JAW MUSCLE WEAKNESS: A DIFFERENTIAL INDICATOR OFNEUROMUSCULAR WEAKNESS—PRELIMINARY OBSERVATIONSSANDIP PAL, DM, and DEBASHIS SANYAL, MD

Department of Neurology, Bangur Institute of Neurosciences, Kolkata, West Bengal, India

Accepted 7 December 2010

ABSTRACT: Introduction: Flaccid quadriparesis is a commonneurological problem. Guillain–Barre syndrome Guillain-Barresyndrome (GBS), polymyositis/dermatomyositis (PM/DM), gen-eralized myasthenia gravis (MG), and hypokalemic periodic pa-ralysis (HPP) constitute the majority of cases of flaccidquadriparesis. Few patients from any of these disease groupslack the cardinal clinical features. We established clinicalmarker(s) that might have significant discriminating power for di-agnosis. Methods: Forty-six patients satisfied all of our criteria.Cases were evaluated clinically followed by laboratory and elec-trophysiological study, and, in selected cases, muscle histopa-thology. Results: Twenty-four patients had GBS, 9 had MG, 7had PM/DM, and 6 had HPP. Jaw-opening weakness was foundin 71.4% of PM/DM, 83.3% of HPP, and 4.1% of GBS cases.Jaw-closing weakness was found in 88.8% of MG cases.Conclusions: Presence of jaw-closing weakness pointedtoward MG, whereas presence of jaw-opening weakness sug-gested muscle disease (PM/DM and HPP). GBS patients veryrarely had jaw muscle weakness.

Muscle Nerve 43: 807–811, 2011

Flaccid quadriparesis is one of the many commonproblems afflicting patients seen in emergency andneurology departments. Excluding acute uppermotor neuron diseases, some common etiologiesto consider include Guillain–Barre syndrome(GBS), polymyositis/dermatomyositis (PM/DM),generalized myasthenia gravis (MG), and hypokale-mic periodic paralysis (HPP). Although there arewell-documented clinical and electrodiagnostic fea-tures for each of these entities, at times diagnosisbecomes difficult due to a paucity of the cardinalfeatures in some patients. For example, general-ized arreflexia may not be present in 20% of casesof GBS. Indeed, in some cases, muscle stretchreflexes may be preserved throughout the illnessand lead to confusion in diagnosis and selection ofappropriate investigations.1 Similarly, thoughinflammatory muscle diseases are characterized byhigh creatine kinase (CK), in some cases of derma-tomyositis, CK activity may remain normal or beonly slightly elevated.2 Although ptosis and diplo-pia are the most common presentation of MG,they may be conspicuously absent in 10% ofpatients3 throughout their disease course.

Although HPP is a rare channelopathy, it is quitecommon in this part of India. HPP is characterizedby short-duration, recurrent, flaccid quadriparesiswith near sparing of bulbar, facial, and respiratorymuscles. At the height of weakness, however, theremay be areflexia and, in cases of secondary hypo-kalemia, the duration of paralysis may be pro-longed and lead to confusion with GBS, particu-larly in an initial attack.4 Rowin indicated that thepattern of weak jaw closure and relatively preservedjaw opening is typical of MG.5 There is occasionaljaw muscle weakness in PM.6 Considering the diffi-culties in making a diagnosis in 10–20% of thecases when looking for classical features and usingestablished investigative methods, there is a needto look for additional clinical features that mayoffer better discriminative value in making a spe-cific diagnosis of acute flaccid quadriparesis. Thisis a vital issue in India and elsewhere, because acost-effective selection of tests to confirm the diag-nosis is of paramount importance.

METHODS

This study was conducted in the Department ofNeurology, Bangur Institute of Neurosciences, Kol-kata, India, between January 2008 and December2009. The cases were recruited prospectively fromconsecutively admitted patients. The inclusion cri-terion was acute (duration �1 month) quadripare-sis. Exclusion criteria were upper motor neurontype weakness, as determined either clinically or byimaging, and grossly altered sensorium of anycause. The clinician who performed the clinical ex-amination was unaware of any investigative find-ings that could have created bias.

Patients were evaluated through detailed his-tory and clinical examination. Special emphasis inthe history was given to duration of illness, dyspha-gia, nasal regurgitation and intonation, chewingdifficulty, neck weakness, and bladder and respira-tory symptoms. During examination, special stresswas placed on muscles related to jaw opening andclosing; facial, palatal, pharyngeal, and tonguemovement; and neck flexion and extension. Jaw-opening weakness was assessed conventionally byasking the patient to open the mouth forciblyagainst resistance from the examiner. Jaw-closingweakness was assessed by asking the patient to for-cibly close the upper and lower jaws against atongue spatula covered with gauze while the

Abbreviations: ACh, acetylcholine; CHAID, chi-square automatic interac-tion detection; CK, creatine kinase; GBS, Guillain–Barre syndrome; MG,myasthenia gravis; MRC, Medical Research Council; PM/DM, polymyosi-tis/dermatomyositis; HPP, hypokalemic periodic paralysis; RNST, repetitivenerve stimulation test; SFEMG, single-fiber electromyography

Correspondence to: S. Pal; e-mail: sandip100_pal@yahoo.co.in

VC 2011 Wiley Periodicals, Inc.Published online 15 May 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mus.21990

Key words: differential indicator, flaccid quadriparesis, jaw-openingweakness, jaw-closing weakness, neuromuscular weakness

Jaw Muscle Weakness MUSCLE & NERVE June 2011 807

clinician pulled against it. Neck muscle weakness(both flexors and extensors) was assessed by con-ventional means, and limb weakness (both proxi-mal and distal groups) was assessed on the basis ofMedical Research Council (MRC) grading. Musclestretch reflexes were classified as lost, decreased(þ), normal (þþ), or increased (>þþ). Sensoryexamination was performed using standard con-ventional procedures, as is done during the evalua-tion of superficial and deep sensations.

Routine hemogram, fasting blood sugar, urea,creatinine, sodium, potassium, serum CK, and liverfunction tests were done for every patient. Detailedelectrophysiological tests, including motor and sen-sory nerve conduction studies, late responses, low-rate (3-HZ) repetitive nerve stimulation, and electro-myography were undertaken for each patient. Serumacetylcholine (ACh) receptor antibody assay wasdone in selected cases. Muscle biopsy with histopath-ological examination was done selectively as well.

Electrophysiology in cases of GBS, low-rate re-petitive nerve stimulation test (RNST) with serumACh receptor antibody for MG, histopathology ofthe muscle biopsy for PM/DM, and serum potas-sium level for HPP were taken as the ‘‘gold stand-ard’’ tests for final diagnosis.

Statistics.. The frequency of various categoricalvariables and mean and standard deviation of vari-ous numerical variables were used as summarymeasures of the data. A chi-square test (or a Fisherexact test when the chi-square test was not applica-ble) was used to test the association between differ-ent diagnostic groups and various categorical varia-bles. Monte Carlo approximation was done if theFisher exact test could not be evaluated com-pletely. In all cases, two-tailed tests were used, andP < 0.05 was considered statistically significant.

Exhaustive chi-square automatic interactiondetection (CHAID) was utilized to identify diagnosticclinical features, which can be used in establishingthe predictive model for etiological diagnosis in casesof acute flaccid paralysis. Statistical analysis was doneusing SPSS software (version 15.0) for Windows.

RESULTS

During the study period, we had 46 patients whofulfilled the inclusion and exclusion criteria.Among them, the final diagnosis was GBS in 24cases, myasthenia gravis in 9 (all generalized),PM/DM in 7, and periodic paralysis in 6 (allHPP). In the inflammatory muscle group, 3 haddermatomyositis, and 4 had polymyositis based onhistopathology. Dermatomyositis was diagnosed onthe basis of perifascicular atrophy, inflammatorycell infiltration in perimysial connective tissue, andperivascular inflammation involving small- and me-

dium-sized vessels along with marked variation indiameter, myophagocytosis, regenerating fibers,and fibers with vacuoles. Polymyositis was diag-nosed on the basis of variable involvement of fas-cicles, myophagocytosis, islands of regeneratedfibers, and fiber vacuolation with inflammatory-cellinfiltration most prominent in relation to musclefiber membrane and endomysium.

In the GBS group, the age of presentationranged from 5 to 70 years (mean 38.5 years). Malesoutnumbered females by a 17:7 ratio. Duration of ill-ness ranged from 3 to 30 days (mean 10 days). Inthe MG group, the age of presentation ranged from14 to 55 years (mean 30.5 years), and the male:fe-male ratio was 5:4. Duration of illness ranged from20 to 30 days (mean 27.6 days). The PM/DM groupconsisted of 7 patients with ages ranging from 22 to78 years (mean 39.2 years). The male:female ratiowas 1:6. Duration of illness ranged from 20 to 30days (mean 28 days). The HPP group consisted of 6patients with age ranging from 16 to 50 years (mean41.3 years). Males and females were equal in num-ber. Duration of illness ranged from 2 to 5 days(mean 3 days). Clinical features are given in Table 1.In the GBS group, muscle stretch reflexes were nor-mally preserved in all four limbs in 2 cases (8.3%),diminished or normal in upper limbs and absent inlower limbs in 5 cases (20.8%), and diminished butpresent at all sites in 1 case (4.2%). In the MGgroup, both diplopia and ptosis were absent in 2cases (22%), 3 had only ptosis, and 1 had onlydiplopia.

Figure 1 shows the decision tree (exhaustiveCHAID) analysis to predict diagnosis using the signif-icant variables shown in Table 1 as predictors. Jaw-closing weakness emerged as the most prominentand significant predictor. All patients who had jaw-closing weakness turned out to be MG cases. If jaw-closing weakness was absent, then the most promi-nent clinical variable for predicting the diagnosiswould be jaw-opening weakness. In patients with nor-mal jaw power, both opening and closing, 85.2%cases were GBS, and if additional lower limb areflexiais considered, 100% had GBS. In the group with jaw-opening weakness without jaw-closing weakness, pha-ryngeal weakness was the best discriminating clinicalvariable. All patients with jaw-opening weakness with-out pharyngeal weakness turned out to be HPP cases.On the contrary, 83.3% of patients with jaw-openingweakness with pharyngeal weakness were ultimatelydiagnosed as PM/DM.

Table 2 shows the predictive value of the clini-cal features found from CHAID for the diagnosisof different diseases. The highest success rate forclinical prediction was for GBS (95.8%), followedby MG (88.9%), HPP (83.3%), and PM/DM(71.4%).

808 Jaw Muscle Weakness MUSCLE & NERVE June 2011

Table 3 shows the predictive value of the clini-cal features excluding jaw muscle weakness foundfrom CHAID for the diagnosis of different dis-eases. The success rates for clinical prediction ofGBS, MG, HPP, and PM/DM were 87.5%, 66.7%,66.7%, and 0%, respectively.

DISCUSSION

In this study, acute-onset quadriparesis with gener-alized areflexia (66.7%) was the predominant pre-sentation in the GBS group, which is consistentwith previous reports.1 Patients who did not havegeneralized areflexia were assessed during presen-tation, which ranged from 7 to 20 days after onset.This suggests that duration of illness had nothingto do with the preservation of reflexes. Bifacialweakness was present in 41.7% of patients in ourGBS series, which is again consistent with the pre-vious reports.4,6 Jaw-closing weakness was absent inall cases, whereas only 1 patient had mild jaw-open-ing weakness. This relative preservation of jaw-mus-cle power is in sharp contrast with frequentinvolvement of nearby muscles such as facial(41.7%) and neck flexor (79.1%) muscles in GBSgroup. We can say that jaw muscle weakness goesvery much against the diagnosis of GBS, irrespec-tive of the severity of spinal and other craniocervi-cal muscle involvement.

In the MG group, the cardinal features (bothdiplopia and ptosis) were absent in 22% of cases,although the literature reported rates of around10%.3 These patients had jaw-closing weakness, butnone had facial weakness. In another 4 patientswho had either ptosis or diplopia, facial weaknesswas present in 2, with jaw-closing weakness presentin all 4. From this we might say that weakness of

jaw closing is a very strong clinical marker of gen-eralized MG even when other cranial muscles,including extraocular muscles and elevators of eye-lids, are clinically not involved. Another notablefinding is the absence of jaw-opening weakness inevery MG patient. The observation by Rowin,5 whonoted that the pattern of weak jaw closure and rel-atively preserved jaw opening is typical of MG,held true for our series as well and we are cur-rently looking to confirm Rowin’s observation witha specific prospective study. In a different study,investigators found that single-fiber electromyogra-phy (SFEMG) of the masseter muscle was abnor-mal in 6 of 9 ocular MG patients and in all gener-alized cases (overall sensitivity: 27 of 30 cases, or90%; confidence interval 79.3–100.0%; P ¼ 0.95)and they proposed that the masseter should beconsidered for SFEMG in the diagnosis of MG,especially in cases with bulbar onset.7 The reasonfor this preferential jaw-closing weakness is notknown. The proposed reasons for preferential ocu-lar involvement include: (i) slight weakness in alimb may be unnoticed, but slight weakness in theextraocular muscles would lead to misalignment ofthe two eyes; (ii) differences in the antibodies inocular MG vs. generalized MG that may favor themuscles responsible for eye movement and eyelidelevation; (iii) compared with extremity muscles,extraocular muscles are smaller, served by morenerve fibers, and are among the fastest contractingmuscles in the body, which would lead to a higherlevel of activity and would predispose them to fa-tigue in MG; and (iv) some reports indicated thatthere may be fewer ACh receptors in extraocularmuscles than in limb muscles. What we can pro-pose for the preferential involvement of jaw

Table 1. Clinical features in each group.

Diagnostic group GBS (n ¼ 24) MG (n ¼ 9) PM/DM (n ¼ 7) PP (n ¼ 6) P-value

Sensory symptoms 4 (16.6) 0 0 1 (16.6) NS*,†

Sensory impairment 3 (12.5) 0 0 0 NS*,†

Respiratory compromise 4 (16.6) 2 (22.2) 2 (28.5) 0 NS*,†

Diplopia 0 4 (44.4) 0 0 0.001‡,†

Ptosis 0 6 (66.6) 0 0 <0.001‡,†

Jaw-opening weakness 1 (4.1) 0 5 (71.4) 5 (83.3) <0.001§,†

Jaw-closing weakness 0 8 (88.8) 0 0 <0.001§,†

Bifacial weakness 10 (41.6) 3 (33.3) 4 (57.1) 0 NS*,†

Pharyngeal weakness 5 (20.8) 4 (44.4) 6 (85.7) 0 0.002‡,†

Palatal weakness 5 (20.8) 4 (44.4) 5 (71.4) 0 0.02‡,†

Tongue weakness 2 (8.3%) 0 0 0 NS*,†

Neck-flexor weakness 19 (79.1) 9 (100) 7 (100) 4 (66.6) NS*,†

Neck-extensor weakness 10 (41.6) 5 (55.5) 6 (85.7) 4 (66.6) NS*,†

Upper limb-muscle stretch reflex absent 16 (66.6) 0 5 (71.4) 3 (50) 0.002§,†

Lower limb-muscle stretch reflex absent 21 (87.5) 0 4 (57.1) 2 (33.3) <0.001§,†

Figures in parentheses indicate respective percentage of column total.*Not significant.†Fisher exact test with Monte Carlo approximation.‡significant.§highly significant.

Jaw Muscle Weakness MUSCLE & NERVE June 2011 809

muscles are: (i) the difference in tissue-specific my-osin in carnivore muscles of mastication from thatof limb muscles8; and (ii) possible difference inantibody type. Moreover, jaw-closing muscles areanti-gravity muscles and are in constant contrac-

tion, and hence they fatigue and become weakcompared with jaw-opening muscles.

In general, retained reflexes, particularly theankle jerk, characterize muscle diseases. In ourstudy, of the total 13 patients with PM/DM and

Table 2. Predictive value of clinical features for each group.

Observed

Predicted

PP PM/DM GBS MG Percent correct

PP (6) 5 0 1 0 83.3%PM/DM (7) 0 5 2 0 71.4%GBS (24) 0 1 23 0 95.8%MG (9) 0 0 1 8 88.9%

Table 3. Predictive value of clinical features excluding jaw muscleweakness for each group.

Observed

Predicted

PP PM GBS MG Percent correct

PP 4 0 2 0 66.7%PM 3 0 4 0 0.0%GB 3 0 21 0 87.5%MG 3 0 0 6 66.7%

FIGURE 1. Decision tree (exhaustive CHAID) analysis for predicting diagnosis. The significant variables from Table 1 were used as

predictors. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

810 Jaw Muscle Weakness MUSCLE & NERVE June 2011

HPP, generalized arreflexia was found in 6 cases(46.1%). In this combined group, jaw-openingweakness was present in 10 cases (76.9%), whereasnone had jaw-closing weakness. On the contrary,in the GBS group, only 1 patient had jaw-openingweakness. This is a noteworthy finding, but the rea-son behind this preferential jaw-opening involve-ment in muscle disease is not clear. The jaw-open-ing muscles are intrinsically weaker than jaw-closing muscles, and this may explain the preferen-tial involvement of jaw-openers in muscle diseases(inflammatory muscle disease and periodicparalysis).

This observation of predominant jaw-openingweakness with preserved jaw-closing power in pri-mary muscle disease (PM/DM and HPP) is inopposition to what we found in the MG group.Again, this contrasts sharply to what we found inthe GBS group, where both opening and closureof the jaw were nearly spared.

A small sample size, lack of quantification ofdegree of weakness of jaw muscles, and non-avail-ability of SFEMG somewhat limit the validity of ourfindings. However, the clinical data clearly supportthe correct diagnosis.

In conclusion, in our study we found that jaw-opening or -closing muscle involvement or lack ofit was a strong discriminator among the differentdiseases involving the nerve roots, neuromuscularjunction, and muscles. As clinical cardinal features

may be absent in 10–20% of patients, this findingmay help to direct the clinician to the correct diag-nosis. This diagnostic tree may help us to properlyutilize scarce investigation resources for thosepatients who need it most.

The authors thank Professor Shyamal Kumar Das (Head OfDepartment), Department of Neurology (Bangur Institute of Neu-rosciences, Kolkata), and Dr. Sumitava Samanta (DM Neurologyresident, Bangur Institute of Neurosciences, Kolkata) for proof-reading and other valuable contributions.

REFERENCES

1. Donaghy M. Polyneuropathy. In: Donaghy M, editor. Brain’s diseasesof the nervous system, 12th ed. Oxford: Oxford University Press;2009. p 563–568.

2. Karpati G, Currie GS. The inflammatory myopathies. In: Walton J,Karpati G, Hilton- Jones D, editors. Disorders of voluntary muscles,6th ed. Edinburgh: Churchill Livingstone; 1994. p 620.

3. Oosterhuis H. The ocular signs and symptoms of myasthenia gravis.Doc Ophthalmol 1982;52:363–378.

4. Yadollah H, Bosh PE. Disorders of peripheral nerves. In: Bradley WG,Daroff RB, Fenichel GM, Jankovic J, editors. Neurology in clinicalpractice, 5th ed. Philadelphia: Butterworth Heinemann Elsevier;2008. p 2288–2296.

5. Rowin J. Approach to the patient with suspected myasthenia gravis orALS: a clinician’s guide. Academy of Neurology continuum series.Myasthenic Disord ALS 2009;15:13–34.

6. Ropper AH, Samuels MA. The infectious and inflammatory myopa-thies. In: Sydor AM, Davis KJ, editors. Adams and Victor’s principlesof neurology, 9th ed. New York: McGraw Hill; 2009. p 1355–1361.

7. Khuraibet AJ, Rousseff RT, Behbehani R, Al-Shubaili AFK, Khan RA.Single-fiber electromyography of masseter muscle in myastheniagravis, Muscle Nerve 2008;37:522–525.

8. Porter JD, Khanna S, Kaminski HJ, Rao JS, Merriam AP, RichmondsCR, et al. Extra-ocular muscle is defined by a fundamentally distinctgene expression profile Proc Natl Acad Sci USA 2001;98:12062–12067.

Jaw Muscle Weakness MUSCLE & NERVE June 2011 811