in patients suffering with thymoma. Studies with a longer follow-up are warranted to evaluate the oncological outcomes considering its indolent course.

Author contribution

All the authors have contributed in the development, reviewing and finalising this manuscript.

Funding

None.

Data availability

Materials and data, if required, will be provided at the request of the editor.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical approval

The ethical approval was obtained by the institutional ethical committee before commencement of the study.

Consent to participate

Informed and written consents were obtained from all the participants.

Consent for publication

Consent to publish the details of the patients without disclosing the identity of the patient was taken.

Statement of human and animal rights

The corresponding author declares that the procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration.