British Journal of Anaesthesia 87 (6): 885±9 (2001)

Difference in sensitivity to vecuronium between patients with

ocular and generalized myasthenia gravis
H. Itoh1, K. Shibata2 and S. Nitta3
1
Department of Anesthesiology and Intensive Care Medicine, Kanazawa University School of Medicine,
Kanazawa, Japan. 2Department of Emergency and Critical Care Medicine, Kanazawa University School of
Medicine, Kanazawa, Japan. 3Division of Anesthesia, Ishikawa Prefectural Central Hospital, Kanazawa,
Japan
*Corresponding author: Department of Anesthesiology and Intensive Care Medicine, Kanazawa University School of
Medicine, 13-1 Takara-machi, Kanazawa 920-8641, Japan

Patients with myasthenia gravis show sensitivity to non-depolarizing neuromuscular blocking

drugs, but little is known about differences in this sensitivity between types of myasthenia. In
10 patients with ocular myasthenia gravis and 10 with generalized myasthenia gravis, twitch
tension was monitored in the adductor pollicis muscle by supramaximal train-of-four stimula-
tion of the ulnar nerve during anaesthesia with sevo¯urane 2.5% and nitrous oxide 60%. After
baseline measurement, an initial dose of vecuronium 10 mg kg±1 was given. When the twitch
height stabilized (maximum block after the ®rst 10 mg kg±1), the next incremental dose of
10 mg kg±1 was given and repeated until block, de®ned as [1±(®rst twitch/baseline ®rst
twitch)]3100 reached 90%. Maximum block after the ®rst dose of vecuronium in ocular
patients was signi®cantly less than that in generalized patients (median 51 vs 91%; P<0.05).
Onset of block after the ®rst dose of vecuronium was signi®cantly slower in ocular than in
generalized myasthenic patients (mean 300 vs 200 s; P<0.05). Doses required to attain a block
of 90% or more were signi®cantly higher in ocular than in generalized patients (median 20 vs 10
mg kg±1; P<0.05). Clinicians should consider the type of disease according to the Osserman
classi®cation when using non-depolarizing neuromuscular blocking drugs in patients with
myasthenia gravis.
Br J Anaesth 2001; 87: 885±9
Keywords: equipment, force transducer; complications, myasthenia gravis; neuromuscular
block, Osserman classi®cation
Accepted for publication: July 23, 2001

Myasthenia gravis is an autoimmune disorder characterized
by muscular fatigue that worsens with exertion. Patients
with myasthenia gravis require special care during anaes-
thesia because of their sensitivity to non-depolarizing
neuromuscular blocking drugs,1±5 resulting from a reduced
number of functioning acetylcholine receptors (AChR) at
the neuromuscular junction.6 A wide range of effective
doses of vecuronium have been shown to be required to
obtain 95% neuromuscular block (ED95) in myasthenic
patients, varying from 6 to 44 mg kg±1.2 3 The wide variation
in the severity of myasthenia, which is categorized using the
Osserman classi®cation,7 may be one explanation for the
varied sensitivity to vecuronium, but little has been reported
about this possibility. Hypothesizing that ocular myasthenic
patients would be less sensitive to vecuronium than
generalized myasthenic patients, we compared neuro-
muscular responses to vecuronium in the adductor pollicis
muscle between these groups.
Methods
After we had received approval of the study protocol from
the Institutional Ethics Committee of our hospital and
obtained individual patients' informed consent, we studied
20 myasthenic patients (10 with ocular disease and 10 with
generalized disease, representing types I and II in the
Osserman classi®cation7 respectively) who were scheduled
to undergo thymectomy. The diagnosis was con®rmed by
the presence of circulating antibody to the AChR and a
positive response to edrophonium in addition to typical

Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2001
 Itoh et al.
clinical and laboratory ®ndings (ptosis, diplopia, limb
weakness, and a decremental conduction response on
electrical stimulation of the nerve supply to the orbicularis
oculi or the deltoid muscle). The generalized patients
showed limb weakness or a decremental response in the
deltoid muscle to repetitive motor nerve stimulation.
Binding and blocking antibodies to AChR were measured
preoperatively in serum samples. To determine binding
antibody, a-bungarotoxin labelled with 125I was bound to
prepared antigenic human AChR (125I-AChR). Sera from
patients were incubated with 125I-AChR. Antigen±antibody
complexes were precipitated by incubation with anti-human
IgG. Radioactivity in the precipitate was measured with a
gamma counter.8 When measuring binding antibody, it is
impossible to detect the antibody in the area near the
acetylcholine binding site of the receptor. Thus, blocking
antibody to the acetylcholine binding site was also
measured. Sera from patients and controls were incubated
with prepared antigenic AChR followed by incubation with
125
I-labelled a-bungarotoxin. Aliquots were then applied to
Sepharose columns to measure the percentage inhibition of
labelled toxin binding with AChR.9
Anticholinesterase medication was continued until the
morning of surgery. The patients were premedicated with
hydroxyzine 50 mg and atropine 0.5 mg, given 1 h before
induction of anaesthesia. Anaesthesia was induced with
thiopental 4 mg kg±1 and fentanyl 2 mg kg±1 followed by
inhalation of sevo¯urane 2.5% and nitrous oxide 60% in
oxygen. The lungs were ventilated by face-mask so that the
end-tidal carbon dioxide pressure was maintained at about
5.3 kPa. Skin temperature over the thenar region was
monitored and was maintained at 32±33°C. Non-invasive
blood pressure was measured every 5 min, and when
systolic blood pressure had decreased to <80 mm Hg
ephedrine 5 mg was administered i.v.
Neuromuscular transmission was monitored by measur-
ing twitch tension in the adductor pollicis muscle
(Myograph 2000; Biometer International, Odense,
Denmark). The ulnar nerve was stimulated at the wrist
using surface electrodes with supramaximal train-of-four
(TOF) square-wave pulses of duration 0.2 ms every 12 s
(Myotest, Biometer International). When the twitch
response had remained stable for at least 10 min, the
baseline height of the ®rst twitch (T1) was measured. The
TOF ratio, expressed as a percentage, was de®ned as the
ratio of the fourth to the ®rst response to TOF stimulation.
After baseline measurements, an initial dose of vecuronium
10 mg kg±1 was administered i.v. Maximum block was
de®ned as the maximum effect of the ®rst dose of
vecuronium 10 mg kg±1. The next incremental dose of
vecuronium 10 mg kg±1 was given after the twitch had again
stabilized. As long as neuromuscular block, de®ned as
[1±(T1/baseline T1)]3100, was less than 90%, additional
doses of vecuronium were given until 90% or greater block
had been achieved. The trachea was then intubated and the
inspired concentration of sevo¯urane was decreased from
886
2.5% to between 0.5 and 1.5%. Responses in the adductor
pollicis muscle were monitored continuously. Maximum
block produced by vecuronium 10 mg kg±1 and the time to
the onset of maximum block were recorded.
Statistics
Parametric data are presented as mean (SD) and non-
parametric data as median (range). Parametric data were
analysed with the t test and non-parametric data with the
Mann±Whitney U test. A P value less than 0.05 was
considered to indicate signi®cance.
Results
No signi®cant differences were found between the ocular
and generalized groups with respect to age, sex, weight,
duration of disease or serum concentration of binding
antibody to AChR. Blocking antibody was signi®cantly
higher in the generalized group than in the ocular group. The
generalized group included ®ve type IIa (mild, generalized)
and ®ve type IIb (moderate, generalized) patients according
to the Osserman classi®cation. Two patients in the ocular
group and six patients in the generalized group received
pyridostigmine (Table 1). Before vecuronium was admin-
istered, median TOF ratios were 95% (range 85±98%) in the
ocular group and 91% (55±100%) in the generalized group.
No signi®cant difference in TOF ratios was evident between
groups. After vecuronium 10 mg kg±1, the ocular group
showed a lower maximum block [median (range) 51
(0±98)% vs 91 (40±100)%], a slower onset of block [mean
(SD) 300 (73) vs 200 (74) s] and greater total doses needed to
exceed 90% block [median (range) 20 (10±30) vs 10 mg kg±1
(10±20 mg kg±1)] than the generalized group (P<0.05). Two
patients in the ocular group had at least 90% block, as did six
patients in the generalized group. No patient showed 100%
block in the ocular group, while three patients did in the
generalized group. Mean onset of maximum neuromuscular
block took over 200 s in all patients in the ocular group,
whereas six patients in the generalized group took less than
200 s. In the ocular group, three patients required
vecuronium 30 mg kg±1 to exceed 90% block. In contrast,
in the generalized group no patient required 30 mg kg±1
(Table 2).
Discussion
Although one might expect to see a difference in sensitivity
to neuromuscular blocking drugs between patients with
different types of myasthenia gravis, such differences have
not been established.10 11 In this study of 20 myasthenic
patients, the 10 patients in the ocular group showed
signi®cantly less neuromuscular block after vecuronium
10 mg kg±1 than the 10 in the generalized group. The ocular
group also showed a more gradual onset of block. The
median dose required to produce at least 90% block in the
 Varying sensitivity to vecuronium in myasthenia

Table 1 Patient characteristics. Summary values are mean (SD) or median (range). *Binding antibody to AChR (normally <0.37 pmol l±1); ²blocking antibody
to AChR (normally <10%); ³P<0.05 vs generalized group

Patient Age (yr), Weight Osserman Duration Binding Blocking Pyridostigmine

sex (kg) classi®cation of disease antibody* antibody² therapy
(months) (pmol ml±1) (%) (mg day±1)

Ocular group
1 50, M 49 I 5.5 0.67 21 ±
2 40, F 70 I 2.5 1.6 12 ±
3 72, F 45 I 84 0.9 1.4 ±
4 73, F 52 I 1.5 24 12 120
5 40, F 43 I 6 35 9.5 ±
6 62, F 59 I 5 0.6 0 180
7 63, M 69 I 5 2.1 7.5 ±
8 68, M 66 I 24 3.5 11 ±
9 70, F 56 I 15 5.1 10 ±
10 72, F 48 I 6 17 15 ±
61 (13) 56 (10) 15 (25) 9 (12) 10 (0±12)³
Generalized group
1 43, M 58 IIa 40 0.93 40 ±
2 58, M 55 IIa 34 6.0 41 180
3 69, F 50 IIa 80 63 10 90
4 52, F 48 IIa 12 48 29 120
5 57, F 57 IIa 3 35 3.2 ±
6 37, M 62 IIb 2.5 2.4 19 180
7 23, F 54 IIb 15 6.5 39 ±
8 53, F 64 IIb 11 2.0 39 240
9 48, F 47 IIb 90 13 29 ±
10 53, F 49 IIb 7 0.9 10 90
49 (13) 54 (6) 29 (32) 17 (23) 29 (10±41)

ocular group was twice that required in the generalized

group. Thus, ocular myasthenic patients were less sensi- Table 2 Mechanomyographic data and total vecuronium dose required to
tive to vecuronium than the patients with generalized achieve at least 90% block in each of the patients studied. Summary values
are median (range) or mean (SD). Maximum block (%) = maximum
myasthenia. depression of [1±(T1/baseline T1)]3100 after vecuronium 10 mg kg±1, where
As sensitivity to non-depolarizing neuromuscular block- T1 is the ®rst TOF response. *Interval between injection of vecuronium and
maximum block; ²type I in the Osserman classi®cation; ³type II in the
ing drugs is increased in myasthenic patients,1±6 we used an Osserman classi®cation; §P<0.05 vs generalized group
initial dose of vecuronium 10 mg kg±1, which is smaller than
the priming dose used ordinarily.12 13 Sensitivity to non- Patient Baseline Maximum Onset of Total dose
TOF ratio block block (mg kg±1)
depolarizing neuromuscular blocking drugs is known to (%) (s)*
vary greatly between myasthenic patients. Several studies
using a cumulative technique have shown the ED95 for Ocular group²
1 98 31 276 30
vecuronium to range from 6 to 44 mg kg±1.2 3 Although one 2 94 87 264 20
patient in our ocular group showed 98% block with 3 97 0 262 30
vecuronium 10 mg kg±1, exceeding the median value for 4 97 90 264 10
5 85 98 300 10
maximal block in the generalized group, the ocular 6 96 46 348 20
myasthenic patients as a whole showed less sensitivity to 7 98 38 264 30
vecuronium than the generalized myasthenic patients. 8 98 63 216 20
9 93 55 288 20
In myasthenia gravis, the ocular muscles are affected 10 95 37 480 20
most commonly, being involved initially in 40% of cases. It 95 (85±98) 51 (0±98)§ 300 (73)§ 20 (10±30)§
is rare for the initial symptoms to be limited to limb Generalized group³
muscles.6 However, intraoperative neuromuscular monitor- 1 100 40 288 20
2 95 47 192 20
ing is commonly performed in the upper limb. Our data 3 92 100 149 10
show that the limb muscles of ocular myasthenic patients 4 94 71 229 20
are signi®cantly less sensitive to vecuronium than those of 5 87 100 168 10
6 80 100 168 10
generalized myasthenic patients. Differences in sensitivity 7 55 98 60 10
between ocular and generalized myasthenia gravis patients 8 77 90 252 10
probably occur because the disease induces different 9 90 92 168 10
10 100 79 312 20
degrees of change in the margin of safety for neuromuscular 91 (55±100) 91 (40±100) 200 (74) 10 (10±20)
transmission in the limb muscles. In ocular patients,

887
 Itoh et al.
clinically important decreases in this margin of safety occur
only in ocular muscles. In contrast, generalized myasthenic
patients show a decreased margin of safety in all skeletal
muscles, including the adductor pollicis.
The ED50 for vecuronium has ranged from 0.3 to
19 mg kg±1 and the ED90 from 5 to 35 mg kg±1 in myasthenic
patients.1±3 In these studies, most patients represented type
II in the Osserman classi®cation, like the patients with
generalized myasthenia in our study. Additionally, type IV
patients (with late, severe generalized myasthenia) were
included in the study of Eisenkraft and colleagues.2 Only
one patient of this type was studied by Buzello and
colleagues.1 Patients in our generalized group were similar
in disease severity to most subjects in previous studies.
However, patients in our ocular group had milder
myasthenia than the individuals studied previously.
Although we did not calculate effective doses of vecu-
ronium, our data showed a difference in sensitivity between
ocular and generalized patients.
Our data were obtained under nitrous oxide±sevo¯urane
anaesthesia; in previous reports of myasthenic patients,
neuromuscular variables were measured in the absence of
potent anaesthetic agents. Generally, inhaled anaesthetics
decrease the availability of acetylcholine at the neuromus-
cular junction14 and increase the neuromuscular block
produced by non-depolarizing neuromuscular blocking
drugs,15±17 but this effect has not been speci®cally con-
®rmed in myasthenic patients. Vanlinthout and colleagues15
reported ED50 and ED95 values for vecuronium in
neurologically healthy patients during anaesthesia with
sevo¯urane 1.7% of 14.4 and 34.9 mg kg±1 respectively,
which are 42 and 32% less than when inhaled anaesthetics
are not used. The generalized myasthenic patients in this
study, who were apparently similar to the myasthenic
patients in previous studies of vecuronium, showed 91%
block from vecuronium 10 mg kg±1. Although comparisons
between studies have limitations, the block produced
appeared to be greater in our study than in previous studies
of myasthenia gravis; the sensitivity to vecuronium was
probably increased by sevo¯urane in myasthenic patients as
well as in our patients with normal neuromuscular trans-
mission.
However, the use of inhalational anaesthetics does not
affect the results of this study, which compared the
sensitivity to vecuronium of two types of myasthenia. One
limitation of this study, however, may be that we did
continue the morning dose of anticholinesterase. The
incidence of TOF fade before administration of vecuronium,
which is well tolerated in myasthenia gravis patients, was
low in our patients; this may have been caused by the
administration of pyridostigmine. The other limitation was
the interval between TOF stimuli. It has not been established
in patients with myasthenia gravis that an interval as long as
12 s between TOF stimuli is suf®cient. The twitches may
thus have been affected by preceding stimuli. We
standardized the interval at 12 s, as used in neurologically
888
healthy patients, as any effect of myasthenia in this respect
has not been shown.
In summary, we estimated neuromuscular block produced
by vecuronium in the adductor pollicis muscle in patients
with ocular and generalized myasthenia. Ocular myasthenic
patients were less sensitive to vecuronium than generalized
myasthenic patients. During general anaesthesia, neuro-
muscular monitoring is essential to avoid problems of
prolonged neuromuscular block in myasthenic patients;
however, the degree of block produced by a given dose of
vecuronium is likely to be less in ocular than in generalized
myasthenic patients.
References
1 Buzello W, Noeldge G, Krieg N, Brobmann GF. Vecuronium for
muscle relaxation in patients with myasthenia gravis.
Anesthesiology 1986; 64: 507±9
2 Eisenkraft JB, Book WJ, Papatestas AE. Sensitivity to vecuronium
in myasthenia gravis: a dose±response study. Can J Anaesth 1990;
37: 301±6
3 Nilsson E, Meretoja OA. Vecuronium dose±response and
maintenance requirements in patients with myasthenia gravis.
Anesthesiology 1990; 73: 28±32
4 Baraka A, Tabboush Z. Neuromuscular response to
succinylcholine±vecuronium sequence in three myasthenic
patients undergoing thymectomy. Anesth Analg 1991; 72: 827±30
5 Seigne RD, Scott RP. Mivacurium chloride and myasthenia gravis.
Br J Anaesth 1994; 72: 468±9
6 Miller JD, Lee C. Muscle Disease. In: Katz J, Benumof JL, Kadis LB,
eds. Anesthesia and Uncommon Disease, 3rd Edn. Philadelphia: W.
B. Saunders, 1990; 590±644
7 Osserman KE, Genkins G. Studies in myasthenia gravis: review of
a twenty-year experience in over 1200 patients. Mt Sinai J Med
1971; 38: 497±537
8 Lindstrom JM, Seybold ME, Lennon VA, Whittingham S, Duane
DD. Antibody to acetylcholine receptor in myasthenia gravis:
prevalence, clinical correlates, and diagnostic value. Neurology
1976; 26: 1054±9
9 Mittag T, Kornfeld P, Tormay A, Woo C. Detection of anti-
acetylcholine receptor factors in serum and thymus from
patients with myasthenia gravis. N Engl J Med 1976; 294: 691±4
10 Itoh H, Shibata K, Yoshida M, Yamamoto K. Neuromuscular
monitoring at the orbicularis oculi may overestimate the
blockade in myasthenic patients. Anesthesiology 2000; 93: 1194±7
11 Mann R, Blobner M, Jelen-Esselborn S, Busley R, Werner C.
Preanesthetic train-of-four fade predicts the atracurium
requirement of myasthenia gravis patients. Anesthesiology 2000;
93: 346±50
12 Brady MM, Mirakhur RK, Gibson FM. In¯uence of `priming' on
the potency of non-depolarizing neuromuscular blocking agents.
Br J Anaesth 1987; 59: 1245±9
13 Martin C, Bonneru JJ, Brun JP, Albanese J, Gouin F. Vecuronium
or suxamethonium for rapid sequence intubation: which is
better? Br J Anaesth 1987; 59: 1240±4
14 Waud BE, Waud DR. The effects of diethylether, en¯urane, and
iso¯urane at the neuromuscular junction. Anesthesiology 1975;
42: 275±80
15 Vanlinthout LE, Booij LH, van Egmond J, Robertson EN. Effect of
iso¯urane and sevo¯urane on the magnitude and time course of
 Varying sensitivity to vecuronium in myasthenia

neuromuscular block produced by vecuronium, pancuronium
and atracurium. Br J Anaesth 1996; 76: 389±95
16 Taivainen T, Meretoja OA. The neuromuscular blocking effects
of vecuronium during sevo¯urane, halothane and balanced
anaesthesia in children. Anaesthesia 1995; 50: 1046±9
17 Suzuki T, Munakata K, Watanabe N, Katsumata N, Saeki S,
Ogawa S. Augmentation of vecuronium-induced neuromuscular
block during sevo¯urane anaesthesia: comparison with balanced
anaesthesia using propofol or midazolam. Br J Anaesth 1999; 83:
485±7

889