j., evaluation ofthe reflex measurement as an index …

POSTGRAD. MED. J., (1965), 41, 518

A CRITICAL EVALUATION OF THE TENDONREFLEX MEASUREMENT AS AN INDEX OF

THYROID FUNCTIONA. M. ROBSON, M.D. Ncle., M.R.C.P.

Luccock Research FellowR. HALL, M.D., B.Sc., Durh., M.R.C.P.

Wellcome Senior Research Fellow in Clinical Scienceand Honorary Lecturer in Medicine

G. A. SMART, M.D., B.Sc., Durh., F.R.C.P.Professor of Medicine

The Department of Medicine, University of Newcastle upon Tyne

IT IS now accepted that the tendon reflex isprolonged in patients with hypothyroidism.Chaney (1924) was the first worker to recordthe Achilles tendon reflex and prove what hadpreviously been detected clinically. Since thenseveral systems have been devised for recordingthe ankle jerk (Lambert, Underdahl, Beckett,and Mederos, 1951; Lawson, 1958; Gilson, 1959;Sharpe, 1961; Smart and Robson, 1963; Moulo-poulos, Koutras, and Kralios, 1964). Two ofthese methods, the Photomotograph (Gilson,1959) and the Kinemometer (Larson, 1958)are available commercially and are widely usedin the United States.Although many authors have confirmed

Chaney's original findings in adults (Lambertand colleagues, 1951; Lawson, 1958; Mann,1963; Sherman, Goldberg and Lawson, 1963)and in children (Bowers, Gordon and Segaloff,1959), the value of measuring the Achillestendon reflex as an index of thyroid functionis still in dou'bt. Measurement of the tendonreflex is a test which can be performed rapidlyand which causes the patient little incon-venience. The result is available immediatelyand is not invalidated by the administration ofthyroid replacement therapy or antithyroiddrugs. The test would therefore be potentiallyuseful if it were found to be reliable and diag-nostic. In this paper an attempt is made toassess the reliability and usefulness of the testin patients suspected of being hypothyroid orwho were thyrotoxic. Results obtained over aperiod of more than three years are summarised.

MethodsTendon reflexes were elicited and recorded using

the method described by Smart and Robson (1963).In all cases reflexes were elicited with the patientkneeling and, if possible, were obtained from theright leg. An interval of at least 15 seconds wasallowed between successive reflexes. Although theduration of the tendon reflex was unrelated to the

strength of stimulus used to elicit it, tendon reflextime was calculated from the mean of six satisfaotorytracings, the result being expressed to the nearestfive milliseconds.A normal range of results, using the 95% con-

fidence limits, was calculated from a series of 200normal subjects who had no clinical evidence ofthyroid disease. Half were hospital patients and theremainder members of the hospital staff and visitorsto hospital. The age range of the normal subjectswas from 14 to 84 years and there were equalnumbers of men and women.

Tracings were recorded from 223 patients (Table 1).Twenty-six of these patients were known to bethyrotoxic and were selected so that the effect ofthyrotoxicosis on the tendon reflex could be studied.The remaining 197 patients were referred by severalphysicians for measurement of their tendon reflexesbefore the state of their thyroid function was ascer-tained. All patients referred for measurement of theirtendon reflexes are included in this series with theexception of those receiving antithyroid drugs orthyroid hormone.The patient's thyroid status was assessed clinically

and confirmed by investigations which included 131uptake by the thyroid, protein-bound iodine estima-tions, resin uptake of 311-labelled triiodothyronine(Clark, 1963), serum cholesterol, electrocardiography,basal metabolic rate and antithyroid antibodies, buteach of these investigations was not performed onevery patient.

ResultsAnalysis of tracings

Typical tracings obtained from normal, hypo-thyroid and hyperthyroid subjects are shownin Fig. 1 and a diagrammatic representationof a typical tracing is shown in Figure 2. Theinitial upright deflection, produced by thestimulus of the tendon hammer stretching themuscle, is followed by a much larger deflectionresulting from contraction and relaxation ofthe muscle during the reflex.We have measured the time taken from the

onset of contraction to the end of relaxation(distance B-D in Fig. 2) and have called thisthe Achilles tendon reflex duration (A.T.R.D.).

by copyright. on July 31, 2022 by guest. P

rotectedhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.41.479.518 on 1 Septem

ber 1965. Dow

nloaded from

http://pmj.bmj.com/

September, 1965 ROBSON, HALL and SMART: Tendon Reflex Measurement

TABLE 1SUBJECTS INVESTIGATED IN THIS STUDY

Normal subjectsPatients with absent ankle jerksThyrotoxicsPatients suspected of hypothyroidism

(a) Hypothyroidi) Primary hypothyroidism 90ii) Following 131I therapy 5iii) Following drug ingestion 8 110iv) Other 1v) Panhypopituitarism 6

(b) Euthyroid 82(c) Undiagnosed 5

Total 427

Previous workers, using other methods forrecording the tendon reflex, found difficulty indetermining the end point of relaxation(Lambert and others, 1951; Gilson, 1959). Be-cause of this they measured the time takenfrom the point of stimulus to the point of halfrelaxation-distance A-C in Figure 2. Althoughwe have not experienced similar difficultiesusing our apparatus, we have also measuredthe "half-relaxation time" from our tracings(see Table 2) and have analysed both sets ofmeasurements.The distributions of A.T.R.D. and "half-

relaxation time" for the normal subjects weremarkedly skew but when expressed logarith-mically, these parameters were distributednormally (Fig. 4). Because of this all analyseswere performed after logarithmic transforma-tion but, for clinical convenience, figures usedin the text are the antilogarithms of the log-arithmic values.The two parameters produced similar co-

efficients of variation for the results obtainedfrom the normal subjects (Table 2) and theyproduced statistically similar degrees of separa-tion between the hypothyroid, hyperthyroidand normal subjects. In practice we found thatwhen trying to differentiate between hypo-thyroid and euthyroid subjects the "half-relaxation time" produced erroneous results in18% of cases whereas the A.T.R.D. measure-ment, in the same patients, was discordant in15%. In thyrotoxic subjects A.T.R.D. pro-duced 31% of results below the normal rangeand the equivalent figure for "half-relaxationtime" was 38%. In this paper all further resultsrefer to the A.T.R.D. parameter.

Normal subjectsThe mean values and 95 percentiles for the

normal subjects, calculated from log,o values,are shown in Table 2. Results from the hospital

'* i Ai l \IiC

FIG. 1.-Typical tracing from (a) normal (b) hypo-thyroid and (c) thyrotoxic subjects. The divisionson the horizontal axes represent 100 millisecondintervals.

patients and the normal volunteers, and fromthe males and females were analysed separately,but since no statistically significant differencecould be shown between these groups they werere-analysed together. No correlation betweenlength of A.T.R.D. and age could be demonst-rated in the 200 normal subjects whose agesranged from 14 to 84 years.Thyrotoxic subjectsWith one exception the A.T.R.D. values for

the thyrotoxic subjects fell in the lower normal

2004

26

197

-

I . I ,-,^ I Ia a . I

..

B

519

.A


rotectedhttp://pm

j.bmj.com

/P


j.41.479.518 on 1 Septem

ber 1965. Dow

nloaded from

http://pmj.bmj.com/

POSTGRADUATE MEDICAL JOURNAL

ANALYSIS OF RESULTSPARAMETER

Mean value ('logarithmic')Upper 95 percentileLower 95 percentileMean value (log,n)Standard deviationCoefficient of variation

ABA BIC D

FIG. 2.-Diagrammatic representation of a typicaltracing.

range or were lower than normal (Fig. 2). The'logarithmic' mean for this group (240 milli-seconds) was significantly less than that for thenormal subjects (301 milliseconds, P<0.001)but of the 26 thyrotoxic patients studied, 18(69%) had results in the normal range. Manythyrotoxic patients show a marked overswingand oscillation at the end of relaxation (Fig. 1)but this was not present in all thyrotoxicsubjects and could not be considered a diag-nostic feature.The effect of treatment with antithyroid

drugs on the A.T.R.D. was followed in tenthyrotoxic subjects. Tendon reflexes progres-sively lengthened and when the patient wasclinically euthyroid the A.T.R.D. was in thenormal range in each instance (Fig. 3). How-ever, estimation of the A.T.R.D. did not con-tribute to the management of these cases.

Hypothyroid subjectsOne hundred and ten of the 197 patients

referred for suspected hypothyroidism weresubsequently found to be hypothyroid. TheA.T.R.D. measurements from this group of

TABLE 2FROM THE 200 NORMAL SUBJECTS

A.T.R.D. '-RELAXATION300 milliseconds 275 milliseconds390 milliseconds 360 milliseconds230 milliseconds 205 milliseconds

2.4784 2.43620.05805 0.061512.34% 2.52%

FIG. 3.-The distribution of A.T.R.D. measurementsin the 26 thyrotoxic patients and the effect oftreatment in ten of these patients. The inter-rupted lines represent the 95% confidence limitsfor the normal subjects.

patients varied widely, the extreme values be-ing 250 milliseconds and 3,450 milliseconds.The values are shown in Fig. 4 which com-pares these results with those from the 200normal subjects. The 'logarithmic' mean forthis group-596 milliseconds-is significantlydifferent from that of the normal subjects (301milliseconds, P<0.001) but 13 (12%) hypo-thyroid patients had A.T.R.D. values withinthe normal range.The aetiology of the hypothyroidism in this

group of patients varied (Table 1). Most patients(90) had primary myxoedema; five had beenthyrotoxic and had been given 131I therapeu-

September, 1965520


rotectedhttp://pm

j.bmj.com

/P


j.41.479.518 on 1 Septem

ber 1965. Dow

nloaded from

http://pmj.bmj.com/


z 602c

NORMALS (200)

HYP

AT.R.D(M 00iS 2500AJ.R.D(MLUSECONDS)

FIG. 4.-The distribution of A.T.R.D. for the normaland hypothyroid subjects. One hypothyroidpatient with an A.T.R.D. measurement of 3,450milliseconds is not included in the figure.A.T.R.D. is plotted on a logarithmic scale.

tically; eight had iodide-induced hypothyroid-ism usually following the ingestion of 'Felsol'powders (Begg and Hall, 1963); one patientwas hypothyroid after a thyroidectomy forthyrotoxicosis and six patients had pan-hypopituitarism with evidence of hypothyroid-ism. We were unable to demonstrate anysignificant difference in the A.T.R.D. valuesbetween the groups of patients with differentcauses for their hypothyroidism, this findingagreeing with that of Lambert and colleagues(1951).With treatment of hypothyroidism the

patient's A.T.R.D. fell and with adequatetherapy it returned to the normal range. InFig. 5 the results of the A.T.R.D. before andafter treatment with L-thyroxine are comparedin 38 patients. In every case the tendon reflexduration shortened and in 34 (89%) the resultafter treatment was within the normal range.In the remaining four cases the A.T.R.D. valueafter treatment was just above the upper normal95 percentile, the longest value being 420milliseconds. In view of our recent experience,we think it is probable that many of thesepatients' A.T.R.D. values might have falleninto the normal range if the dose of L-thyroxinehad been increased further.

In many patients the A.T.R.D. value wasestimated at regular intervals after treatmentwith L-thyroxine was started and typical re-sults from four patients are shown in Fig. 6.In the majority of patients the tendon reflexprogressively shortened (patient J.B., Fig. 6),this improvement parallelling the improvementin the patient's general condition. The time

taken for the A.T.R.D. value to regain thenormal range varied from three weeks to fourmonths and at this time the patients wereclinically euthyroid. Sometimes the A.T.R.D.did not fall progressively with treatment buttended to fluctuate. However, it eventually fellto normal with adequate replacement therapyand remained in the normal range (see B.T.in Fig. 6). Although we have observed thisfluctuation in several patients we have neverobserved a lengthening of the tendon reflexabove the pre-treatment level.Measurement of the A.T.R.D. appears to be

a sensitive index of response to treatment. Inpatient S.R. the dose of thyroid therapy wasprogressively increased to 0.3 mg. L-thyroxineper day. On this dose the patient's A.T.R.D.hardly altered after six weeks of treatment andthe daily dose of L-thyroxine was increased to0.4 mg. The A.T.R.D. then rapidly fell to thelower normal range in one month and re-mained at this level. Similarly in patient N.I.a daily dose of 0.3 mg. L-thyroxine was in-sufficient to reduce the patient's A.T.R.D. tothe normal range. Although the patient feltbetter on this therapy, the dose of L-thyroxinewas increased to 0.4 mg. per day because histendon reflexes were still prolonged. When seentwo months later the patient had noticed amarked improvement and his A.T.R.D. hadfallen into the normal range.Euthyroid patients suspected of hypothyroidism

Eighty-two of the patients referred for sus-pected hypothyroidism were found, on furtherinvestigation, to be euthyroid. The remainingfive patients could not be allocated to eitherthe euthyroid or hypothyroid groups sincethere was insufficient information about theirthyroid function. Of the 82 euthyroid patients,16 (20%) had A.T.R.D. values above the normalrange (Table 3). In six of these patients theresult was either 395 or 400 milliseconds, i.e.,just above the normal upper 95 percentile. Infive further patients the A.T.R.D. was between400 milliseconds and 450 milliseconds, whereasin the remaining five patients the tendon reflexprolongation was even more marked. Clinicaldetails of the euthyroid patients with the mostmarkedly prolonged A.T.R.D. values are shownin Table 4.

DiscussionAlthough Sherman and others (1963) found

that the tendon reflex was shorter than normalin 75% of their thyrotoxic subjects and con-cluded that measurement of the tendon reflex

521by copyright.

on July 31, 2022 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.41.479.518 on 1 S

eptember 1965. D

ownloaded from

http://pmj.bmj.com/


FIG. 5.-The response of theA.T.R.D. to treatment in38 hypothyroid patients.The interrupted lines re-present the 95 % confidencelimits for the normal sub-jects.

was valuable in the diagnosis of thyrotoxicosis,our figures do not support this view. We foundthat 69% of thyrotoxic subjects had normalA.T.R.D. values and our findings are similarto those of Lambert, Underdahl, Beckett andMederos (1951), whose equivalent figure was75%, whereas Sabeh, Sarver, Moses andDanowski (1964) found that only one of theirthyrotoxic patients had a shorter than normalreflex. A variety of drugs have been reported toshorten the tendon reflex (Lawson, 1958; Fejerand Kun, 1964) and shortening of the reflexhas also been reported after episodes of stress(Fejer and Kun, 1964; West, 1964). We con-

clude, therefore, that measurement of the tendonreflex in thyrotoxicosis is of little diagnosticvalue.

In hypothyroidism the Itest produces a greaterdegree of separation between normal and ab-normal subjects. However, 12% of hypothyroidpatients had reflex times which fell below theupper 95 percentile for normal subjects. Sinceseveral euthyroid subjects had prolonged re-flexes it would appear wise not to diagnosehypothyroidism on the basis of this test alone.Although Mann (1963) found prolonged re-flexes in all his hypothyroid patients, tfiis isnot a universal finding. Most authors agree

z

IV)

r--__I

c:

September, 1965522by copyright.

on July 31, 2022 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.41.479.518 on 1 S

eptember 1965. D

ownloaded from

http://pmj.bmj.com/


.I-J

aC

Ni

:3 ..A . Z

.70

t2S

400

.' . - . :

6 8. i.-. : , : .. .

FIG. 6.-Detailed response to treatment in four hypo-thyroid patients. The interrupted lines representthe 95 % confidence limits for the normal subjects.

that there is an overlap between normal andhypothyroid subjects. Lawson (1958) found thisoverlap to be minimal. Sherman and others(1963) found it to be 6%; Lambert and others(1951) 23%; whilst Sabeh and others (1964)found prolonged reflexes in only half the hypo-thyroid patients investigated. We have foundthat it is the patients with obvious hypo-thyroidism who tend to have the longestreflexes and in the mildly hypothyroid patients,in whom there is diagnostic difficulty, thetendon reflexes tend to be in the upper normalrange or only slightly prolonged. This limitsthe diagnostic value of the test although similarcriticisms can be applied to many of the testsof thyroid function.Measurement of the Achilles tendon reflex

appears to have considerable value in assessing

the effectiveness of treatment in patients withthyroid disease. In nearly all hypothyroidpatients the tendon reflex time fell into thenormal range with adequate thyroid replace-ment therapy and in several patients the testindicated inadequate therapy when there waslittle clinical evidence of this. In thyrotoxicpatients we have found that the A.T.R.D.measurement will indicate overdosage withantithyroid drugs but this is of limited valuesince such a problem should not occur veryoften.

Weissbein and Lawson (1960) have demonst-rated the lengthening of the hypothyroid re-flex shortly after the institution of thyroidreplacement therapy. Our failure Ito observe thisphenomenon may have been because we didnot record the reflexes as soon after starting

-.. ..

523

*1


rotectedhttp://pm

j.bmj.com

/P


j.41.479.518 on 1 Septem

ber 1965. Dow

nloaded from

http://pmj.bmj.com/


TABLE 3THE DISTRIBUTION OF THE A.T.R.D. VALUES IN 192OF THE 197 PATIENTS SUSPECTED OF HYPOTHYROIDISM.Thyroid function in the remaining five patients could

not 'be determined.The interrupted line represents the upper 95 per

cent confidence limit for the normal subjects.A.T.R.D. RANGE No. OF PATIENTS IN RANGE

(milliseconds)

201251301331361 -371 -381

391 -

401411 -421431441451501601 -701801901

1,001Above

250300330360370380390

4004104204304404505006007008009001,0002,0002,000

Hypothyroid1015051

313334

1120179

10481

Total 110

Euthyroid21613156410

61210113000010

82

O

82

treatment as is necessary to demonstrate thisparadoxical lengthening.Approximately 20 of the 82 euthyroid

patients who were investigated for hypothyro-idism had A.T.R.D. values very close to thenormal upper 95 percentile (Table 3). It ispossible that several of these patients weremildly hypothyroid and that this was notdetected by other tests of thyroid function.This has been observed in one patient whohad an A.T.R.D. measurement (380 milli-seconds) in the upper normal range. Althoughinvestigations at this time failed to confirmthe diagnosis of hypothyroidism, she was re-viewed one year later when her A.T.R.D. valuehad lengthened to 425 milliseconds and re-investigation confirmed the suspected diagnosis.Another explanation for the clustering of

llany euthyroid results around the normalupper 95 percentile is that several of thesepatients were thought clinically to have pro-longed reflexes and were referred, and in-vestigated, for this reason. It is not surprisingthat several of these euthyroid subjects hadprolonged tendon reflexes. Two and a half percent of any normal group of patients would beexpected to have A.T.R.D. values above thenormal upper 95 percentile and this group of

524 September, 1965TABLE 4

CLINICAL DETAILS OF THE EUTHYROID PATIENTS WHOHAD MARKED PROLONGATION OF THEIR TENDON

IREFLEXES.PATIENT A.T.R.D. CLINICAL DETAILS

(milliseconds)S.A. 450 Obesity, Hypertensive. Myo-

cardial IschaemiaM.F. 500 Diabetes Mellitus-no peri-

pheral neuropathyH.K. 540 Hypertensive, Arterio-

sclerosisA.S. 590 Psoriasis, Obesity, Positive

antithyroid antibodiesA.D. 600 Obesity, HypertensiveE.L. 1,950 Clinically Hypothyroid -

investigations negativeeuthyroid subjects was highly selected sincethey were suspected of being hypothyroid.

Previous workers have reported prolonga-tion of tendon reflexes in euthyroid patients.Lambert and others (1951) found that obesepatients, as a group, had significantly longerreflexes than normal subjects and three ofour six euthyroid patients who had markedprolongation of their tendon reflexes wereappreciably overweight. One of our euthyroidpatients with prolonged reflexes was a diabeticand Bearwood and Schumacher (1964) foundthat diabetics often had prolongation of theirtendon reflexes outside the normal range and,as a group, they had a significantly longerreflex time than had normals. The euthyroidpatient wilth the most marked prolongation ofher reflexes was diagnosed clinically as beinghypothyroid, but investigations failed to con-firm this diagnosis. The sixth patient in thisgroup was hypertensive and arterioscleroticand recently a patient has been reported whohad ischaemia of a limb associated withatheroma and hypotension and was found tohave prolongation of her tendon reflex (Galpinand O'Brien, 1964). Prolongation of the tendonreflex has also been reported in sarcoidosis(Richards, 1962), neurosyphilis (Simpson, Blairand Nartowicz, 1963a), myasthenia gravis(Simpson and others, 1963b) and in hypo-kalaemia (Carr, Gill, Henkim and Bartter,1963).Although we have pointed out that measure-

ment of the Achilles tendon reflex has distinctlimitations, prolongation of the reflex is sosimple to measure and can be so strikingclinically, that it can often be a most helpfulaid in the diagnosis of hypothyroidism.SummaryTendon reflexes were recorded from 200

normal subjects, 26 thyrotoxic patients and 197patients who were thought to be hypothyroid.


rotectedhttp://pm

j.bmj.com

/P


j.41.479.518 on 1 Septem

ber 1965. Dow

nloaded from

http://pmj.bmj.com/

September, 1965 ROBSON, HALL and SMART: Tendon Reflex Measurement 525

In the latter group 110 patients were sub-sequently found to be hypothyroid, 82 to beeuthyroid and five patients could not be classi-fied because of inadequate data. The hypo-thyroid patients, as a group, had significantprolongation of their tendon reflexes irrespect-ive of the cause of the hypothyroidism, but12% had results below the upper 95 percentilefor normal subjects. Twenty-six per cent ofthe 82 euthyroid paitients, referred because ofsuspected hypothyroidism, had prolongedtendon reflexes and in six of these the reflexwas markedly prolonged. Three of this groupof patients were obese, one had diabetes mellitusand one was arteriosclerotic, prolonged tendonreflexes previously having been described ineach of these conditions. The mean Achillestendon reflex duration (A.T.R.D.) for thethyrotoxic patients was significantly shorterthan that for the normal subjects but 69%of these patients had results in the normalrange.We conclude that measurement of the Achil-

les tendon reflex is of little value in the diag-nosis of thyrotoxicosis. Despite its limitations,the technique can be helpful in the diagnosisof hypothyroidism and it is especially usefulin assessing 'the effect of treatment and inregulaiting the dosage of thyroid replacementtherapy given to hypothyroid patients.We would like to thank Dr. T. H. Boon, Dr. J.

Vallance-Owen and the many other physicians whoreferred patients for measurement of their tendonreflexes. We would also like to thank Dr. S. G.Owen for his help with the statistical analyses.

REFERENCESBEARDWOOD, D. M., and SCHUMACHER, L. R. (1964):Delay of the Achilles Reflex in Diabetes Mellitus,Amer. J. med. Sci., 247, 324.

BEGG, T. B., and HALL, R. (1963): Iodide Goitre andHypothyroidism, Quart. J. Med., 32, 351.

BOWERS, C. Y., GORDON, D. L., and SEGALOFF, A.((1959): The Myxedema Reflex in Infants andChildren with Hypothyroidism, J. Pediat., 54, 46.

CARR, A. A., GILL, J. R., JR., HENKIN, R. I., andBARTrER, F. C. (1963): Relationship of PotassiumMetabolism to the Achilles Tendon Reflex in Man,Clin. Res., 11, 215.

CHANEY, W. C. (1924): Tendon Reflexes in Myx-oedema: Valuable Aid in Diagnosis, J. Amer. medAss., 82, 2013.

CLARK, F. (1963): Resin Uptake of 131I-triiodothyro-nine: An In-vitro Test of Thyroid Function, Lancet,ii, 167.

FEJER, A. G., and KuN, M. (1964): The AchillesReflex, Lancet, ii, 695.

GALPIN, 0. P., and O'BRIEN, P. K. (1964): Prolonga-tion of Tendon Reflexes in Ischaemia, Lancet, ii,209.

GILSON, W. E. (1959): Achilles-reflex;IRecording witha Simple Photomotograph, New Engl. J. Med.,260, 1027.

LAMBERT, E. H., UNDERDAHL, L. O., BECKETT, S.,and MEDEROS, L. 0. (1951): Study of Ankle Jerkin Myxoedema, J. clin. Endocr., 11, 1186.

LAWSON, J. D. (1958): The Free Achilles Reflex inHypothyroidism and Hyperthyroidism, New Engl.J. Med., 259, 761.

MANN, A. S. (1963): The Value of Kinemographyin the Diagnosis of Thyroid Dysfunction, Amer. J.med. Sci., 245, 317.

MOULOPOULOS, S. D., KOUTRAS, D. A., and KRALIOS,A. C. (1964): A Simple Inexpensive Method forRecording Achilles Tendon Reflexes, Lancet, i,85.

RICHARDS, A. G. (1962): The "Myxedema Reflex"as a Presenting Sign in Sarcoidosis, Canad. med. Ass.J., 86, 32.

SABEH, G., SARVER, M. E., MOSES, C., and DANOWSKI,T. S. (1964): Triiodothyronine Binding to RedBlood Cells and Achilles Tendon Reflex as ThyroidIndices, Amer. J. med. Sci., 248, 253.

SHARPE, A. R. (1961): A Simple Isotope Methodfor Recording the Achilles Tendon Reflex inMyxedema, J. Lab. clin. Med., 57, 165.

SHERMAN, L., GOLDBERG, M., and LARSON, F. C.(1963): The Achilles Reflex: A Diagnostic Test ofThyroid Dysfunction, Lancet, i, 243.

SIMPSON, G. M., BLAIR, J. H., and NARTOWICZ, G. R.(1963): Prolonged Achilles Reflex in NeurosyphilisSimulating the "Myxedema Reflex". New Engi. J.Med., 268, 89.

SIMPSON, G. M., BLAIR, J. H., and NARTOWICZ, G. R.(1963): Diagnostic Limitations of Achilles TendonReflex in Thyroid Disease, N.Y. St. J. Med., 63,1148.

SMART, G. A., and ROBSON, A. M. (1963): A SimpleMethod for Recording Achilles Tendon Reflexes,Lancet, i, 363.

WEISSBEIN, A. S., and LAWSON, J. D. (1960): Para-doxical Lengthening of Muscle Contraction Timein Hypothyroid Patients given Smali Doses ofL-triiodothyronine, Thyroxine and DessicatedThyroid, J. clin. Endocr., 20, 429.

WEST, E. D. (1964): The Achilles Reflex, Lancet, ii,861.


rotectedhttp://pm

j.bmj.com

/P


j.41.479.518 on 1 Septem

ber 1965. Dow

nloaded from

http://pmj.bmj.com/

j., evaluation ofthe reflex measurement as an index …

Documents