Comparison of inte-
grated evoked EMG
between the hypothenar
and facial muscle groups
M.D. Sharpe MD, C.A. Moote MD, A.M. Lam MO,
P.H. Manninen MD
In 17 healthy patients undergoing 02" N20' isoflurane anaes-
thesia, following atracurium or vecuronium administration, we
compared simultaneous integrated evoked electromyograms
(IEEMGs) during spontaneous recovery of the adductor digiti
minimi (ADM) and orbicularis oris (OOM) muscle groups in
response to train-of-four (TOF) stimulation of the ulnar and
facial nerves, respectively. In all patients, the onset of neuro-
muscular recovery occurred first in the OOM. The time required
to recover to a T4/TI = 0.70 • 0.01 (SD) was earlier in the
OOM compared with the ADM muscles in both the atracurium
(33.4 • 5 vs46.5 +. 8, P < O.O05) andvecuronium(46.5 • 12
vs 60.3 • 20, P < 0.005) groups. When the OOM attained a
Tn/Tt = 0.70 +-- 0.01, the simultaneous T4/Tt in the ADM was
0.29 • 0.15 (P < 0.05) in the atracurium group and 0.41 •
0.16 (P < 0.01) in the vecuronium group. We conclude that (I)
the facial musc&s (OOM) recover earlier than the hypothenar
muscles (ADM) and (2) an EMG T4/Tt = 0.70 in the facial
muscles may not indicate adequate recover), of neuromuscular
Nous avons compare les traces dlectromyographiques int~SgrEs
de l'orbiculaire des paupiOres (ORB) et de I'adducteur du
petit doigt (ADD) dvoquEs simultanEment par stimulation en
train-de-quatre des nerfs facial et cubital; cela, pendant
I'anesthEsie avec 02, NzO et isoflurane de 17 sujets sains aprEs
injection d'atracurium ou de vEcuronium. Chez tousles pa-
tients, la recuperation de la fonction neuromusculaire survenait
d'abord d I'ORB. Le temps requis pour retrouver un ratio de
T4/TI +- 0,7 +- O, I (Ecart-type) Eta#plus court pour I'ORB que
pour I'ADD, ~ la fois avec I'atracurium (33,4 • 5 vs 46,5 • 8
min, P < 0,005)et avec le vEcuronium (46,5 +- 12 vs 60,3 • 20
min, P < 0,005). Quand le T4/TI de I'ORB atteignaitO, 7 • O, 1,
celui de rADD n'Etait que de 0,29 +-. 0,15 (P < 0,05) avec
l'atracurium et que de 0,41 • O, 16 avec le vdcuronium (P <
0,01). AprEs I' injection de myorela.rants, la fonction neuromus-
culaire des muscles du visage (ORB) r~cupEre plus rapidement
que celle des muscles de I' dminence hypothdnar (ADD) de plus,
un ratio T4/TI = 0,7 ~ I'EMG de la foce n'est pas synonyme
d'une r~cup~ration adequate de la fonction neuromusculaire.
MEASUREMENT TECHNIQUES" electromyography,
NEUROMUSCULAR RELAXANTS: atracurium, vecuronium.
From the Department of Anaesthesia, University Hospital,
University of Western Ontario, London, Ontario, Canada and
Department of Anesthesiology, University of Washington,
Address correspondence to: Dr. Michael Sharpc, Department
of Anaesthesia, University Hospital, 339 Windermere Road,
London, Ontario, N6A 5A5.
Accepted for publication 22nd November, 1990.
The importance of adequate reversal of muscle relaxants
following a surgical procedure cannot be overempha-
sized. Accurate detection of any residual neuromuscular
blockade predisposing the patient to respiratory muscle
weakness is essential in order to prevent hypoventilation
and/or upper airway obstruction. ~.z With the introduction
of the peripheral nerve stimulator using train-of-four
(TOF) and tetanic stimulation, the anaesthetist has a
bedside monitor of neuromuscular function. 3'4 Studies
stimulating the ulnar nerve and measuring the mechanical
response of the thenar muscles have correlated degrees of
neuromuscular blockade with indices of respiratory
function. 5-8 Due to its accessibility and convenience, the
facial nerve/muscle has also become a popular site for
monitoring of neuromuscular blockade. Its efficacy,
CAN J ANAESTH 1991 / 38:3 / pp318-23
Sharpe etal.: FACE VERSUS HAND EMG 319
however, as an accurate monitor to assess the degree of
residual neuromuscular blockade and therefore to predict
postoperative respiratory muscle function has not been
well defined. Our study was therefore designed to study
the relationship between the integrated evoked electromy-
ograms (IEEMGs) of the hypothenar and facial muscle
groups during recovery from neuromuscular blockade.
We wished to characterize the time course of recovery of
the T4T~ ratio from TOF facial nerve stimulation in
relationship to ulnar nerve stimulation, and in particular,
determine whether TOF stimulation of the facial and ulnar
nerves provided similar information regarding recovery
of neuromuscular function.
With institutional human subjects review board approval,
17 healthy patients (ASA physical status l-II) with no
known renal, hepatic or neuromuscular disease, aged
18-61 yr were studied. General anaesthesia was induced
with thiopentone (4-5 mg.kg-t), lidocaine (I.0 mg.
kg -l) and fentanyl (3 I.Lg- kg-~). Tracheal intubation was
facilitated with succinylcholine (0.5-0.75 mg.kg- t).
Maintenance anaesthesia consisted of oxygen/nitrous
oxide (FIO2 0.30-0.40 to maintain SaO2 - 96% (Nelcor
oximeter, Model NIO00) and isoflurane (0.5-1.0%
spired). Positive-pressure ventilation was used to main-
tain normocapnia (Nelcor Capnograph, Model NI000).
The two muscle relaxants were studied sequentially; nine
patients received vecuronium and eight patients received
Simultaneous IEEMGs were recorded with surface
electrodes from the OOM (facial) and ADM (hypothenar)
using two Datex NMT 221 monitors (Puritan-Bennett)
(Figure I). This monitor automatically delivers a supra-
maximal stimulus, stores the control response in memory
and displays the subsequent response to TOF stimulation
both as the ratio of the first twitch to control (TI/Tc) and
the ratio of the fourth twitch to first twitch (TJT0. During
the study continual recordings were made of the response
to TOF supramaximal stimuli delivered at a frequency of
2 Hz to the ulnar and facial nerves every 20 sec.
Before control IEEMGs were recorded, at least 30 min
were allowed to elapse following succinylcholine admin-
istration to ensure full recovery of the EMG. Atracurium
TABLE I Anthropometric characteristics
Age (yr) Weight (kg)
Group Mean ~ SD Range Mean • SD Range
Atracurium n = 8
Vecuronium n = 9
34 • 13
23 - 14
72 • 14
74 • 10
I j I ~-----
R Ref. S
and recording (R) electrodes on the face and hand. UN = ulnar
nerve, FN = facial nerve, ADM = adductor digiti minimi muscle,
OOM = orbicularis oris muscle, Ref = reference electrode.
Schematic representation of placement of stimulating (S)
(0.06-0.1 mg'kg -I) or vecuronium (0.03-0.05 mg-
kg -I) was administered (time 0) to abolish T4. The time
at which maximal blockade was achieved in each muscle
group as indicated by the T4/T~ digital display was
recorded as mark 3 on the recording strip. Spontaneous
recovery of neuromuscular function was monitored with
continuous display of Ta/TI and TdTc ratios. When the
T4/Tt ratio of the orbicularis oris muscle attained 0.70 -+
0.01, the simultaneous Ta/T~ ratio of the hypothenar
muscle was recorded. The recovery time (from time 0 to
T4/T~ ratio of 0.70) was measured for each muscle group.
For statistical analysis, comparison of the recovery
time for each muscle group was made using paired
Student's t test. Comparison between T4/T~ ratios of the
hypothenar and orbicularis oris muscles when the latter
was 0.70 was made using the Wilcoxin ranked-sum test.
Results are expressed as mean --- SD.
The anthropometric characteristics of the two groups are
shown in Table !. The time of onset of neuromuscular
blockade was similar in both muscle groups. Figure 2 is a
representative study comparing the TOF response in the
CANADIAN JOURNAL OF ANAESTHESIA
FIGURE 2 Simultaneous integrated evoked electromyograms of the adductor digiti minimi (hand) and orbicularis oris muscles (face) following
vecuronium administration. 1 = baseline IEEMG, 2 = vecuronium injection, 3 = point of maximum depression, 4 = TJT~ = 0.70 --- 0.01.
face and hand following vecuronium administration.
Although this particular study suggests a more intense
block in the facial muscle, we found no significant
difference in intensity of block between the two muscle
groups following atracurium or vecuronium administra-
tion. Note the earlier recovery of the facial EMG. The
orbicularis oris muscle group was first to recover in all
cases. The OOM recovered to a T4TI ratio = 0.70 --- 0.01
approximately 13 min earlier than the ADM muscle in
both the atracurium and vecuronium groups (P < 0.01)
(Table II). When the OOM had recovered to a T4TI =
0.70 --- 0.01, the corresponding T4TI of the ADM was
only 0.29 +-- 0.15 in the atracurium group (P < 0.05) and
0.41 --- 0.16 in the vecuronium group (P < 0.01) (Table
During the conclusion of a general anaesthetic, it is
essential to detect any residual neuromuscular blockade
prior to extubation of the trachea in order to prevent
respiratory insufficiency as a result of obstruction of the
airway due to upper airway muscle weakness. Presently
there is no direct bedside method to measure the integrity
of the upper airway muscles involved in maintaining
patency and providing protection of the upper airway as
well as the muscles of respiration. Historically, several
clinical tests such as head-lift, 9 armlift I~ tongue protru-
sion and intensity of hand-grip ~ have been utilized to
assess the adequacy of neuromuscular function and as
indirect tests of respiratory muscle function. However,
TABLE 11 Time (min) required to recover to a TJTt ratio =
0.70 --- 0.01 for facial and hand muscle groups
A tracurium (rnin) Vecuronium (rnin)
33.4 • 5*
46.5 --- 8
46.5 • 12"
60.3 4 20
Mean --- SD.
*P < 0.005, between muscle groups.
Sharpe etal.: FACE VERSUS HAND EMG 321
these tests require an awake and cooperative patient and
are not always accurate. 2'~2 Moreover, Pavlin et al. have
recently demonstrated that despite having adequate spon-
taneous respiration during recovery from d-tubocurarine
neuromuscular blockade, the patient may not be capable
of maintaining a patent airway. ~3 It is, therefore, an
established practice to monitor the neuromuscular junc-
tion using a peripheral nerve stimulator and observe the
mechanical response (MMG) of the thenar muscles to
TOF stimulation of the ulnar nerve. ~4 Studies have shown
that MMG T4TI ratio of 0.70 measured in the thenar
muscle groups is consistent with adequate recovery of
respiratory function. 5-7
Our results indicate that the facial nerve/muscle prepa-
ration recovers faster than the ulnar nerve/hypothenar
muscles; the OOM attained a T4T~ ratio of 0.70 by
approximately 13 min earlier in both atracurium and
vecuronium groups. At this time the T4T~ ratios of the
ADM muscles were only 0.41 in the atracurium group and
0.29 in the vecuronium group. Our results are in agree-
ment with previous studies 15- ~7 which qualitatively indi-
cated that the facial nerve recovered earlier than the hand
muscles. Caffrey et al. k6 had previously compared the
orbicularis oculi with the adductor pollicis as a monitoring
site for neuromuscular blockade, and although they
studied the MMG (mechanical myogram) and we used the
IEEMG, the results are remarkably similar with the hand
muscles lagging behind the facial muscles by approxi-
mately 13 min during recovery. In contrast, Paloheimo et
al. m8 attempted to study the comparative IEEMG of
orbicularis oculi and the hypothenar muscles and found
the results to be too variable to be conclusive, although
their results are consistent with an earlier recovery of the
facial muscles. The variability in their study, an observa-
tion we had also made in earlier pilot studies, may be due
to direct muscle stimulation. Such variability was not
observed using the orbicularis oris as we had done in this
study. Based on these findings, we conclude that adequate
recovery of the EMG response to facial nerve stimulation
is compatible with considerable residual neuromuscular
blockade and therefore does not ensure adequate respira-
tory function. It should be mentioned that not all facial
TABLE Ill T4/T~ ratios of hypothenar muscle when orbicularis otis
muscle T,/T~ = 0.70 --- 0.01
0.70 ~ 0.01
0.29 -+ 0.15;
0.70 --- 0.01
0.41 --- 0.16"
Mean --- SD.
*P < 0.05, between muscle groups.
tP < 0.01, between muscle groups.
muscles behave in the same manner, as it has been
demonstrated that the masseter muscle is more sensitive to
neuromuscular blockade than the adductor pollicis. 19
Indeed, as a muscle partly responsible for maintenance of
the airway, this sensitivity may account for the partial
airway obstruction observed during recovery from neuro-
muscular blockade reported by Pavlin et al. ~3 However,
clinically the masseter muscle is seldom used as a
monitoring site for neuromuscular blockade.
As previous studies concluding that a T4TI ratio 0f0.70
is consistent with adequate recovery of respirator), func-
tion are mostly measured by MMG, a-6 our assumption
that the information provided by the IEEMG is compara-
ble to MMG needs to be addressed, Four separate studies
comparing these two monitoring modalities, EMG vs
MMG, have shown that the MMG consistently lags
behind the EMG during recovery from a non-depolarizing
neuromuscular blockade. 2~
studies, 22"23 the magnitude of these differences was
relatively small and considered clinically unimportant.
Our own study on correlation between IEEMG and
respiratory function also confirmed that a T4T~ of >0.7 is
consistent with adequate recovery. 7 In contrast, Kop-
man 21 demonstrated a difference of 0.15 between EMG
and MMG T4Tt ratios which was consistent throughout
the study. A recent clinical study by Dupuis et al.
confirmed that the EMG lags behind the MMG by
approximately 20% which is compatible with Kopman's
observations. 24 This disparity between EMG and MMG
measurement of neuromuscular blockade may be related
to the type of anaesthetic administered. For example, the
studies by Kopman 2~ and Caffrey et al. ~6 indicating a lag
of the MMG utilized a halogenated agent to maintain
anaesthesia whereas the studies by Harper 22 and Weber 23
indicating no significant disparity between the EMG and
MMG, utilized a narcotic technique. Since the MMG is
not only dependent upon neuromuscular transmission but
also upon the contractility and loading of the muscle, the
lag of the MMG may be explained by the effects of the
halogenated agents upon these latter two factors. These
considerations do not affect the validity of our study,
since we did not use MMG, and the anaesthetic technique
used would have little influence on EMG. However, they
do lend further support to our contention that the OOM
electromyography may not be a reliable indicator of the
integrity of neuromuscular transmission as its recovery
preceded ADM recovery in all cases. As well, observa-
tions made by Laycock et al. 25 and Donati et al. 26 that the
diaphragm is more resistant to neuromuscular blockade
than the thenar muscles suggest that this important
respiratory muscle would also recover earlier than the
thenar muscles. It is therefore probable that the recovery
of OOM EMG may parallel that of the diaphragm and
However, in two of these
322 CANADIAN JOURNAL OF ANAESTHESIA
indicate at least the onset of recovery of respiratory
function. Indeed, the relative time course of recovery of
the diaphragm, orbicularis oculi (OOC) and adductor
pollicus was recently studied by Donati et al. and the
recovery potential of the diaphragm and OOC were shown
to be quite similar. 27 More importantly, however, the
recovery of the adductor pollicus occurred later than that
of the diaphragm or OOC which suggests adequate
recovery of other important muscle groups, i.e., upper
airway muscles involved in airway patency, has not yet
In summary, the facial nerve/muscle preparation is a
convenient and accessible site of monitoring neuromuscu-
lar blockade, and the orbicularis oris may be a better
alternative to orbicularis oculi for EMG recording.
However, similar to orbicularis oculi, its earlier recovery
probably reflects the time course of recovery of the
diaphragm and underestimates the recovery of other
important muscles of respiration necessary to maintain a
patent airway. We conclude that recovery of facial
nerve/muscle EMG may not represent adequate recovery
of respiratory function and emphasize the need for further
clinical evaluation of adequate muscle strength.
I Ali HH. A new device for monitoring force of thumb
adduction. Br J Anaesth 1970; 142: 83-5.
2 Waits LF, Levin N, Dillon JB. Assessment of recovery
from curare. JAMA 1970; 213: 1894-6.
3 Ali HH, Utting JE, Gray TC. Quantitative assessment of
residual antidepolarizing block (Part I). Br J Anaesth
1971; 43: 473-6.
4 Ali HH, Wilson RS, Savarese J J, Kitz RJ. The effect of
tubocurarine on indirectly elicited train-of-four muscle
response and respiratory measurements in humans. Br J
Anaesth 1975; 47: 570-3.
5 BrandJB, Cullen D J, Wilson NE, Ali HH. Spontaneous
recovery from non depolarizing neuromuscular block-
ade: correlation between clinical and evoked responses.
Anesth Analg 1977; 56" 55-8.
6 Ali HH, Kitz RJ. Evaluation of recovery from non-
depolarizing neuromuscular block, using a digital
neuromuscular transmission analyzer: preliminary report.
Anesth Analg 1973; 52: 740-4.
7 Sharpe MD, Lain AM, Nicholas JF et al. Correlation
between integrated evoked EMG and respiratory func-
tion following atracurium administration in unanaesthe-
tized humans. Can J Anaesth 1990; 37: 307-12.
8 Lee CM. Train-of-4 quantitation of competitive neuro-
muscular block. Anesth Analg 1975; 54: 649-53.
9 Ali HH, Utting JE, Gray TC. Quantitative assessment of
residual antidepolarizing block (Part II). Br J Anaesth
1971; 43: 478-85.
10 Bar ZG. The armlifi test. Anaesthesia 1985; 40: 630-3.
11 Viby-Mogensen J. Clinical assessment of neuromuscular
transmission. Br J Anaesth 1982; 54: 209-23.
12 Viby-Mogensen J, Jorgensen BC, Ording H. Residual
curarization in the recovery room. Anesthesiology
1979; 50: 539-41.
13 Pavlin EG, Holle RH, Schoene RB. Recovery of airway
protection compared with ventilation in humans after
paralysis with curare. Anesthesiology 1989; 70:38 I-5.
14 Day NS, Blake G J, Standaert FG, Dretchen KL. Charac-
terization of the train-of-four response in fast and slow
muscle: effect of d-tubocurarine, pancuronium, and
vecuronium. Anesthesiology 1983; 58: 414-7.
15 Ho LC, Crosby G, Sundaram P, Ronner SF, Ojemann
RG. Ulnar train-of-four stimulation in predicting face
movement during intracranial facial nerve stimulation.
Anesth Analg 1989; 69: 242-4.
16 Caffrey RR, Warren ML, Becket KE. Neuromuscular
blockade monitoring comparing the orbicularis oculi
and adductor pollicis muscles. Anesthesiology 1986; 65:
17 Stiff el P, Hameroff SR, Blitt CD, Cork RC. Variability in
assessment of neuromuscular blockade. Anesthesiolo-
gy 1980; 52: 436-7.
18 Paloheimo MPJ, Wilson RC, Edmonds HL Jr, Lucas LF,
Triantafillou AN. Comparison of neuromuscular block-
ade in upper facial and hypothenar muscles. J Clin Monit
1988; 4: 256-60.
19 Smith CE, Donati F, Bevan DR. Differential effects of
pancuronium on masseter and adductor pollicis muscles
in humans. Anesthesiology 1989; 71 : 57-6 I.
20 Katz RL. Electromyographic and mechanical effects of
suxamethonium and tubocurarine on twitch, tetanic and
post-tetanic fade. Br J Anaesth 1973; 45: 849-59.
21 Kopman AF. The relationship of evoked electromyo-
graphic and mechanical responses following atracurium
in humans. Anesthesiology 1985; 63: 208-II.
22 Harper NJN, Bradshaw EG, Healy TEJ. Evoked elec-
tromyograpic and mechanical responses of the adductor
pollicis compared during the onset of neuromuscular
blockade by atracurium or alcuronium, and during
antagonism by neostigmine. Br J Anaesth 1986; 58:
23 Weber S, Muravchick S. Electrical and mechanical
train-of-four responses during depolarizing and non
depolarizing neuromuscular blockade. Anesth Analg 1986;
24 Dupuis JY, Martin R, Tdtrault JP. Clinical, electrical and
mechanical correlations during recovery from neuro-
muscular blockade with vecuronium. Can J Anaesth 1990;