Breatlvng Disorders during Sleep in Myasthenia Gravis M. A. Quera-Salva, MD,* C. Guilleminault, MD,? S. Chevret, MD,$ G. Troche, MD," C. Fromageot, MD," C. Crowe McCann, MD," R. Stoos, MD,? J. de Lattre, MD," J. C. Raphael, MD,* and Ph. Gajdos, MD" ~~~~

~~~~~~~

~~

~

Twenty consecutive patients (16 women and 4 men), with a mean age of 40 years, who were diagnosed and treated for myasthenia gravis were enrolled in a prospective investigation aimed at determining the amount of respiratory disturbance occurring during sleep while they received treatment. Patients were clinically evaluated to determine body mass index, presence of upper airnay anatomical abnormalities, level of functional capacity and activity scored from 1 to 5 , and presence of sleep-related complaints. They underwent daytime pulmonary function tests, determination of mavimal static inspiratory pressure, measurement of transdiaphragmatic pressure, and measurement of arterial blood gas levels. Polygraphic monitoring during sleep, evaluating respiration and oxygen saturation, was also performed. RrsuIts indicated that in the studied population, all subjects had evidence of daytime diaphragmatic weakness as demonstrated by transdiaphragmatic pressure measurements, independent of the degree of autonomy and functional capacity and activity level reached. Older patients with moderately increased body mass index, abnormal total lung capacity, and abnormal daytime blood gas concentrations were the primary candidates for development of diaphragmatic sleep apneas and hypopneas, and oxygen desaturation of less than 9096 during sleep. However, these clear indicators were not found in all subjects with sleep-related disordered breathing. Rapid-eye-movement sleep was the time of highest breathing vulnerability during sleep. Sleep-related complaints may also help identify subjects at risk for abnormal breathing during sleep, even when daytime functional activity is judged normal. Quera-Salva MA, Guilleminault C, Chevret S, Troche G , Fromagcot C, Crowe McCarin C , Stoos R, de Lattre J, Raphael JC, Gajdos Ph. Breathing disorders during :ilecp in myasthenia gravis. Ann Neurol 1932;3 1:80-32

Myasthenia gravis (MG) is characterized by recurrent episodes of weakness due to the fatigability of voluntary muscles. In this condition, IgG autoantibodies bind to acetylcholine receptors and interfere with neuromuscular transmission { 1). Diaphragm and accessory respiratory muscles may be involved, and respiratory failure has been observed during exacerbations of the syndrome [2, 33. In the recent past, an increased interest in sleep-related breathing led to the investigation of patients with neuromuscular disorders during nonrapid-eye-movement (NREM) sleep and rapid-eyemovement (REM) sleep. We present the results of a bicenter study of 20 patients with MG who were investigated during wake and sleep. The aims of the investigation were to assess the presence of breathing disorders during sleep and to determine whether a correlation existed between daytime pulmonary function and any sleep-related breathing disorder.

Materials and Methods Criteria f i r Inclzcszon

From the 'HBpital Raymond PoincarC, Centre Hospitalo-Universiraire Paris-Ouest, Garches, France; ?Stanford University Medical School, Stanford, CA; and the $Department de Biostatktique et Informatique, HBpitaJ St-Louis, Paris, France.

Received Apr 26, 1990, and in revised form Feb 14 and Jun 27, 1991. Accepted for publication Tun 29. 1991.

The study was designed as a prospective investigation, during wakefulness and sleep, of patients with MG. Informed consent was obtained from each patient. Each consecutive patient who was seen over a 12-month period and fulfilled the following criteria was included in the study: 1. Each patient had to have generalized MG. The diagnosis of M G had to have been based on findings from clinical history, clinical examination, electromyography, and pharmacological testing. (Circulating antiacetylcholine receptor antibodies might have been measured, but this t8est was not a prerequisite for inclusion.) 2. All patients had to have a stable clinical status and wellcontrolled MG during wakefulness, with no change in treatment for at least 8 months. Patients would be studied while they received their currently prescribed medications.

Address correspondence to D r Quera-Salva, Service de Rkanimarion Medicale, HGpital Raymond Poincare, 104, bd Raymond PoincarG. 92380 Garches, France.

86 Copyright 0 1992 by the American Neurological Association

Table 1 . Clinical Characteristics and Pulmonary Function Test Results of the Patient Population

Patient No.

Age (yr)

22 31

1 2 3 4 5

56 49 68 28 30 24 49 29 53 39 52 59 53 36 27 40 37 44

6 7 8 9 10 11 12 13 14 I5 16 17 18 19 20

Sex

BMI (kg/cm2)

F F F F M F F F F F F F M M F F F M F F

19.5 15.0 32.0 30.0 29.5 16.5 20.0 22.0 20.0 24.0 25.0 17.5 28.0 23.5 19.5 16.0 18.5 22.0 23.0 23.5

Duration since Diagnosis of Disease (yr)

2 2 2 28 4 4

9 10 11

8 15 7 4 4 23

FC

TLC(%)

FRC(%c)

2 2 3 3 3 4 3 2 2

110 76 75 72 49 84 94 75 109 83 92 73 68 92 95

113 100 53

1

4

2 3 1 3 3 1

2

1

9

2 2 1

8

12

100

62 110 103 47 132 92

10 1

109 80 72 75

111

92 60 112 114 100 102 78 74 68

Pao, (mmHg)

Paco, (mmHg)

Pdi Muller" (cmH20)

95 90 81 75 70 90 92 96 84 85 79 69 80 78 87 89 88 84 81 78

38 37 38 47 45 40 40 39 41 37 39 56 44 33 44 38 38 39 40 40

20 -

23 10 -

32 38 23

6.5 30 8

30 18 37 40 49 21 25 20

"See texr for description.

FC

=

functional capacity and activity (I-normal, 5-completely incapacitated); TLC = total lung capacity, expressed in % of predicted; FRC 9T of predicted; Paoz = arterial oxygen pressure; Pam, = arterial carbon dioxide pressure.

= functional residual capaciry in

3. The final selection criterion used was different for the two groups, as the clinics from which the patients were recruited had different orientations. The French patients came from an intensive care unit follow-up clinic and all of them had a history of at least one episode of acute respiratory distress that required intubation. They were not enrolled as a result of sleep complaints. The Stanford patients came from a neurology clinic and were recruited on the basis of having complaints of breathing problems during sleep.

Patient Populution During the 12-month period considered (from June 1988 to June 1989), 20 patients fulfilled the inclusion criteria and entered the study. There were I 5 French and 5 American patients, including 16 women and 4 men. The French patients were recruited from a pool of 104 patients with M G seen in the clinic during the same period. Eight American patients with M G were seen during the 12 months and 5 fulfilled the selection criteria for the study. The mean age was 4 1 & 13 years (range, 22 to 6 8 years). The mean duration of MG was 7 -t 5.5 years. Body mass index (BMI) was calculated using Khosla and Lowe's formula { 4 ] , weight in kilograms times 10,00O/(height in centimeters)'. The mean was 22 i- 5 kg/m2 and the range, 15.0 to 32.0 kg/m2; 28 to 33 kg/m2 was considered to be moderate obesity. Four patients had a BMI in this range (Table 1).

All patients received daytime treatment for MG: All received anticholinesterase medications (pyridostigmine bromide or ambenonium chloride), 9 also received corticosteroids, and 13 patients had had a thymectomy. In each case, the last anticholinesterase medication was taken at bedtime.

Procedure After a review of the clinical history (including sleep-related symptoms) and a clinical evaluation, including examination of the upper airways, and grading of functional capacity and activity level on a 5-point scale (from 1, complete remission, to 5, major deficit without autonomy), all patients underwent pulmonary function tests (PITS)during the daytime. PFTs were performed while the patient was in a sitting position and included spirometry, measurement of functional residual volume (FRC) by the helium dilution technique, arterial blood gas levels determined from blood samples drawn within 2 hours of awakening, measurement of transdiaphragmatic pressure (Pdi), and maximal static inspiratory pressure (Pi max, performed according to Black and Hyatt's technique [ 5 , 6 } ) . Pi max was considered to be reduced when it was less than 75% of that predicted 15, 61. The Pdi was measured using Lapporta and Grassino's technique [7}, with a differential transducer connected to gastric and esophageal balloons (Pdi: gastric pressure-esophageal pressure). Pdi values at rest and expiratory pressure were taken at the zero level. In addition to spontaneous breathing,

Quera-Salva et al: Breathing Disorders during Sleep in MG 87

patients were asked to perform a maximal inspiratory effort against a closed glorris (Pdi during the Muller maneuver) [7]. These maneuvers were repeated several times until a best-obtained result had been reproduced. Patients were allowed to rest between each trial. The so-called “diaphragmatic fatigue” level (Pdi critical = Pdi/Pdi Muller > 40%) was then calculated using Roussos and Macklem’s formula IS]. Pdi and Pi max were measured with biofeedback, the order being at random.

Nocturnal Pobgraphic Recording Electroencephalograms (EEG) (C3/A, and C4/A, derivations), electro-oculograms, chin electromyograms, and electrocardiograms (modified V, lead) were systeniatically recorded. Airflow was measured by buccal and nasal thermistors. Respiration was measured by uncalibrated, inductive respiratory plethyiimography (Respitrace, Anrlsley, NY), aith, in addition, esophageal pressure recording and/or measurements of the accessory respirarory muscles and diaphragm with a nonintegrated electromyogram, using subcutaneous needle electrodes. Oxygen saturation (SaC),) was measured by oxinietry at the finger (Biox-Ohmeda TM).

Data Analysi, Sleep and sleep stages were scored following the international criteria of IRechtschaffen and Kales [9]. Apnea was defined as a cessation of airflow at the nose and mouth for at least 10 seconds. Hypopnea was defined as a reduction in airflow by at least 50% associated with a decrease in SaO, of at least 4% below the preceding baseline. Apneas and hypopneas were characterized as central or diaphragmatic by: (1) a decrease or absence of thoracoabdominal efforts, and (2) a decrease or absence of the negative inspiratory esophageal pressure deflection and/or a decrease or absence of accessory respiratory muscle activity. Several indices were calculated: (a) the classic respiratory disturbance index (RDI), that is: No. of apneas + No. of hypopneas Total sleep time

X

60;

(b) the REM slcep RDI (RDI REM), that is, an index of respiratory disturbance during total REM sleep time, (c) the percentage of tclral sleep time (TST) with an SaO, of less than YO‘/, that is, the percentage of time spent during sleep with an SaOz of less than 90%, and (d) the SaO, 90 index and SaOl 80 index, that is, the number of decreases in SaO, below 90(/1 and below 80% during total sleep time [lo]

Statis&-al Analysis Statistical analysis was performed using the SAS statistical software (SAS Institute, Carey, NC). Comparisons between patients with or without severe oxygen desaturacion during sleep, called group A and group B, respectively, were based o n the nonparametric Wilcoxon test I1 11. Relationship between breathing indices during sleep, clinical data, and daytime PET results was tested using the nonparametric Spearman rank test. However, considering the small sample size, we computed the Tukey’s Jacknife estimates of the correlation coefficient, in order to provide a better estimate [ 12, 131. Results are expressed as mean 5 standard deviation (SD).

88 Annals of Neurology

Vol 31 No 1 ,January 1992

Results Clinical Status Patients’ clinical status is reviewed in Table 1. Eight patients had a score of 3 (out of a possible 5) or more on the functional scale (i.e., partial activity due to MG), 7 patients had a score of 2 (i.e., minimal deficit with normal activity), and all others were considered to be in remission with no deficit during the daytime with treatment intake (score 1).

Sleep Complaints Twelve patients (Patients 3, 4, 5, 10, 12, 13, and 15 through 20) reported disrupted nocturnal sleep. Most commonly, patients reported waking up in the middle of the night and during the early-morning hours with a sensation of breathlessness. The sleep complaints may have existed for as long as 10 years in some patients. However, nocturnal sleep disturbances were always noted after the onset of MG. Four (Patients 3, 4, 5, and 12) of the 12 patients also reported daytime somnolence, and 4 (Patients 3 , 4, 12, and 15) had morning headaches, symptoms often seen in connection with repetitive deep apneas. Only 1 male patient (Patient 13) presented a history of snoring over the years, and he had a long soft palate with redundant soft tissues, suggesting an anatomical abnormality of the upper airway. Pulmonary Function Testing (see Table I ) Patients’ mean vital capacity was 80 -+ 17% of predicted (range, 5 1-1 12%), the mean total lung capacity (TLC) was 84 5 16% of predicted (range, 49-1 lo%), mean residual volume was 104 15% of predicted. (range, 59-110%), and FRC was 91 5 23.5% (range, 47- 132%). Nine patients had a restrictive respiratory syndrome (TLC < 80% of predicted). Daytime arterial blood gas levels were within the normal range in 17 subjects; however, 3 patients had hypoxia and hypercapnia. The mean Pi max was 7 1.5%; of predicted (range, 27 to 100%). Pi max was considered to be within the range of predicted values (> 75% of predicted) in 11 patients (Patients 1, 3 , 6, 8, 9, 11, and 15 through 19). Pdi was not measured i n 3 patients who refused to undergo the test. Pdi wais reduced in all 17 patients in whom it was measured. Normal values of the Muller maneuver Pdi range from 91 to 141 cm H 2 0 [7]. In our patients, the mean Muller Pdi was 26.5 cm H,O (range, 6.5 to 49 CIS-I H20), indicating diaphragmatic weakness at the moment of examination in all subjects studied.

*

Polygraphic Monitoring during Sleep (Table 2) Mean TST was 382 -+ 70.5 minutes (range, 212-525 min), mean percentage of stage 1 NREM sleep was 17.5% (range, 8-38%), mean percentage of stages 3 and 4 NREM (slow-wave) sleep was 12% (range,

Table 2. Polygraphic Monitoring Results Patient

TST

No.

(min)

% REM

5 7 s3-4

RDI

REM RDI

52 TST SaO, < 90%

433 431 418 363 446 525 453 2 12 369 298 340 450 355 272 349 40 1 396 406 378 361

17.7 22.0 14.0 15.0 14.0 17.0 20.8 32.1 10.8 26.3 17.0 9.9 10.9 15.0 13.0 17.0 18.0 15.0 16.5 14.0

14.5 23.0 4.3 15.0 17.0 15.0 10.4 32.8 12.0 7.3 17.3 11.0 11.7 26.7 14.5 4.0 6.0 0.0

0.0 0.0 5.0 13.0 11.5 0.0 0.5 0.0 3.4 2.0 0.7 18.0 6.5 1.3 7.0 9.0 8.0 9.0 11.0 10.0

0.0 0.0 35.0 44.2 70.0 0.0 0.0 0.0 10.0 0.0 3.9 85.0 24.5 0.0 53.0 30.0 30.0 42.0 46.0 36.0

0.0 0.0 19.5 50.0 58.7 0.0 0.0 0.0 0.0 0.0 0.0 13.3 1.4 1.5 3.0 1.7 1.5 1.9 2.6 2.2

1 2 3 4 5 6 7 8

9 10 11 12 13 14 15 16 17 18 19 20

0.9 0.0

~~

~~~~

TST = total sleep rime, % FGM = percentage of total sleep time spent in rapid-eye-movement (REM) sleep, 9 S-3-4 = percentage of TST spent in stages 3 or 4 non-REM, RDI = respiratory disturbance index, REM RDI = RDI during total REM sleep, 9 TST Sa02 < 90% = percentage of total sleep time spent with saturated oxygen level below 90%

0-332), and mean percentage of total REM sleep was 17(2 (range 9.9-32.I%). These results indicate a moderately disturbed nocturnal sleep [ 141, with an overall increase in stage 1 NREM sleep, and a decrease in stages 3 and 4 NREM and REM sleep percentages. This sleep disturbance must, at least in part, be related to the examination conditions and to first-night effect. Investigation of breathing during sleep showed that 11 patients had an RDI of 5 or higher; 4 of them had an RDI greater than 10. A raised REM sleep RDI (REM RDI > lo), with a mean REM sleep RDI of 45 (range, 30 to 70), was also seen in these 11 patients, indicating that they had a significant number of apneas and hypopneas during REM sleep. All 11 patients, with the exception of Patient 13, who presented a predominance of mixed and obstructive apneas and hypopneas, had diaphragmatic (central) apneas and hypopneas. SaO, measures showed that these respiratory disturbances were associated with decreases in oxygen saturation. When the SaO, was examined during sleep, it was found that 12 patients passed greater than 1% of their TST with an SaO, of less than 90%. This group of 12 patients comprised the 1I patients who had an RDI of 5 or higher and 1 patient who did not have a pathological RDI. This group of 12 patients with decreases in SaO, during sleep can be divided into two subgroups:

Subgroup A comprised 8 patients who spent from 1 to 3% of their TST with an SaO, of less than 909%.Of these 8 patients, 6 patients (Patients 15 through 20) presented central-type apneas and hypopneas exclusively related to REM sleep and SaO, dropped mildly in association with these respiratory disturbances. Patient 14 presented non-EM-related central-type apneas and hypopneas, and spent 1.5% of TST with an SaO, of less than 90%. Patient 13, who had signs and symptoms of “obstructive sleep apnea syndrome” (OSAS), presented obstructive sleep apneas at the time of polygraphic recording, which occurred predominantly during stages 1 and 2 NREM sleep. His breathing abnormalities during sleep were thought to be, at least in part, related to the upper airway abnormality and not entirely due to MG. The 6 patients with REM sleep-related respiratory disturbances and mild decreases in SaO, in association with diaphragmatic apneas had nighttime arousals associated with the predominantly REM sleep respiratory events and complaints of nocturnal disrupted sleep. SunGRouP A .

n. Subgroup B included 4 patients (Patients 3, 4, 5, and 12) who presented a more severe condition, spending an average 35.5% of TST (range, 13.3-58.7s) with an SaO, of less than 90%. T h e nocturnal polygraphic investigation indicated that 3 (Patients 4, 5, and 12) of these 4 patients had a clear

SUBGROUP

Quera-Salva et al: Breathing Disorders during Sleep in MG

89

Table 3 . Correlation between Total Lung Capacity and Respiratory Disturbance Index: Potential Influence of Indit miduar' Obsewations

2o

1

7

2

16

Patient No. 1 2 2

4 5 6 7

I

8 9 10 11 12

13 14 15 16 17

18

19 20

Partid Spearman Correlation Coefficients" - 0.404

-0.513 -0.455 - 0.398 -0.388 - 0.484 - 0.481 -0.530 - 0.475 -0 . 4 3 - 0.482 - 0.401 - 0.494 - 0.482 -0.526 -0.529 -0.527 -0.452 - 0.398 - 0.408

"Caku1att.d for cdch patient o n the remaining 19 patients.

central-type sleep apnea syndrome f l r j ) . They presented a combination of clinical complaints (nocturnal disrupted sleep, daytime somnolence, and morning headaches) with polygraphic abnormalities including a significant RDI (RDI = 13.0, 11.5, and 18.0, respectively) and a decrease in SaO, (percentage of TST with an SaO, of less than 90% of 50.0, 58.7, and 13.3%, respectively). Patient 3 had a predominantly REM sleep-related central type of respiratory events (RDI = 5 ; REM RDI = 35) with a combination of hypoventilation and apneas and hypopneas, with a lower SaO, SaO, < 70% := 19.5). baseline during sleep (2,

Cowelution of Test Results Duration of disease and level of functional capacity showed no correlation with disturbed breathing during sleep or decrease of SaO, during sleep, hut age correlated with the REM sleep RDI (Spearman coefficient, r = 0.48, p = 0.03) and with percentage of TST with an SaO, less than 90% ( Y = 0.57, p := 0.009). In addition, BMI was correlated to age ( Y = 0.60, p = 0.005). The older the subjects, the greater the risk of presenting an elevated REM RDI and an increased amount of time spent with an SaO, less than 90%. Groups A and B were thereafter compared. Slight imbalances, which were not significant statistically, were observed between the two groups. They con90 Annals of Neuroiogy Voi 3 1 N-o 1 January 1992

Correlation between total lung capacity ITLCI and respiivatory disturbance index (RDIi, calculatedfor each subset oJ 19 patients. Each point corresponds t o the &set of 19 patient.! and the number indicates the exrluded pdient. ?'he partial Spearman correlation codjcient ii indicated in Table .?.

sisted of older patients in Group B, where the mean age was 54 versus 37 years in Group A ( p = 0.07), and higher BMI in Group B, where the mean BMI was 20 versus 27 in Group A ( p = 0.08). Restrictive lung disease, as determined by TLC,correlated negatively with RDI ( Y = -0.466, p = 0.04), with REM RDI ( Y = --0.448, p = 0.05), anal with indices of SaO, decreases, the highest correlation being against the percentage of TST spent with an SaO, of less than 90% ( r = -0.524, p = 0.02). In order to assess the influence of each subject, we reperformed the latter analyses and examined the effect on the body of data that resulted by omitting each patient. Thus, Spearman correlation coefficients were calculated for every subset of 19 patients (Table 3 ) , and then an overall correlation coefficient was estimated by computing the mean of the 20 partial coefficients. It appeared that these Jacknife estimates did not differ markedly from those previously shown. T h e Figure displays the plot of TLC against RDI, indicating for each patient the value of the correlation coefficient that resulted from omitting that subject. The Jacknife correlation coefficient between TLC and RDI was unchanged I(Y = -0.465, p = 0.04). T h e correlation of TLC against percentage of TST spent with an SaO, of less than 90%; yielded similar results (corrected r = - 0.5.23, p = 0.02). Finally, the negative correlation between TLC and REM RDI was still observed, but statistical significance disappeared (corrected r =; - 0.40 1, p = 0.08). Daytime arterial blood gas concentrations were altered in 3 Group B patients. They had dayrime hypoventilation with combined daytime hypoxia and hypercapnia. Measurements of daytime diaphragmatic efforts

(Muller Pdi, Pdi critical) and Pi max measurements showed no correlation with RDI, nor with indices of desaturation during sleep. All 17 patients tested had a lower-than-expected Muller Pdi, and this measurement could not differentiate patients with and those without disordered breathing during sleep.

Discussion Sleep studies are rarely performed on patients with MG, and patients who appear to be stable during the daytime are considered to have well-controlled disease. However, patients with M G may have sleep-related complaints such as sleep disruption [ 161, nocturnal awakenings, early-morning awakenings, and daytime somnolence, which may be related to their poorly controlled muscle disorder during sleep, with heightened problems during REM sleep. All of our patients with REM-related respiratory events complained of disturbed nocturnal sleep, with a sensation of breathlessness. All of the Group B patients also complained of daytime somnolence, and 3 of the 4 had earlymorning headaches. A patient may have MG and an abnormality of the upper airway, which may lead to development of OSAS, as occurred with Patient 13. But our study emphasizes that this is not the rule. All other patients with sleep-related disturbed breathing had diaphragmatic (central-type) apneas and hypopneas during sleep, as already indicated by Shiozawa and colleagues [ 17) in 6 of 10 stable, daytime-well-controlled patients with MG. The fact that the respiratory events were mostly diaphragmatic may be related to the chronic impairment of diaphragmatic and accessory respiratory muscles. Mier-Jedrze jowicz and associates [ 3 ] reported that a severe impairment of respiratory muscles may exist in patients with MG, even when the involvement of other skeletal muscles is mild. Our patients had a significant reduction of Pdi and Muller Pdi. At the same time, the values for the Pi max were often at the lower limit of normal. These discordant results may be explained by a poorly functioning diaphragm in the presence of normally functioning intercostal and accessory muscles. However, the Pdi maneuvers are difficult to perform, which may also have contributed to the results. The diaphragmatic dysfunction may be the reason for the nocturnal respiratory events, particularly during REM sleep when the physiological REM sleep-related muscle atonia is present, with diaphragmatic movements contributing the bulk of the air exchange r18). Despite the fact that all of our patients had a low Muller Pdi, not all of them had a large number of diaphragmatic apneas and hypopneas during sleep, o r spent a large amount of sleep time with an SaO, of less than 90%. The long periods of time when SaO, was less than 90% in the 3 Group

B patients with existing daytime hypoventilation can be explained by a combination of increased hypoventilation during sleep and sleep apneas. The fourth patient (Patient 3) had predominantly REM-related apneas but had a tendency to hypoventilate throughout sleep. PFTs, performed during the daytime, were informative and indicated those subjects who were more at risk of having disordered breathing during sleep. O n the basis of our study, the existence of a restrictive lung disease and the presence of abnormal daytime blood gas levels in the patients should raise the concern of a worsening of breathing during sleep. We pointed out that restrictive lung disease was negatively correlated with disturbed breathing during sleep and with decreases in SaO,. One may note that since the data are few, these findings should be interpreted cautiously. We argue that even though several observations are influential, particularly those in Patients 5 and 12 (see Fig), results were not markedly affected, and persistent negative correlations were observed. Also, it is easy to understand why patients with M G who have a higher BMI (i.e., higher loaded breathing and, more important, chest bellow disease) will be at a higher risk of disordered breathing during sleep. However, one must emphasize that BMI was never massively elevated in our patients. But a mild elevation of BMI may not be well tolerated during sleep in older patients with MG. Age is a variable that is not considered often enough as a risk factor for abnormal breathing during sleep. The majority of patients with M G who had a higher RDI and RDI REM were older. Despite identification of risk factors that appear to correlate with breathing disorders during sleep in our patients with MG, such as abnormal PFI' results, abnormal daytime blood gas levels, elevated BMI, and older age (factors that have also been found to be significant for predicting abnormal breathing during sleep in other respiratory disorders such as chronic obstructive pulmonary disease or myotonic dystrophy), we cannot explain in our population the dichotomy in normal and abnormal breathing during sleep solely o n these daytime indices. For example, Patient 12, 37 years old, with a TLC of 7 3 % ~of predicted value and a BMI of 17.5 (i.e., very slim), had an RDI of 18, a REM sleep RDI of 85, and a percentage of time spent during sleep with an SaO, of less than 90% of 13.3. She also had the highest amount of REM sleep-related apnea. In conclusion, patients with MG, even if appropriately treated during the daytime and possessing good functional capacity and activity level, may have abnormal breathing during sleep. Sleep-related complaints (disrupted nocturnal sleep, nocturnal awakening, and so on) may allow us to suspect the problem. Older patients with a moderate increase in BMI and abnormal PFT results and daytime blood gas concentrations are the primary candidates for sleep-disordered breathing

Ouera-Salva et al: Breathing Disorders during Sleep in MG

91

and oxygen desaturation during sleep. O n e issue that was not addressed in this report, but which will be in the second phase of this investigation, is the limited duration of action of some of the treatments prescribed for MG. Several prescribed drugs have too short a half-life to maintain appropriate coverage throughout the TST. This study was supported by I'Association Frangaise contre les Myopathies (AFM) an,J by the grant NS-07772 from the National Insritute of Aging. We are indebted to Daniele Morlat, Jean-Marc Sadriri, ant1 Gilles Macadoux for their technical support, and Corine Redelsperger for typing the manuscript.

References I

-3

Lindstrom JM, Seybold ME, Lennon VA, et al. Antibody to ACh-R in myasthenia gravis. Neurology 1976;26:1054--1059 Newsom-Davis J, Goldman M, Lohn L, Casson M . Diaphragm function and alvcdar hypoventilation. Q J Mcd 1976;45:

x 7- 1oo 3. Mier-Jedrzejowicz A, Brophy C, Green M. Respiratory muscle function in myasthenia gravis. Am Rev Respir Dis 1988; 139867-873 4. Khosla T, Lowe FR. Indices of obesity derived from body weight and height. Br Soc Med 1967;21:122-128 5. Black LF, Hyatt RE. Maximal respiratory pressures. normal values and relationship to age and sex. Am Rev Respir Dis 1969;99:696-702 6. Black LF, Hyatt RE. Maximal static respiratory pressures in

92 Annals of Neurology

Vol 31

No 1 January 1992

generalized neuromuscular disease. Am Rev Respir Dis 197 I ; 103:641-650 7 Lapporta D, Grassino A. Assessment of transdiaphragmatic pressure in humans. J Appl Physiol 1985;58: 1469-1475 8. Roussos CS, Macklem PT.Diaphragmatic fatigue in man. J Appl Physiol Respir Environ Exerc Physiol 1977;43: 189-197 9. Rechtschaffen A, Kales A, eds. A mmual of standardized terminology. Techniques and scoring system for sleep stages of human subjects. Brain Information Service (NINCDS Neurological Information Network, UCLA, Los Angeles, 1968), B r a n Research 10. Partinen M, Guilleminault C, Quera-Salva MA, Jamieson A. Obstructive sleep apnea and cephalometric roentgenograms: the role of anatomic upper airway abnormalities in the definition of abnormal breathing during sleep. Chest 1988;93:119'+1205 11. Wilcoxon F. Individual comparisons by ranking methods. Biometrics 1945 ;1:80-8 3 12. Efron 8. Non parametric estimates of standard error: the Jacknife, the boot strap and other methods. Biometrika 1981;68: stlo-599 13 Mosteller F, Tukey JW. Data analysis and regression. A second course in statistics. Reading, MA: Addison Wesley, 1977 14. Williams RL,Karacan I, Hursch CJ. Electroencephalography of human sleep: clinical applications. New York: Wiley, 1974 15. Guilleminault C, Dement WC, eds. Sleep apnea syndromes. New York: Allan R. Liss, 1986 16. Mennumi G, Morante MT, Scoppeta C, Bergonzi P. Night sleep organization in myasthenic patients not undergoing therapy. Sleep Res 1983;12:84 (Abstract) 17 Shiozawa 2, Shintani S, Tsunoda S, Shindo K. Sleep apnca in well-controlled myasthenia gravis. Sleep Res 1987;16:301 (Ahstract) 18 Tisiewicz K, Moldofsky H , Bryan A, Bryan M. Mechanics of rib cage and diaphragm during sleep. J Appl Physiol Rcspir Environ Exerc Physiol 1977;43:600-602k

Breathing disorders during sleep in myasthenia gravis.

Twenty consecutive patients (16 women and 4 men), with a mean age of 40 years, who were diagnosed and treated for myasthenia gravis were enrolled in a...
659KB Sizes 0 Downloads 0 Views