Changes in visual evoked potentials induced by alterations of serum electrolytes and osmolality in dogs
It has been reported that EEG changes including light sensitivity and electrographic seizure discharges occurred in uremic or normal dogs during hemodialysis against law sodium dialysate. This paper is one of the efforts to elucidate basic mechanisms of dialysis dysequilibrium syndrome, and is concerned
with changes of visual evoked potentials or VEP in dogs induced by artificial alterations of urea, serum electrolytes, and osmolality.
Sixty-seven female adult mongrel dogs were used. Throughout the each experiment, the dogs were immobilized with intravenous drip of succinylcholine chloride and respiration was maintained artificially using 100% oxygen. The EEG, blood pressure, and EKG were monitored. EEGs were recorded in the array suggested by Pampiglione, using subdermal needle electrodes. Blood was analyzed for electrolytes, osmolality, and several other constituents. A single or double layer Kiil artificial kidney was
used for hemodialysis.
Visual evoked potentials were elicited by direct photic stimulation with a Grass PS-2 photostimulator. The responses to photic stimulation and samples of the EEG were stored on Mnemotron magnetic tape. For analysis, fifty or one-hundred
responses were summated from the tape using a CAT 400B, and were written out by a XY plotter.
The experimental conditoins were: (Table Ⅰ) urea loading followed by normal dtalysis or overhydration, normal dialysis, dialysis of uremic dog against normal or low sodium bath, low sodium dialysis with either weight maintained or with osmolality maintained.
VEPs obtained by extracranial leads in curarized dogs were bilaterally symmetric both in amplitude and components in most of the experimental animals. VEPs obtained using scalp-to-scalp linkage between central and occipital electrodes on each aide
were analyzed. The prominent components were an initial major peak at 43±13 msec. followed by a second major peak at 100±37 msec. the first relatively negative at the central as compared with the occipital lead and the second relatively positive (Fig. 1). There were also inconsistent low amplitude components and occasional longer latency components before and after the most prominent peaks respectively.
The normal dogs were dialyzed against a normal bath for three to six hours. There was no significant changes in VEP, EEG, or serum constituents (Fig. 2).
In order to rapidly lower the increased serum urea and osmolality, intravenous administration of 37% urea solution was fellowed by 2-4 hours' dialysis against normal or standard dialysate in 9 dogs. Amplitude of the VEPs, which were decreased
during urea infusion, gradually increased to equal or mildly higher levels in comparison with those of pre-urea infusion in 7 out of 9 dogs. Latency changes persisted during dialysis in most of the animals (Fig. 3).
Forced diuresis and overhydration by infusion of 5-8 liters of 0.45% saline over 2-5 hours following urea loading produced significant increase of VEP amplitudes and electrographic seizure discharges in 2 out of 3 dogs. Marked alterations of serum electrolytes and osmolality were present in all dogs.
Nine dogs were made uremic either by bilateral ureteral ligation or bilateral nephrectomy 3-4 days preceding dialysis: Among 6 dogs dialyzed against the normal bath, one had mild increase of amplitude and latency of the prominent negative
peak, while the others did not show any definite changes of major deflections. However, short latency negative-positive peaks at 13-20 and 21-30 msec. respectively became prominent(Fig. 4). On the other hand. all 3 dogs dialyzed against low sodium dialysate had consistent and gradual increase of amplitude and latency of the prominent peaks, and two had paroxysmal EEG discharges. These
changes are similar if not identical to the findings in normal dogs subjected to low sodium dialysis, which will be described below.
Thirty-one dogs were dialyzed against low sodium dialysate and the body weight was maintained by intravenous 5% glucose in water. Serum sodium was gradually decreased from the mean of 144.7 to 89.7 mEq/L with concomitant significant decrease of chloride and osmolality. Photic driving responses were enhanced in all animals and photoconvulsive responses and spontaneous or light-induced seizure discharges occurred in 16 dogs. The most consistent and unequivocal changes of VEPs were observed in this group. There was a consistent progressive increase of amplitude and latency of VEPs. Sequential VEP changes approximately paralleled the changes of gross photic responses in the EEG (Fig. 5). The prominent positive peaks showed the most significant change.
The maximum chases appeared immediately preceding the seizure discharges(Fig. 6); whereas subsequently little response could be elicited, except for short latency, low amplitude negative or negative-positive deflections. The animals which did not have seizure discharges showed the same successive changes with sudden or gradual decrease of VEP amplitudes after reaching the maximum.
The amplitude changes in VEPs occurred significantly earlier and were more pronounced in dogs which had seizure discharges as compared with those that did not, while there was no significant difference of latency changes between the two groups (Fig. 7).
If saline or a hypertonic agent such as urea-or mannitol was liven during early stages of the VEP changes, recovery from the induced changes could be achieved (Fig. 8).
Dialysis using 2 layer Kill induced maximum VEP changes or paroxysmal EEG discharges earlier than the slower dialysis using 1 layer Kiil.
In 9 dogs, serum osmolality was maintained by intravenous urea or mannitol during low sodium dialysis. Photoconvulsive responses or spontaneous paroxysmal abnormalities occurred in 5. VEP changes were similar to the group where osmolality was not maintained, but were less pronounced.
This study shows that changes of visual evoked potentials in dogs induced by artificial alteration of serum components during hemodialysis correspond closely to the changes in the EEGS. Of the factors studied, hyponatremia induces the most significant alterations of VEP, probably by enhancing neuronal irritability,
however, osmolality changes contribute to the total effect.
The finding that animals which showed an earlier prominent increase of VEP amplitudes were more apt to develop full blown EEG seizure discharges may be of practical value in the management of uremic patients undergoing dialysis, if similar changes can be detected in human subjects.