Vestibulo-ocular reflexes (VOR) were evaluated with a reactive torque helmet that imposed high-frequency oscillation (2-20 Hz) or step displacements of the head in the horizontal plane. The present paper describes the results in patients with vestibular deficiencies (labyrinthine defective; LD); experimental and analytical techniques and results for normal subjects were described in Part 1 of this paper. The patient groups included: total unilateral LD (related to acoustic neuroma; n = 40); severe (clinically total) bilateral LD (n = 7); bilateral hyporeflexia (n = 14); unilateral hyporeflexia (n = 11); and patients with LD phenomena that had subsided (n = 3). Helmet-induced head steps provided the most specific information. Characteristically, gain was lowered in one direction or both directions after unilateral or bilateral vestibular lesions, respectively; in general, the magnitude of the gain reduction correlated well with the degree of complaints and disability. Surprisingly, delay was systematically prolonged (up to several tens of milliseconds) in all groups of subjects with manifest vestibular pathology. These results suggest that the determination of delay, in addition to gain of the VOR, is feasible and important in the evaluation of vestibular function. The results of head oscillation generally supported the results for steps, but were somewhat less specific. The responses to manually generated head steps roughly agreed with those to helmet-induced steps, but because of the non-uniform acceleration they allowed a less exact analysis of VOR function.

We evaluated changes in the subjectively perceived gravitational vertical as an index of imbalance in the function of the right and left otolith organs. In addition to normal subjects (n = 25), we measured patients with a longstanding (mean 4.5 year +/- 3.2 SD; range 0.5-11.5 years) unilateral vestibular loss after surgery for acoustic neuroma (n = 32), patients with partial unilateral vestibular loss (n = 7) and patients with bilateral vestibular hyporeflexia (n = 8). Normal subjects could accurately align a vertical luminous bar to the gravitational vertical in an otherwise completely dark room (mean setting -0.14 degree +/- 1.11 SD). Patients with left-sided (complete; n = 13) or right-sided (complete; n = 19 and partial; n = 7) unilateral vestibular loss made mean angular settings at 2.55 degrees +/- 1.57 (SD) leftward and 2.22 degrees (+/-1.96 SD) rightward, respectively. These means differed highly significantly from the normal mean (p < 0.00001). In the time interval investigated (0.5-11.5 years) the magnitude of the tilt angle showed no correlation with the time elapsed since the operation. The mean setting by patients with clinically bilateral vestibular loss (-1.17 degrees +/- 1.96 SD; n = 8) did not significantly differ from the control group. The systematic tilts of the subjective vertical in patients with a unilateral vestibular impairment were correlated with their imbalance in canal-ocular reflexes, as reflected by drift during head-oscillation at 2 Hz (r2 = 0.44) and asymmetries in VOR-gain for head-steps (r2 = 0.48-0.67). These correlations were largely determined, by the signs of the asymmetries; correlation between the absolute values of the VOR gain asymmetries and subjective vertical angles proved to be virtually absent. We conclude that the setting of the subjective vertical is a very sensitive tool in detecting a left-right imbalance in otolith function, and that small but significant deviations towards the defective side may persist for many years (probably permanently) after unilateral lesions of the labyrinth or the vestibular nerve.

The response of the human vestibulo-ocular reflex system to a constant angular acceleration is calculated using a second order model with an adaptation term. After first reaching a maximum the peracceleratory response declines. When the stimulus duration is long the decay is mainly governed by the adaptation time constant T_a, which enables to reliably estimate this time constant. In the postacceleratory period of constant velocity there is a reversal in response. The magnitude and the time course of the per- and postacceleratory response are calculated for various values of the cupular time constant T₁, the adaptation time constant T_a, and the stimulus duration, thus enabling their influence to be assessed.

Two middle-aged female patients with active stapedial otosclerosis and sensorineural hearing loss have been treated with aminohydroxypropylidene bisphosphonate (APD). No effect on the Schwartze sign was observed either during or following therapy. No changes were seen in the audiometric data of the patients during the 1st year. Then, within a few weeks, both patients complained of sudden deterioration of the existing hearing losses in both ears, with an alteration of tinnitus. One patient became totally, bilaterally deaf, while the other retained minimal auditory function in the low frequencies in one ear. A sudden bilateral hearing loss in cochlear otosclerosis never develops under normal circumstances and we are convinced that in the cases that we have observed it is due to the treatment with APD.

The literature on cupulometry shows inconsistent data as to (1) the magnitude of the time constant T₁; (2) the iníer-individual variation between subjects; (3) the difference between clockwise and counter-clockwise rotation, and (4) the linearity of the system. In our investigations, the nystagmus was measured after a start or a stop in angular velocity. All determinations were carried out six times for 5 subjects. The time constant T, was calculated from the slope of the decay of the slow-phase eye velocity. Statistical evaluation of the results showed: a test-retest variability which is greater than the inter-individual variability; significant inter-individual differences between subjects; mean values of T, for clockwise and counter-clockwise rotation of 13.2 (SD 1.4) and 12.8 (S.D. 1.7) sec, respectively; a significant difference of T₁ between clockwise and counter-clockwise rotations for individual subjects at the 95% level; no relation between T, and the amplitude of the step in angular velocity, and a linear increase of the maximum slow-phase eye velocity with the amplitude for velocities below 100 °/sec, whereas at higher velocities, saturation occurs.