J7ournal ofNeurology, Neurosurgery, and Psychiatry 1993;56:1323-1326 1 323

SHORT REPORT

Primitive in Parkinson's disease

Fred W Vreeling, Frans R J Verhey, Peter J Houx, Jellemer Jolles

Abstract snout reflexes was not significantly different A standardised protocol for the exami- in patients with Parkinson's disease and nation of 15 in which healthy, age matched control subjects.' the amplitude and the persistence were Although a positive glabellar tap is con- scored separately, was applied to 25 sidered to be an important diagnostic sign of patients with Parkinson's disease and an parkinsonism,'4 it is also found in patients equal number ofhealthy matched control with intracranial disease who do not show any subjects. Most reflexes were found con- other signs, or who show symptoms of a siderably more often in the patients than clearly symptomatic parkinsonism.36 In one in the control subjects, especially the study, the glabellar tap reflex correlated best snout, the glabellar tap, and its variant, with the extent of the lesion and not with the the nasopalpebral reflex. Only the mouth site of the lesion.3 Several authors have open finger spread reflex was present reported the reversal of this reflex in patients more often in the control subjects. For after therapy with levodopa," 1215 all reflexes except this last, the scores for amantadine,II or lisuride'6; other authors, amplitude and persistence of the reflexes however, have not confirmed these findings.' 8 for the control group never exceeded the In another study, the palmomental reflex was scores for the patient group. Reflexes found to be even more reliable than the persisted more often in the patients than glabella reflex as a clinical indicator of in the control subjects. Parkinsonism Parkinson's disease.'3 alone can explain a large number of To summarise, the published findings and primitive reflexes, irrespective of the conclusions on primitive reflexes in parkin- severity or duration of the disease. In sonism are often confusing, controversial, or contrast, the number of reflexes was not readily compatible. This is not because of related more closely to cognitive scales. false diagnoses or heterogeneity of the patient It is concluded that such reflexes may be groups studied, but mainly because of a lack helpful in diagnosing Parkinson's dis- of compatibility of methodology used to elicit ease. In addition, a standardised protocol and score the primitive reflexes and the small for eliciting and scoring is essential for numbers of (mutually different) reflexes in the study of these reflexes in parkin- the various studies. sonism and other neuropsychiatric Recently, we found that experienced neu- conditions. rologists differ considerably in how they elicit and judge primitive reflexes.'7 Reliable mea- (7 Neurol Neurosurg Psychiatry 1993;56:1323-1326) surement in adult patients with neurological disease therefore requires a very elaborate protocol. A standardised protocol for the The prevalence and clinical value of primitive study of a 'primitive reflex profile' has not reflexes in Parkinson's disease have been dis- been applied in Parkinson's disease. Most University of Limburg, cussed by many authors."A Although some of studies have examined only one, and rarely Maastricht, The these reflexes can be elicited in normal adults5 two or three, primitive reflexes.'-3 8 9 Netherlands and in patients with focal lesions,6 they are The aim of the present study was therefore Department of found more often in patients with diffuse, to apply a standardised and semiquantified F W Vreeling hemispherical disease.7-'0 test battery of 15 primitive reflexes to patients Department of The glabellar tap,23" 12 the snout,'210 and with Parkinson's disease to determine the Neuropsychology and palmomental'24 '3 reflexes are especially fre- prevalence and the clinical value of these Psychobiology quently found in Parkinson's disease. The reflexes, compared with healthy controls, by F R J Verhey P J Houx proportion of glabellar tap, snout, nasopalpe- correlating them to parameters such as sever- J Jolles bral, suck, and grasp reflexes increases with ity and duration of the disease, cognitive Correspondence to: the severity of the disease.38 Findings for the functioning, and depression. The battery that Dr F W Vreeling, reflex are was chosen was found to have inter- Deparanent of Neurology, palmomental controversial.2 The high University of Limburg, presence of primitive reflexes increases with observer and intra-observer reliability.'7 The PO Box 616, 6200 MD No reflexes included: the Maastricht, The cognitive impairment.2810 relationship has primitive glabellar tap; Netherlands. been found between the reflexes and the palmar and plantar grasp; palmomental and Received 11 September 1992 duration of the disease or the degree of pollicomental; rooting; snout; suck; head and in revised form depression of the patient.28 In other studies, retraction; nuchocephalic; asymmetrical tonic 11 February 1993. Accepted 24 February 1993 the incidence of the palmomental and the neck; mouth open finger spread (MOFS); 1324 Vreeling, Verhey, Houx, J7olles

and palmar and plantar support reflexes. SCORING Most of these are well known in research on Amplitude and persistence of the reflex were neurological ageing.7 The last four reflexes scored separately. A three point scale for both have potential value for use in adults.'819 A characteristics was used. The scoring was as variant of the glabellar tap reflex, the follows: for amplitude, 0 = absent; 1 = a nasopalpebral reflex, was added to the bat- weak to moderate response; and 2 = a strong tery, because of its presumed clinical value in response. A well defined description for the parkinsonism.3 20 weak to moderate and strong amplitudes is given for each reflex. For persistence, 0 = absent; 1 = response for one to four consecu- Method tive times; and 2 = response for four or more SUBJECTS consecutive times. The glabellar tap and the Twenty-five patients with a diagnosis of pri- nasopalpebral reflexes were considered to be mary, degenerative Parkinson's disease were present but exhaustible after four to 10 con- selected at random from the neurological out- secutive responses and to be persistent after patient clinic. All had undergone an extensive more than 10 consecutive responses, respec- general and neurological examination, bio- tively. This is in accordance with the pro- chemical analysis and CT of the brain, to cedure followed by others.2 A detailed exclude other causes of parkinsonism. description of instructions, position, method Twenty-five healthy control subjects were of elicitation, and scoring of each reflex is matched to the patients with respect to age given elsewhere.26 and sex. All patients underwent the following examinations (table): the reflex battery; the Results Webster rating scale for severity of disease21; The prevalence of primitive reflexes in the Hoehn and Yahr scale for staging of patients and control subjects is shown in the parkinsonism22; the global deterioration scale figure. The prevalence increased with age in (GDS)23 and the mini mental state examina- the control group (p < 0.01), but not in the tion (MMSE)24 for assessment of cognitive patient group. The average number of functioning; and the Zung depression scale.25 reflexes per individual was, irrespective of The control subjects underwent a neuro- age, however, considerably higher in the logical examination including the reflexes; patients than in the control subjects: 4-6 and none of them showed any neurological sign, 0-8, respectively. For patients younger than or mental deterioration or depression in 60 years this was 3 9 (SD 0 2), between 60 neurocognitive testing. and 70 years 5-4 (SD 0 5), and for those older than 70 years 3-7 (SD 1 9). There was PROTOCOL OF THE PRIMITIVE REFLEX BATTERY no difference between men and women. Briefly, a basic position was described and the All but one of the reflexes occurred more subject was given instructions as to what was frequently in patients than in the control sub- expected of him or her-for example, sitting jects. The overall difference in frequencies or standing; eyes open or closed, etc.). The per reflex was significant (p < 0 01). The subject was not informed about the nature of glabellar tap and snout reflexes occurred in the expected response, but was always nearly all patients (96 and 92%, respectively). informed about the nature of the in These reflexes were also present in 12 and order to prevent startle reactions, which 20% of the control subjects, respectively. The might influence the required response. Every nasopalpebral reflex occurred in 88% of the reflex was measured at least three times, with patients but in none of the control subjects. about two seconds between each elicitation, The palmomental and pollicomental reflexes except for the glabellar tap and the were also found more often in patients than nasopalpebral reflexes, which were applied in the control group; so were the suck and left two times per second. All reflexes were nuchocephalic reflexes, albeit to a lesser assessed for amplitude and persistence, as extent. Only the left MOFS was present more proposed by other investigators.413 If apt, the often in the control subjects (20%) than in reflexes were elicited on both sides. the patients (4%). Six reflexes (asymmetric tonic neck, palmar and plantar grasp, head retraction, rooting, and plantar support) were Table Characteristics ofpatients and control subjects absent in both groups. Patients Controls In the control subjects, amplitudes were Number of subjects 25 25 never scored as 'strong' and only four out of Age (years) 66-5 (9-5) 67-5 (9-5) 21 responses were persistent (19%). In the Age range (years) 40-84 40-82 Sex 18 M17 F 18 M/7 F patients, the amplitude was scored as 'strong' GDS 2-0 (0-9) 1.0 six times, of which four were for the snout MMSE 27-4 (2 4) N/A MMSE (range) 21-30 N/A reflex; more than half of the responses, how- Zung depression scale 32-6 (8R0) N/A ever, (64 out of 118) were scored as 'persis- Hoehn and Yahr 2-3 (0 9) 0 Webster 11-5 (4 4) N/A tent' (54%). The most frequently persisting Parkinsonism since (years) 8-0 (6 2) N/A reflexes were the glabellar tap (19/24), the Diagnosed since (years) 6-4 (5 4) N/A nasopalpebral (11/22), the snout (19/23), and Unless indicated otherwise, mean values are given. Numbers the suck reflexes (5/8). Only the persistence in parentheses denote standard deviations. GDS = Global deterioration scale; MMSE = Mini mental of reflexes, not the amplitude, was related to state examination; N/A = not applicable. the patient group. Primitive reflexes in Parkinson's disease 1 325

Figure Percentage of patients with Parkinson's Glab Glab disease (left) and Naso Naso persistent healthy control subjects Palm L Palm L m N~~~~~not persistent (right) showing primitive R R I reflexes. Glab = glabellar Poll LI Poll L I tap; Naso = nasopalpetral t II reflex; Palm = R R I palmomental reflex; Snoul Snout Poll = pollicomental reflex; Suck Suck Snout = snout reflex; Suck = suck reflex; MOFS L MOFS L MOFS = mouth open Rtt R II finger spread reflex; Nuch L c Nuch L Nuch = nuchocephalic R I r n xs e n n r n rr R Icontrols reflex; SuPa = palmar SuPa L SuPa L support reflex. R R 0 10 20 30 40 50 60 70 80 90 100'%^%fv 0 10 20 30 40 50 60 70 80 90 100%

The average number of reflexes did not of the nuchocephalic reflex was not associated increase with the duration of the disease, with unilateral parkinsonism, in contrast to which was estimated retrospectively after the the palmomental reflex in Maertens de initial symptoms (range 2-25 years, mean Noordhout's study.4 8&0, SD 6 2), or with the number of years We could not test the negative correlation since the diagnosis of Parkinson's disease had between the glabellar tap and the palmomen- been established by a physician (range 1-22 tal reflexes and dyskinesia reported by Iriarte years, mean 6-4, SD 5.4). Hoehn and Yahr, et al,28 since too few patients showed dyskine- and Webster scores were closely related (p < sia. As for the reversal of reflexes-for exam- 0.01), but did not show a significant correla- ple, the glabellar tap, after starting levodopa, tion with the number of primitive reflexes. lisuride, or amantadine treatment, we could The number increased with the GDS: not test this because our patients were on a patients in stage 1-2 (n = 18) showed an stable drug regimen. Levodopa was taken by average of 4-2 reflexes; patients in stage 3-4 56% of the patients, amantadine by 48%, and (n = 9) showed 5-6 (p < 0 01). MMSE rat- the combination of both drugs by 16%. The ings were also weakly related (p < 0-05). glabellar sign was present in 96% of our Depression did not correlate with reflexes, patients, and 79% of these positive responses but it did with the time since the diagnosis showed persistence. According to some (p < 0-05), was established. Age did not cor- authors, looking for these reflexes could give relate with the number of reflexes, but it an objective, although indirect, evaluation of did-slightly-with the Webster, Hoehn and the patient's dopaminergic status.41' 12 This Yahr, GDS, and MMSE scores (p < 0 05). was not confirmed by Huber and Paulson.'9 Our findings do not support the view of Messina et alP", and Klawans et a!2 about the Discussion habituation or reversal of the glabellar sign. Once an individual has definite symptoms of The present findings suggest that the Parkinson's disease, some primitive reflexes glabellar tap, nasopalpebral, and snout show up and persist. The number of reflexes reflexes, and especially their persistence, may does not increase with the duration or sever- be of relevance in the examination of patients ity of the disease. The correlation with the with Parkinson's disease, in view of sugges- MMSE and GDS is compatible with the view tions that persistence or amplitude, or both, that these reflexes are a sign of diffuse cere- are correlated with the degree of cerebral bral dysfunction, rather than a symptom of a degeneration.268101217 A standardised proto- distinct neurological disease.3 6 9 10 27 col on how to elicit and score primitive Our results confirm other findings con- reflexes is required for the study of a broad cerning the most frequently found reflexes. profile of these signs in neuropsychiatric dis- Some state that the persisting glabellar tap orders. sign is probably the best correlative test in Parkinson's disease.' 014 Gimenez-Roldan et 1 Gossmann MD, Jacobs L. Three primitive reflexes in aP3 found the palmomental reflex to be an parkinsonism patients. Neurology 1980;30: 189-92. even more reliable clinical indicator. Our 2 Huber SJ, Paulson GW. Relationship between primitive reflexes and severity in Parkinson's disease. Neurol results are not in agreement with this, but Neurosurg Psychiatry 1986;49: 1298-300. they do agree on the amplitude and persis- 3 Pearce J, Aziz H, Callagher JC. Primitive reflex activity in primary and symptomatic parkinsonism. Neurol tence of the palmomental reflex. The (re-) Neurosurg Psychiatry 1968;31:501-8. appearance of the nasopalpebral reflex is 4 Maertens de Noordhout A, Delwaide PJ. The palmo- mental reflex in Parkinson's disease. Arch Neurol 1988; interesting, from an ontogenetic as well as 45:425-7. from a historical point of view.20 In our study, 5 Jacobs L, Gossmann MD. Three primitive reflexes in this normal adults. Neurology 1980;30: 184-8. compared with healthy control subjects, 6 van Tiggelen CJM. The Bracha reflexes. Akt Gerontol reflex seems to have an almost equally great 1983;13: 195-200. 7 Jenkyn LR, Reeves AG. Signs of cortical inhibition in sensitivity, and an even greater specificity for neuropsychiatric disorders. Psychiat Med 1984;1: parkinsonism than the glabellar tap reflex. 389-405. 8 Bakchine S, Lacomblez L, Pallison E, Laurent M, The snout reflex equals the glabellar sign as Derouesnet C. Relationship between primitive reflexes, to specificity and sensitivity. The asymmetry extra-pyramidal signs, reflective apraxia and severity of 1 326 Vreeling, Verhey, Houx, JoUles

cognitive impairment on of the Alzheimer 20 Simchowicz T. Ueber den Nasenaugenreflex und der type. Acta Neurol Scand 1989;79:38-46. Nasenkinnreflex. Dtsch Z Nervenheilkd 1922;75:342-55. 9 Jenkyn LR, Reeves AG, Warren T, et al. Neurologic signs 21 Webster DD. A method of measuring the dynamic charac- in senescence. Arch Neurol 1985;42: 1154-7. teristics of muscle rigidity, strength and tremor in the 10 Girling DM, Berrios GE. Extrapyramidal signs, primitive upper extremity. IRE Trans Med Electr 1959;6:159-64. reflexes and function in senile dementia of 22 Hoehn MM, Yahr MD. Parkinsonism: Onset, progression the Alzheimer type. Br_' Psychiatry 1990;157:888-93. and mortality. Neurology 1967;17:427-42. 11 Messina C, Di Rosa AE, Tomasello F. Habituation of 23 Reisberg B, Ferris SH, Crook T. The Global blink reflexes in Parkinsonian patients under levodopa Deterioration Scale (GDS): an instrument for the and amantadine treatment. J Neurol Sci 1972;17:141-8. assessment of primary degenerative dementia (PDD). 12 Klawans Jr. HL, Goodwin JA. Reversal of the glabellar AmJPsychiatry 1982;139:1136-9. reflex in Parkinsonism by L-Dopa. Jf Neurol Neurosurg 24 Folstein MR, Folstein SE, McHugh PR. Mini-Mental Psychiatry 1969;32:423-7. State, a practical method for grading the cognitive state 13 Gimenez-Roldan S, Esteban A, Abad JM. Reflejo palmo- of patients for the clinician. J Psychiat Res 1975; mentoniano en enfermedad de Parkinson. Arch 12:189-98. Neurobiol 1976;39:233-48. 25 Zung WWK, Durham NC. A self-rating depression scale. 14 Garland HG. Parkinsonism. BrMedJ 1952jan 19:153-5. Arch Gen Psychiatry 1965;12:63-70. 15 Penders CA, Delwaide PJ. Blink reflex studies in patients 26 Vreeling FW, Verhey FRJ, Jolles J. Protocol on the exami- with parkinsonism before and during therapy. J Neurol nation of primitive reflexes: basic position, instructions, Neurosurg Psychiatry 197 1;34:674-8. elicitation, response and scoring. Protocol: 87-0126 16 Sandrini G, Alfonsi E, Martignoni E, Horowski R, Nappi Neuropsychology and psychobiology project. State G. Effects of lisuride on blink reflex habituation in University of Limburg, Maastricht, The Netherlands, Parkinson's disease. Eur Neurol 1985;24:374-9. 1987. 17 Vreeling FW, Verhey FRJ, Houx PJ, Jolles J. Primitive 27 Franssen EH, Reisberg B, Kluger A, Sinaiko E, Boja C. reflexes in healthy, adult volunteers and neurological Cognition-independent neurologic symptoms in normal patients: Methodological issues. J Neurol 1993 (in aging and probable Alzheimer's disease. Arch Neurol press). 1991;48: 148-54. 18 Bronisch FW. Die Reflexe. Stuttgart: Thieme Verlag, 28 Iriarte LM, Chacon J, Madrazo J, Chaparro P, Vadillo J. 1979:86-9. Blink reflex in dyskinetic and nondyskinetic patients 19 Botez MI, Bogen JE. The grasp reflex of the foot and with Parkinson's disease. Eur Neurol 1989;29:67-70. related phenomena in the absence of other reflex abnor- 29 Huber SJ, Paulson GW. Influence of dopamine and dis- malities following cerebral commisurotomy. Acta Neurol ease severity on primitive reflexes in Parkinson's disease. Scand 1976;54:453-63. Eur Neurol 1989;29:141-4.