Neurol India Home 

Year : 2020  |  Volume : 68  |  Issue : 3  |  Page : 648--651

Noninvasive Ventilation in Myasthenia Gravis

Usha K Misra, Surendra Kumar, Varun Kumar Singh, Deepanshu Dubey, Jayantee Kalita 
 Department of Neurology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, Uttar Pradesh, India

Correspondence Address:
Dr. Usha K Misra
Department of Neurology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Raebareli Road, Lucknow - 226 014, Uttar Pradesh


Background: Myasthenic crisis (MC) is mainly managed by invasive ventilation (IV) which is associated with prolonged intubation and ventilation and respiratory complication. Aim: To report the characteristics of patients who can be managed by noninvasive ventilation (NIV) and also predictors of NIV in myasthenia gravis (MG). Patients and Methods: In a retrospective study at tertiary care hospital in India during 2013–2015, patients with MG were evaluated. MG was diagnosed by clinical, repetitive nerve stimulation, and acetylcholine receptor antibodies. Patients were intubated based on arterial blood gas criteria. Demographic variables, Myasthenia Gravis Foundation of America (MGFA) score, and clinical examination were done. NIV success was defined if patient did not need mechanical ventilation (MV) in the management or for 72 h after extubation and was considered unsuccessful if patients required MV during this period. Results: Twenty three out of 68 MG patients had MC, 16 patients were given NIV, and 16 IV. Median intensive care unit stay was 18 (4–94) days. Predictors of crisis were infection, bulbar weakness, MGFA > 2b, and history of crisis. NIV was the only respiratory support in 7 patients and reintubation was prevented in 3 patients. NIV failed in 9 patients. Comparison of NIV success and failure did not reveal any difference. Conclusion: With NIV, intubation was prevented in 7 and reintubation in 3 patients. NIV has lower incidence of complications. NIV should be tried in patients with impending MC.

How to cite this article:
Misra UK, Kumar S, Singh VK, Dubey D, Kalita J. Noninvasive Ventilation in Myasthenia Gravis.Neurol India 2020;68:648-651

How to cite this URL:
Misra UK, Kumar S, Singh VK, Dubey D, Kalita J. Noninvasive Ventilation in Myasthenia Gravis. Neurol India [serial online] 2020 [cited 2020 Sep 25 ];68:648-651
Available from:

Full Text

Myasthenia gravis (MG) is an important neurological condition due to antibody and complement-mediated damage to postsynaptic acetyl choline receptors (AChRs) at the neuromuscular junction. MG has a prevalence of 20/100,000 population. Its incidence has been reported as 1.7–10.4/100,000. Earlier, MG was regarded as a grave disease, with a mortality of about 40%, but with modern treatment its mortality has been reduced to <5%.[1] Myasthenic crisis (MC) refers to weakness in MG severe enough to require intubation and invasive ventilation (IV).[2] MC accounts for 8–63% patient's presentation to the hospital with MG and corresponds to Class V of Myasthenia Gravis Foundation of America (MGFA) classification.[3],[4] There is no consensus on how to define MC.[5] MC should ideally be defined as exacerbation of myasthenic weakness causing respiratory failure, because not all patients will actually require mechanical ventilation (MV).[6] Osserman's classification was introduced before noninvasive ventilation (NIV) was available for the management of MG. MG is associated with muscle fatigue and weakness, reduced airway protection reflex, and reduced secretion clearance which result in poor alveolar ventilation, hypercarbia, acidosis, and hypoxia. NIV is an alternate strategy, in which oxygen is delivered at the peak of inspiration and pressure is reduced at the end of expiration. NIV in this way prevents alveolar collapse, improves oxygenation, and is devoid of complications of IV. There are several reports suggesting the usefulness of NIV in MG. These reports are based on retrospective chart reviews [3],[6],[7],[8] and isolated case reports.[9] These reports are mostly from the developed countries. In the developing countries, there is scarcity of hospitals with IV and the role of NIV has not been systematically evaluated in the developing countries. In this communication we report the characteristics of the patients who can be managed by NIV and compare them with patients who were managed by IV. We also report the predictors of patients with MC who can be managed by NIV.

 Patients and Methods

In this retrospective single-center hospital-based study, the patients with MG managed during 2013–2015 were included. The study was approved by the Institutional Ethics Committee. The diagnosis of MG was based on clinical, repetitive nerve stimulation study, and acetylcholine receptor (AChR) antibodies. The patients were intubated and kept on IV based on arterial blood gas (ABG) criteria pH <7.3, hypoxia <70 cm water, and pCO2>50 cm water.[10]

Exclusion criteria

Subjects were excluded if IV was conducted because of cardiac arrest or during surgery. The medical records were reviewed and the demographic details (age, gender), clinical (duration of myasthenic symptoms, MGFA score, severity of weakness, swallowing and breathing difficulty), history of MC, and precipitating factors (infection, stress, and metabolic alterations) were recorded. On admission ABG findings were noted. The duration of NIV and IV and associated complications were recorded.

The success of NIV was defined if the patient was free from intubation in intensive care unit (ICU) or did not need intubation for 72 h. Weaning and extubation was decided by the treating physician. The patient received corticosteroids, plasma exchange, and intravenous immunoglobulin (IVIG) as indicated. The patient also received symptomatic and supportive treatment such as antibiotics for infection, supplementation of calories, fluid, and electrolytes. The patients following intubation did not receive pyridostigmine for a week, and it was restarted before weaning in a dose of 30 mg tid and increased as indicated.[11]

T-piece trial/spontaneous breathing test: This trial included T-piece with low pressure support <6 cm water, and peak end expiratory pressure <5 cm water, the patient having calm appearance, had normal pulse and blood pressure, O2 saturation >90% on pulse oximetry, with fraction of O2 supplementation below 40%, and infrequent need of suctioning during spontaneous breathing trial, extubation was considered.

Extubation was considered successful if patient maintained spontaneous breathing for 72 h, otherwise it was considered as extubation failure. The physician decided the need of reintubation depending on the respiratory pattern, vital signs, and ABG parameters. The patient who required NIV for > 6 h per day after extubation, inability to withdraw NIV support even after 72 h following extubation was also considered as extubation failure.

Statistical analysis

The results were presented as mean ± standard deviation or median and range depending on the normal distribution or otherwise, respectively. The categorical data were analyzed by χ2 test. The means were compared using independent t- test or Mann–Whitney U test. Predictors of NIV and poor outcome were evaluated by logistic regression analysis assessed using clinical, laboratory, and ABG parameters. Two-tailed P value <0.05 was considered significant. The statistical analysis was done using Statistical Package for the Social Sciences version 16 and graph pad Prism V.


During the study period, 23 out of 68 MG patients (33.8%) had MC. The median age of the patients was 45 (15–69) years, 15 (65.2%) were males. The diagnosis of MG was based on repetitive nerve stimulation (22 patients), presence of AChR antibodies (19 patients), and muscle specific kinase (MuSK) antibodies in two patients. Ten patients had thymoma and 2 had thymic hyperplasia. Comorbidities were present in 8 patients and included diabetes mellitus in 4, hypertension in 5, coronary artery disease in 4, chronic obstructive pulmonary disease in 3, and thyroid disorder in 2 patients. Fifty nine patients out of 68 received steroids. Additional immunomodulator was given to 17 patients, 13 received azathioprine, and 4 were treated with mycophenolate mofetil. Sixteen patients were managed using MV and 16 using NIV. Six patients received IVIG, 3 plasma exchange, and thymectomy was done in 12 patients.

Twenty three out of 68 patients had MC and required ICU admission or respiratory support and 45 patients were managed in the general wards or did not require hospitalization. The patients who had MC had higher frequency of infection on admission (22 out of 23 vs 3 out of 45, P < 0.01), bulbar weakness (21 out of 23 vs 21 out of 45), MGFA stage >2b (19 out of 23 vs 21 out of 45), and history of MC (18 out of 23 vs 2 out of 45) [Table 1].{Table 1}

Patients with MC have median ICU stay of 18 (4–94) days and more than 7 days of ICU stay in 21 patients (91.3%). Seven (30.4%) patients of MC were managed by NIV only. In 3 other patients, NIV was used to avert reintubation. The patients who were managed with NIV only had a median age of 45 (15–68) years of whom only one was >60 years of age and 3 were females. The median duration of myasthenic symptoms before the crisis was 24 (1–60) months. The median duration of hospitalization before respiratory support was 4 (0–8) days and most common MGFA grade was IVb (5/7 patients). The admission ABG showed pH 7.37 ± 0.05, pO269.97 ± 14.66 mm Hg, pCO244.54 ± 9.49 mm Hg, and bicarbonate was 25.59 ± 3.44 meq/l. Serum potassium on admission was 3.49 ± 0.34 mmol/l and ionized calcium 1.08 ± 0.25 mmol/l. In 3 patients NIV was used after extubation to prevent reintubation. One out of the three developed crisis after extubation following thymoma surgery and the other 2 were extubated after IV and had reappearance of respiratory distress requiring NIV to avert reintubation. Nine patients who were initially managed on NIV later required IV. The cause of NIV failure was patient discomfort in 3, increasing drowsiness in 3, pneumonia in 1, and hypotension in 2 patients. The baseline characteristics of these patients (age, gender, serum potassium, admission ABG pO2, pCO2, bicarbonate, and MGFA stage) were not significantly different from the patients who could be managed on NIV alone compared to those who needed IV [Table 2].{Table 2}

Out of the 23 patients of MC, patients who were on immunotherapy required intubation more frequently (13 patients receiving immunotherapy vs 7 patients not receiving immunotherapy, P = 0.03). The duration of ICU stay in these patients was 22 (16–94) days, though it was more than the patients managed on NIV, i.e., 13 (8–24) days but did not attain statistical significance (P = 0.12). Three patients on IV died whereas none in NIV group died (P = 0.22) [Figure 1].{Figure 1}

In the 13 patients requiring IV, there was a higher frequency of complications compared to patients who require NIV. The complications in IV group included pneumonia (1 out of 7 vs 6 out of 13; P = 0.39), pneumothorax (0 out of 7 vs 1 out of 13; P = 0.45), tracheostomy (0 out of 7 vs 4 out of 13; P = 0.09), and hypotension (0 out of 7 vs 7 out of 13; P = 0.02). Three patients in MV group died, whereas none in NIV (0 out of 7 vs 3 out of 13; P = 0.27) [Table 3].{Table 3}


In the present study, 23 (33.8%) out of 68 patients of MG had MC. NIV was helpful in managing 10 and MV in 13 patients with MC. Infection, bulbar symptoms, MGFA score >2b, and previous history of crisis predicted MC. IV was associated with higher frequency of complications than NIV. This is the first study from a developing country highlighting the usefulness of NIV in management of MC. The success rate of NIV in MG has been reported to be 15–70%.[3] In our study 7 out of 23 (30.4%) patients with MC could be managed with NIV and in 3 patients reintubation was prevented. The failure of NIV can be attributed to longer duration of myasthenic symptoms and associated comorbidities such as diabetes mellitus, hypertension, chronic obstructive pulmonary disease (COPD), cardiovascular artery disease, though these did not achieve statistical significance.

Early intubation and IV is recommended in the management of MG to avoid fatigue, secondary metabolic complications, and infection. Evidence of MV in MC comes from earlier studies.[1],[12] NIV has been tried in MC to prevent intubation. In a study by Wu et al., NIV was used for the initial treatment in 24 episodes of MC and was considered safe if serum bicarbonate was below 30 mmol/l and APACHE II <6.[3] In our study, the patients who were successfully managed by NIV on admission pCO2 was 44.54 ± 9.49 mm Hg and bicarbonate was 25.59 ± 3.44 meq/l, i.e., <30 mmol/l, which was lower, i.e., 48.39 ± 13.53 mm Hg and 27.24 ± 3.17 meq/l, respectively, who failed NIV trial but it did not attain statistical significance. Lesser degree of bicarbonate rise refers to lesser metabolic compensation, which may suggest time since destabilization of respiratory parameter is less and there is scope for improvement with NIV.

On IV, 10 out of 16 patients could be successfully extubated after a median duration of 9.5 (5–48) days and they stayed in the ICU for 21 (14–55) days. We were unable to extubate 3 patients who died. Three patients required NIV to avoid reintubation. In an earlier study from our center, the duration of MV was 15.5 (16–60) days,[13] which is longer than other studies in which the duration of MV was <2 weeks.[14],[15] It is possible that we were not able to use IVIG and PLEX as indicated because of economic reasons. Prolonged use of corticosteroids and other immune suppressants predisposes to infection, especially in mechanically ventilated patients.

Eight patients had poor outcome and 4 needed tracheostomy. The predictors of poor outcome in the present study were reintubation, NIV failure, infection, MGFA > 2b, recurrent crisis and anti-MUSK positivity. High frequency of complications was found in patients on IV and ventilator-associated pneumonia was present in 6 patients, pneumothorax in 1, and shock in 7 patients. These complications were not present in patients on NIV although NIV patients complained of feeling of suffocation and dryness of mouth.

Moreover, NIV cannot be given to a patient with coma, severe bulbar weakness, hypotension, and pneumonia. We were not able to extubate 3 (23%) patients who were on IV. In MG extubation failure has been reported in 3–19% patients.[16],[17] The reason for extubation failure was prolonged intubation, associated COPD, diabetes mellitus, and immunosuppression in one each. In the study by Wu et al., reintubation was reported in 39% patients. In our study, 5 patients were reintubated. The causes of the patients who were reintubated were CO2 retention in 2, recurrent infection in 2, and poor cough in 1. A high rate of reintubation has been reported in progressive diseases such as amyotrophic lateral sclerosis.[18] However, in MG following reintubation, patient can be managed with good outcome when acute stress and precipitating events such as atelectasis, infection, and electrolyte changes can be managed. Physiotherapy and proper suctioning are essential, but improper and unnecessary suctioning and overzealous physiotherapy has been known to produce trauma and other complications. After extubation sometimes NIV is helpful in supporting the patient and preventing reintubations in 3 of our patients.

Four of our patients in MC were intubated directly without the NIV trial. The contraindications for NIV are hemodynamic instability, decreased consciousness, facial injuries, noncooperation, and apnea. Although bulbar weakness and pooling of secretions are considered relative contraindications for NIV, but we could use NIV in spite of these complications and manage MC using NIV because of nonavailability of MV. In the developing countries where facility of IV is sparse and expensive, the role of NIV should be further explored, especially because it can prevent reintubation. NIV seems to be cost-effective and has fewer complications than IV.

The present study is based on a small sample size, has a retrospective design and is from a tertiary care teaching hospital with a referral bias for advanced and severe cases. The results cannot be generalized to MC in general. Further studies are needed to explore the role of NIV in MG.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Cohen MS, Younger D. Aspects of the natural history of myasthenia gravis: Crisis and death. Ann N Y Acad Sci 1981;377:670-7.
2Kirmani JF, Yahia AM, Qureshi AI. Myasthenic crisis. Curr Treat Options Neurol 2004;6:3-15.
3Wu JY, Kuo PH, Fan PC, Wu HD, Shih FY, Yang PC. The role of non-invasive ventilation and factors predicting extubation outcome in myasthenic crisis. Neurocrit Care 2009;10:35-42.
4Jaretzki A, Barohn RJ, Ernstoff RM, Kaminski HJ, Keesey JC, Penn AS, et al.; for the Task Force of the Medical Scientific Advisory Board of the Myasthenia Gravis Foundation of America. Myasthenia gravis: Recommendations for clinical research standards. Neurology 2000;55:16-23.
5Piastra M, Conti G, Caresta E, Tempera A, Chiaretti A, Polidori G, et al. Noninvasive ventilation options in pediatric myasthenia gravis. Paediatr Anaesth 2005;15:699-702.
6Seneviratne J, Mandrekar J, Wijdicks EF, Rabinstein AA. Non-invasive ventilation in myasthenic crisis. Arch Neurol 2008;65:54-8.
7Rabinstein A, Wijdicks EF. BiPAP in acute respiratory failure due to myasthenic crisis may prevent intubation. Neurology 2002;59:1647-9.
8Rabinstein AA, Wijdicks EF. Weaning from the ventilator using BiPAP in myasthenia gravis. Muscle Nerve 2003;27:252-3.
9Epstein SK, Ciubotaru RL, Wong JB. Effect of failed extubation on the outcome of mechanical ventilation. Chest 1997;112:186-92.
10Mador MJ. Weaning from mechanical ventilation: What have we learned and what do we still need to know? Chest 1998;114:672-4.
11Ropper AH, Gress DR, Diringer MN, Green DM, Mayer SA, Bleck TP. Neurological and Neurosurgical Intensive Care. 4th ed.. Philadelphia: Lippincott Williams & Wilkins; 2004. pp 391.
12Chen R, Grand'Maison F, Strong MJ, Ramsay DA, Bolton CF. Motor neuron disease presenting as acute respiratory failure: A clinical and pathological study. J Neurol Neurosurg Psychiatry 1996;60:455-8.
13Gracey DR, Dievertie MB, Howard FM Jr. Mechanical ventilation for respiratory failure in myasthenia gravis. Two-year experience with 22 patients. Mayo Clin Proc 1983;58:597-602.
14Kalita J, Kohat AK, Misra UK. Predictors of outcome of myasthenic crisis. Neurol Sci 2014;35:1109-14.
15Bedlack RS, Sanders D. On the concept of myasthenic crisis. J Clin Neuromusc Dis 2002;4:440-2.
16Panda S, Goyal V, Behari M, Singh S, Srivastava T. Myasthenic crisis: A retrospective study. Neurol India 2004;52:453-6.
17Aggarwal AN, Gupta D, Behera D, Prabhakar S, Jindal SK. Intensive respiratory care in patients with myasthenic crisis. Neurol India 2002;50:348-51.
18Misra UK, Kalita J, Bhoi SK. Spectrum and outcome predictors of central nervous system infections in a neurological critical care unit in India: A retrospective review. Trans R Soc Trop Med Hyg 2014;108:141-6.