Abstract
The digiti quinti sign (DQS), described originally as a clinical indication of subtle motor deficit, consists of a relatively greater abduction of the fifth finger on the affected side when both arms are extended forwards. This sign was previously observed interictally in three consecutive hemiplegic migraine (HM) patients.
Patients and methods: To verify whether the DQS specifically discriminates HM from non-hemiplegic migraine (nHM), the angle between the fourth and fifth fingers (ANG) was measured interictally in 10 HM patients, 44 migraine with aura and migraine without aura patients, and 45 healthy controls.
Results: The ANG was significantly wider at the symptomatic side in HM as compared with nHM and controls. The differences between the symptomatic and non-symptomatic (for HM) or between the right and left sides (absolute values for nHM and controls) were, respectively, 10.10° ± 9.58°, 4.15° ± 3.95° and 5.37° ± 4.74° (p = .007). The optimal cutoff point for ANG was 15° at the symptomatic side (sensitivity and specificity of 80.0% and 72.2%, respectively), 10.5° at the non-symptomatic side (sensitivity and specificity of 60.0% and 52.3%), and 3° for the difference between sides (sensitivity and specificity of 90.0% and 79.5%).
Conclusion: Data show that the DQS discriminates HM from nHM and controls.
Introduction
Signs and symptoms in migraine may be much richer than just headache, nausea and vomiting, comprising paroxysmal or permanent impairments of various types. Motor dysfunction may occur as part of aura symptoms in both sporadic and familial hemiplegic migraine (HM) (1,2). The digiti quinti sign (DQS), elicitable clinically through a straightforward manoeuvre, was thought to indicate mild paresis even in the absence of overt weakness (3). The DQS has been detected in three consecutive HM patients examined interictally (4), suggesting that this sign may be of importance in headache practice as well.
To further investigate the actual role of this sign in hemiplegic migraine HM, 10 HM subjects were examined specifically for the presence of DQS and compared with 44 non-hemiplegic migraine (nHM) subjects. Data indicate that the DQS may be discriminative in this context.
Patients and methods
Ten consecutive sporadic HM migraine patients, 44 nHM (migraine with aura [MWA[and migraine without aura [MWoA]) and 44 healthy non-migraineurs examined by the authors in 2008–2009 took part in this investigation. The diagnosis of migraine was established according to the International Headache Society (IHS) 2004 criteria (1). Controls were migraine-free subjects with no previous history of more than infrequent, non-specific, mild headaches. All HM subjects had normal neuroimaging (MR and/or CT scan), except for a patient who presented with a low-grade glioma at the right middle cerebellar peduncle, as described elsewhere (4). All subjects were investigated outside attacks.
Patients underwent a comprehensive neurologic physical examination, with passive and active manoeuvres specifically devoted to motor strength and power, including:
Flexion and extension involving the fingers, hands, elbows, knees, ankles, and hip joints; abduction, adduction, flexion and elevation at the shoulder joints. Deep and superficial reflexes, including a search for the Babinski, Chaddock, Oppenheim, Gordon, Schaefer, Trömner and Hoffmann signs.
Two sensible clinical manoeuvres shown to identify mild, otherwise undetectable arm paresis were performed:
The forearm rolling test (5): Patients were instructed to make a fist with both hands holding the forearms horizontally and overlapping the fists and forearms; and then rotate the hands around each other in front of the torso, with an overlap of approximately 15 cm. The movement was performed for 5–10 seconds in each direction of rotation. Normally, the forearms are expected to rotate around each other symmetrically. A positive sign is present when the movement predominates at the normal side and the healthy forearm tends to orbit the weak limb. The pronator drifting test (6): Subjects held the upper extremities outstretched forwards with palms up for at least 20–30 seconds. Asymmetric pronation more pronounced at the symptomatic side indicates hemiparesis.
While seated, subjects were asked to horizontally extend both arms forwards, parallel to the floor, with palms facing down. No instructions as to how the hands or fingers should be placed were given. Digital pictures of the hands were taken perpendicularly from above (approximately 60 cm), having both hands in a single shot. The pictures were printed out, and straight lines were drawn at the axis of both fourth and fifth fingers bilaterally using a ruler. The angle (ANG) formed by the proximal prolongation of these lines was blindly measured with the help of a protractor. The absolute values and right-to-left differences were saved for posterior analysis. In controls and nHM, right-to-left differences were considered as absolute values. In HM, differences were obtained by subtracting the non-symptomatic side from the contralateral value (SS−nSS).
Statistics
Values were stored in a spreadsheet. InStat for Macintosh 3.0b (GraphPad Software, Inc.) and SPSS 12.0 (SPSS, Chicago, IL, USA) for Windows were used for statistical analyses. The Kruskal-Wallis and the Dunn’s Multiple Comparison tests were used to compare the right-to-left differences between the three groups. The Wilcoxon test was used to compare the ANG within each group. A receiver operating characteristic (ROC) curve analysis was performed to address the ANG discriminating power and estimate the best cutoff point with maximum sensibility and specificity to discriminate HM from nHM. Values are presented as mean ± standard deviation (SD); p values <.05 were accepted as statistically significant.
Results
Demographic data
MWA, migraine with aura; MWoA, migraine without aura; MWA + MWoA, co-existence of both forms.
Descriptive statistics: angles between the fourth and fifth fingers (values in degrees)
SS, symptomatic side; nSS, non-symptomatic side; HM, hemiplegic migraine; nHM: non-hemiplegic migraine; MW, migraine with aura, defined as at least one attack with aura; MWoA, migraine without aura.
The SS−nSS (for HM) and the absolute right-to-left differences (for nHM and controls; respectively 10.10° ± 9.58°, 4.15° ± 3.95° and 5.37° ± 4.74°) varied significantly (p = .007). In particular, HM was significantly different from nHM (p < .01). The right-to-left difference values in nHM and controls were not statistically different. Such values did not differ statistically comparing MWA with MWoA (p = .3513).
The ROC curve analysis on the angle between the fourth and fifth fingers is depicted in Table 3 and Figure 1. The optimal cutoff point for ANG was 15° at the symptomatic side (sensitivity and specificity of 80.0% and 72.2%, respectively); 10.5° at the non-symptomatic side (sensitivity and specificity of 60.0% and 52.3%, respectively), and 3° for the difference between sides (sensitivity and specificity of 90.0% and 79.5%, respecitively).
Receiver operating characteristic (ROC) curve analysis on the angle between the fourth and fifth fingers. nSS, non-symptomatic side. SS, symptomatic side; RL, reference line. Receiver operating characteristic curve analysis on the angle between the fourth and fifth fingers for the discrimination between hemiplegic migraine (N = 10) and non-hemiplegic migraine (N = 44) HM, hemiplegic migraine; nHM, non-hemiplegic migraine; ROC, receiver operating characteristic; AUC, area under curve; SS, symptomatic side; nSS, non-symptomatic side; *Maximum sensitivity and specificity for hemiplegic migraine.
Discussion
The DQS, described by Alter in 1973 (3), is elicited by asking the subject to keep arms and fingers horizontally extended with the palms down. It consists of a clear discernible wider space between the fourth and fifth fingers at the side with “mild hemiparesis”. Alter considered that the DQS could be the only objective sign of weakness. In fact, we recently confirmed that this is one of the discriminative signs present in individuals with mono-hemispheric brain tumours, even in the absence of clear hemiparesis (7). Due to methodological differences, however, the DQS sensitivity in that study cannot be contrasted with the present findings. We also reported this sign in three consecutive HM patients (4). The present data show that the angle between the last two fingers when patients hold both arms extended forwards facing the palms down interictally is significantly wider at the symptomatic side in HM, contrary to nHM and healthy controls.
Since no other signs indicative of hemiparesis were obtained (except for one nHM subject with mild hemiparesis, data not shown) and no demonstrable lesions potentially involved with hemiparesis were detected with neuroimaging, the DQS is not necessarily an indication of a lesion or hemiparesis as originally described. On the contrary, as the present series demonstrates, it may simply indicate the presence of a permanent asymmetric dysfunction indicative of paroxysmal hemiplegia susceptibility at that side, independently from the overt paresis detected during HM attacks. In HM, residual post-ictal abnormalities could persist for fairly long periods of time. Because the time span between the last HM attack and DQS examination varied from two days to more than 20 years, and that no other clinical deficits were detected on examination with the expect of one subject, the DQS should not be regarded as indicative of a motor deficit in all patients. All HM patients but one showed a wider space between the last to fingers at the symptomatic side easily observable by the clinician during examination, in contrast to the observations in nHM and controls. The present measurements confirm our previous clinical impression and supports the importance of this clinical sign in HM.
The reasons for the DQS positivity in HM are not known. The posture suggests relative hypotonia of the third interossei volar (adduction of the little finger) and abductor digiti minimi (opposition of the fifth finger with thumb) muscles. Since the neurological examination is otherwise normal, a peripheral lesion, even purely motor as seen in the handle-bar palsy (8), seems very unlikely. In addition, because migraine is a brain disorder, central nervous system–specific disruption of tonus control should underlie the DQS in HM. Speculatively, in migraine with paroxysmal motor deficits, cortical descending motor signals related to the fifth finger movement could be constantly atypical secondary to the cortical hyperexcitability present in migraine. Abnormalities in the hemisphere prone to paroxysmal hemiparesis could disrupt the opposition of the thumb and fifth finger. In knockin mice carrying the familial hemiplegic migraine P/Q type calcium channel abnormality, a gain of function of excitatory neurotransmission secondary to Ca2+ influx and glutamate release at pyramidal cell synapses has been demonstrated (9). It is possible that anomalous cortical and cerebellar (10) input to the striatum would favour posture changes (11). Interestingly, migraine has been reported to be more frequent in extra-pyramidal disorders (12) such as essential tremor (13), Tourette’s syndrome (14), dystonia (15,16) and Sydenham’s chorea (17). Parkinson’s disease was shown to change the course of migraine (18). It remains to be determined whether the DQS plays a predictive role in HM. The effect of prophylaxis is also unknown. Noteworthy is the fact that the sign remained unchanged in some of our few re-examined patients.
We suggest that this very simple test should be performed routinely in practice, as a positive result indicates possible a hemiplegic deficit in migraineurs at the correspondent side, although a negative finding does not rule this possibility out. It also indicates the need for neuroimaging. Mild, doubtful right-to-left differences, however, must not be overestimated during examination, as normal individuals almost never show a strictly symmetric space between the last two fingers. To approach the issue of DQS expression over time according to the number of attacks and disease severity, a higher number of subjects must be observed for a long time.