Auditory brainstem response in hyperbilirubinemic rats: Part II

Auditory Brainstem Response in Hyperbilirubinemic Rats: Part II Masanori Kuriyama, MD, Yukuo Konishi, MD, and Masakatsu Sudo, MD

The effect of an albumin-bilirubin solution on the auditory brainstem response (ABR) of rats was examined and compared with that of rats in which only bilirubin was administered. In albumin-bilirubin-loaded rats, the latencies of waves II and IV were prolonged and the amplitudes decreased, whereas the ABR wave I did not change significantly. The I-IV interpeak latency also was prolonged. These changes were more severe in rats loaded only with bilirubin. Mannitol injection hastened the appearance of ABR changes when only bilirubin was used, but peak levels of ABR changes were almost the same in the mannitoi-injected groups. When the blood-brain barrier is intact, albumin administration is useful in hyperbilirubinemia for the prevention of damage to the central auditory pathways. The reversibility of the ABR changes were demonstrated in all bilirubin-administered groups. Kuriyama M, Konishi Y, Sudo M. Auditory brainstem response in hyperbilirubinemic rats: Part II. Pediatr Neurol 1991;7:375-9.

Introduction The effects of hyperbilirubinemia on the auditory brainstem response (ABR) in infants have been studied [1-6]. We have previously demonstrated that the damage in the auditory pathway is central rather than peripheral, that the ABR changes are reversible when serum bilirubin level is decreased by phototherapy and/or exchange transfusion [5,6], and that central auditory damage occurs in acute hyperbilirubinemia in nonasphyxiated Wistar rats. The reversibility of ABR changes in the last group was demonstrated as well [7]. In neonatal hyperbilirubinemia, hypoproteinemia has been considered to be one of the factors that induce kernicterus. Some authors suggested a relationship between unbound bilirubin (UB) and kernicterus [4,8-10]. No previous animal study has documented an alteration of ABR due to serum protein concentration. We infused rats with bilirubin in an albumin solution and studied the change in ABR following mannitol injection.

From the Departments of Pediatrics; Fukui Medical School; Fukui, Japan.

ABR changes, bilirubin concentration, total serum protein (TP), albumin, and serum osmolality were recorded. These results were compared with previously reported data derived from rats in which albumin was not administered [7]. Methods Male Wistar rats (200-300 gm) were divided into 3 groups of 5 animals each: Group AI received an 11.2% albumin solution; Group A2 received bilirubin in albumin solution (A-B solution); and, Group A3 received an A-B solution following a 20% mannitol injection. Groups AI, A2, and A3 corresponded with previously studied groups B1, B2, and B3 [7], respectively (Table 1). Via a jugular vein catheter, the 20% mannitol (13 ml/kg) was injected through a 0.22 ~tm Millipore ® filter for 1 min. The A-B solution was prepared with an albumin solution and unconjugated bilirubin in a sodium hydroxide solution. This solution was adjusted to pH 8.2 with 0.1 N hydrochloric acid. The bilirubin concentration was 100 mg/dl. Experimental Method (Fig 1). Rats were anesthetized with 35 mg/kg sodium pentobarbital intraperitoneally and both jugular veins were catheterized. After stabilization of the rat's condition, ABR testing was performed, followed by blood sampling. In group A3, 20% mannitol (13 ml/kg) was injected for 1 min. Five minutes after injection, ABR testing was repeated, which was followed by 20 ml/kg of the A-B solution (bilirubin 100 mg/dl, albumin 11.2%) infused over 5 min. ABR testing and blood sampling were performed 5 min after each infusion, which were repeated 4 times. After the administration, ABR testing was performed 4 times every 30 min. In each blood sample, the total serum bilirubin (TB), UB, TP, and serum osmolality were measured

Table 1. Loading substances for each group* Group

Loading Substances

AI

11.2% albumin solution

A2

A-B solution

A3

Mannitol solution + A-B solution

B1

Phosphate buffer solution

B2

Bilirubin solution

B3

Mannitol solution + bilirubin solution

* A-B solution: 11.2% albumin, 100 mg/dl bilirubin. Bilirubin solution: 100 mg/dl bilirubin.

Communications should be addressed to: Dr. Kuriyama; Department of Pediatrics; Fukui Medical School; Matsuoka-cho, Yoshida-gun; Fukui 910-11, Japan. Received January 2, 1991; accepted March 25, 1991.

Kuriyama et al: ABR in Hyperbilirubinemic Rats

375

Experimental Method 20% Mannitol Injection (larnl/kg) for 1minute Bilirubin-Albumin Ioadingfor5minutes Blood Collection ABR Testing

~

~l

~ i ~ ~

i ~

6

3'o

~ i ~

[~ i ~

1

6'0

1

9'0 1riO Time, min

1 150

1 160

Figure 1. Experimental method of group A3. In groups AI and A2, the second ABR testing and blood collection and mannitol injection were not performed. (Semi-Micro-Digital Osmometer, KNAUER, Berlin). TB was measured spectrophotometrically [11], UB by the peroxidase method [12], TP by the biuret method [13], and albumin by the bromcresol green method [14]. All samples were protected from light and were analyzed for hilirubin immediately after sampling. For ABR testing, stainless needle electrodes were placed on the earlobes, vertex, and tail (resistance < 3,000 ohms). Two thousand responses were amplified, filtered (150-3,000 Hz), and averaged (Nicolet CA1,000). The stimuli, presented at 80 dB above the threshold of the normal human population, consisted of clicks presented at 13.5 Hz monaurally. ABR testing was performed in a quiet, sealed box. The room temperature was maintained at 28-30°C. An upward deflection at the vertex was recorded as positive. Positive waves I, lI, and IV were identified and interpeak latencies (IPLs) and latencies were measured using a cursor with a digital read out. The amplitude of each wave was the height from the negative peak just before each positive peak (I,II,IV) to the next positive peak. They were measured using the same methods. The ratios of changes in latencies, IPLs, and amplitudes were calculated. These ratios were compared with our previously reported data [7]. The one-way analysis of variance and the simultaneous multiple comparison (Bonferroni method) [15] was used to test the difference in mean obtained at each time point. Student t test was used to compare these experimental data with previous data. The significance level of P = 0.05 was adopted.

Results In group A2, infusion of A-B solution increased serum TB and UB concentration to 27.6 ± 6.4 mg/dl and 1.17 + 0.4 ~g/dl, respectively. These values increased to 26.8 ± 4.3 mg/dl, and 1.11 + 0.13 ~g/dl in group A3 following A-B infusion. These maximum values were obtained soon after the final loading. There was no significant difference between these 2 groups (Figs 2,3). In group A1, the latencies and amplitudes did not change significantly during the experiment. Wave I in groups A2 and A3 did not change. Maximal and statistically significant prolongation of wave II latencies in group A2 (increased rate: 5.0 _+ 2.2%) were observed 30 min after the final loading. In group A3, the maximum point was reached at the same time as in group A2 (increased rate: 3.2 _+ 2.4%), but this prolongation was not significant (Fig 4A). Wave IV latency in 2 A-B solution-loaded groups were prolonged significantly compared with the preloading values. Maximum values were obtained 30 min after the final loading in both groups A2 and A3. In group A2, significant prolongation of wave IV appeared after the

376

PEDIATRIC NEUROLOGY

Vol. 7 No. 5

second loading, while it was observed only 30 min after the final loading in group A3. Sixty minutes after the final loading, latencies of wave IV in both groups returned to preloading values (Fig 4B). I-IV IPLs, which signify central conduction time. were prolonged significantly in both groups A2 and A3 and maximum values for each were found 30 min and immediately after the final loading, respectively. In both groups, these ABR changes were normalized to preloading values within 60 rain after the final loading (Fig 4C). The amplitudes of waves 11 and IV in 2 A-B solutioninjected groups were significantly decreased except for wave II of group A2 (Fig 5). The amplitudes of wave II revealed a significant and maximal decrease after the final loading in group A3 (78.4 _+ 15.5%) and returned to preloading values 30 min after the final loading. In group A2, a minimum value (74.8 _+ 17.3%) was obtained 30 min after the final loading (Fig 5A). The amplitude of wave IV disclosed maximal changes at the same time as those of wave II in both groups. The amplitudes of wave IV in both groups had not demonstrated significant changes 1 hour after the final loading. Minimum values of wave IV in groups A2 and A3 were 82.4 ± 8.7% and 81.0 ± 11.6% of preloading values, respectively (Fig 5).

-- 1 GroupB2 e---e Gro~A=

3(] .#,~___

0 0 G r o u p B3 0 - - -0 Group A 3

L

I,--

Time, rain

Figure 2. Change of total bilirubin concentrations of groups ,42, ,43, B2, and B3.

llll

osmolality was raised to 321.2 + 9.2 mOsm/kg after the mannitol injection. Comparison Between Groups A and B. Previously, we reported data derived from bilirubin-loaded rats [7]; albumin was not administered (group B). These previous fmdings were illustrated by solid lines in each figure. TB concentrations increased to the same levels in 4 bilirubinloaded groups (A2,A3,B2,B3) and their maximum levels were obtained after the final loading (Fig 2). UB concentrations increased to maximum levels at the same time as in TB. But maximum levels of UB concentration in albumin-administered groups were significantly lower than in groups B2 and B3 (Fig 3); therefore, albumin administration could reduce UB concentration. TP and albumin levels increased in albumin-administered groups and maximum values of TP and albumin were 6.1-6.5 gm/dl and 4.4-4.8 gm/dl, respectively. Conversely, these levels were lower than those in nonalbumin-administered groups and minimum values were 3.6-4.2 gm/dl and 2.1-2.4 gm/dl, respectively. Thus, there were remarkable differences in TP and albumin concentrations between groups A and B. A study of latency changes (Fig 4) demonstrated that maximum prolongation of wave II and IV latencies and I-IV IPLs occurred 30 min after the final loading in groups A2 and B2, but degrees of prolongation were larger in group B2 than in group A2. In mannitol-injected groups (groups A3,B3), the degrees of prolongation of each latency were almost the same, but the peak values appeared later in group A3 than in group B3. Regarding amplitudes, the decrease of waves II and IV were smaller in groups A2 and A3 than in groups B2 and B3 (Fig 5). The minimum values were obtained at the same

H Group B2 (+SD) 11--O Group A2 (+SD)

='2.0 f..

1.0

i1111

O-.-O

Group

93 (+SD)

O==O

Group

A3 (+SD)

:t2.0 e-

.Q I-

P

I

I

30 60 Time, rain

I

I

I

I

90

120

150

180

Figure 3, Change of UB concentrations of (a) groups A2 and B2 and (b) groups A3 and B3 (mean ± S.D.). There were statistically significant differences between albumin-injected groups and nonalbumin-injected groups. Arrows signify bilirubin or A-B injection and the dashed arrow indicates mannitol injection.

The serum osmolality in group A2 did not change during the experiment. In the mannitol-injected group, the serum

Ii

I i

-- ~-GroupB2(+SD) e,--e Group A2(+SD)

e--e

Group B2 (+SD) Group A2 (+SD)

W

Ill

Group B2 (+ SD )

,-~GroupA2~+so,

10 .10

*+

~ 5

J5

~ 5 I

i llll

I

I

I

I

i

o--oGroop~(+so, 0.-=O GroupA 3 ( + S D )

10

O,=-O Group A3 (+SO)

~ 5

•

*

,

I

I

I

I

T O - = O Group A 3 ( + S D )

....* %,,,

ds, _

~0

~ 0

3'0 6'0 9'o 1~o 1~0 1~0

3'0 6'0 9'0 i~0 1~o 1~o

Time, rain A

Zo ao 6'0 9'0 liO 1;o 1~o

Time, rain B

Time, rain C

Figure 4, (A-C) Increased rates of latencies and IPLs (%) by bilirubin or A-B injection: [(latency of each testing~latency of preloading) - 11 x 100. *Statistically significant (P < 0.05) increase of latency from preloading value; +statistically significant (P < 0.05) difference between groups A2 and B2 or between A3 and B3 at each time. Arrows signify bilirubin or A-B injection and the dashed arrow indicates mannitol injection.

Kuriyama et al: ABR in Hyperbilirubinemic Rats

377

Group B2 (--SD) ¢P-,-0GroupA2 (--SD)

100

~ ~ I I

101 -

.

7. 7-GroupB2(-SD) t-,--e GroupA2 (-SD)

~'~.~ ,

~'" Y "

75

* ~ 7!

o.2.-°t"-.

> '~

50

~50 |

'

=

I

I

'

Group B3 (--SD) O,-,,,,,O Group A3 (--SD)

~

10t ~ 7~ >

'~ 50 I

I

30

60

I

I

100

*

'

"

I

I

/ 11/

O---OGroupB3(-SD)

IL V I ~

(W,-~DGroupA3(-SD)

rk.

~75

I

I

90 120 150 180

Time, min

|

iO ~

~

•

i

120 i ~ t ~

Time, min

Figure 5. Decreased rate of amplitude: (amplitude of each testing/amplitude of preloading) x 100. *Statistically significant (P < 0.05) decrease of amplitude from preloading value. Arrows signify bilirubin or A-B injection and the dashed arrows indicate mannitol injection.

time in groups A2 and B2, but were obtained earlier in group B3 than in A3. A significant difference in wave IV amplitude was found only between groups A2 and B2.

Discussion It has been demonstrated that albumin-bound bilirubin does not pass through the intact blood-brain barrier (BBB) and that UB can be transported into brain [8]; therefore, albumin administration to hyperbilirubinemic infants for the purpose of decreasing serum UB is appropriate. In animal experiments, albumin-bound bilirubin has been demonstrated to enter the brain when the BBB was damaged transiently by hyperosmolality and/or hypercarbia [16,17]. Some authors have changed UB concentrations using competitive substances for the bilirubin binding site of albumin and recorded ABR changes [18,19]; however, ABR following albumin administration has not been studied. In this experiment, except for wave II in group A2, the latencies of waves II and IV and 1-IV IPLs were significantly prolonged with an A-B solution infusion and the amplitudes of waves 1I and IV also decreased significantly. These data suggest that the ABR changes in hyperbilirubinemic rats result mainly from damage in the central auditory pathway from A-B solution; however, the latencies and amplitudes in our previous experiment changed more remarkably than in the present experiment. ABR changes in groups A2 and A3 were reversible with a de-

378 PEDIATRICNEUROLOGY Vol.7 No. 5

crease in bilirubin concentration. The same was found in groups B2 and B3. The prolongation of latencies and decrease of amplitudes were more severe in B2. to a statistically significant degree (i.e., the central auditory pathway damage in group B2 was more severe than in group A2). The maximal changes of latencies and amplitudes were observed 30 min after the final loading in both groups. Between these 2 groups, there was no difference except for UB and for serum total protein and albumin concentration induced by albumin administration; however, in groups A1 and B 1. the effect of serum protein concentration on ABR was negative. Therefore, the UB concentration was the primary cause of central auditory pathway damage. There was no statistical difference in peak values of latencies and amplitudes between the mannitol-injected groups. But the maximal changes of latencies and amplitudes of group A3 appeared later than those of group B3. Albumin-bound bilirubin enters the brain through the BBB which was altered by mannitol injection. LIB and albuminbound bilirubin both exhibit neurotoxicity. In our study, both latencies and amplitude derived from the central auditory pathway were changed and these ABR changes were reversible. Hansen and Bratlid reported that bilirubin reduced synaptic transmission and that the reduction was reversible with rat hippocampal slices in vitro [9]. In a part of our studies in vivo was similar to the study by Hansen and Bratlid. Reduced synaptic transmission in the

central auditory pathway is believed to be a cause of these ABR changes. ABR changes in the acute phase of hyperbilirubinemia can be induced by A-B solution injection; however, in the case of an intact BBB, the degree of change of each wave in ABR is more severe in group B2. Conversely, in the case of a damaged BBB, the degrees of change are the same, but the times when peak values are achieved differ. Therefore, albumin administration for hyperbilirubinemia is useful in the reduction of UB and associated central auditory pathway damage when the BBB is intact. When BBB is damaged, albumin administration can delay the appearance of ABR changes. Further study is required to determine whether the ABR change in that instance is due to UB or A-B.

The authors wish to thank Masakazu Saito, MD for his help in statistics and Chiemi Mizumoto for her assistance in the experiment.

References

[1] Wennberg RP, Ahlfors CE, Bickers R, McMurthy CA, Shetter JL. Abnormal auditory brainstem response in a newborn infant with hyperbilirubinemia: Improvement with exchange transfusion. J Pediatr 1982; 100:624-6. [2] Perlman M, Fainmesser P, Sohmer H, Tamari H, Wax Y, Pevsmer B. Auditory nerve-brainstem evoked responses in hyperbilirubinemic neonates. Pediatrics 1983;72:658-64. [3] Nwaesei CG, Van Aerde J, Boyden M, Perlman M. Changes in auditory brainstem responses in hyperbilirubinemic infants before and after exchange transfusion. Pediatrics 1985;74:800-3. [4] Nakamura H, Takada S, Shimabuku R, Matsuo M, Matsuo T, Negishi H. Auditory nerve and brainstem responses in newborn infants with hyperbilirubinemia. Pediatrics 1985;75:703-8.

[5] Kuriyama M, Konishi Y, Mikawa H. The effect of neonatal hyperbilirubinemia on the auditory brainstem response. Brain Dev 1986;8:240-5. [6] Kuriyama M, Tomiwa K, Konishi Y, Mikawa H. Improvement in auditory brainstem response of hyperbilirubinemic infants after exchange transfusion. Pediatr Neurol 1986;2:127-32. [7] Kuriyama M, Konishi Y, Sudo M. Auditory brainstem response in hyperbiliruhinemic rat: Part I. Biol Neonat 1990;58:32-40. [8] Diamond I, Schmid R. Experimental bilirubin encephalopathy. The mode of entry of bilirubin-14C into the central nervous system. J Clin Invest 1966;45:678-89. [9] Hansen TWR, Bratlid D. Bilirubin and brain toxicity. Acta Paediatr Scand 1986;75:513-22. [10] Cashore WJ, Oh W. Unbound bilirubin and kernicterus in lowbirthweight infants. Pediatrics 1986;69:481-5. [11] Jackson SH, Hemandez AH. A new "bilirubinometer" and its use in estimating total and conjugated bilirubin in serum. Clin Chem 1970; 16:462-5. [12] Jacobsen J, Wennberg RP. Determination of unbound bilirubin in serum of newborns. Clin Chem 1974;20:783-9. [13] Reiner M, Cheung HU Fibrinogen. In: Seligson D, ed. Standard methods of clinical chemistry. New York: Academic Press, 1961; 3:114. [14] Doumas BT, Watson WA, Biggs HC. Albumin standards and the measurement of serum albumin with bromocresol green. Clin Chim Acta 1971;31:87-96. [15] SAS Users Guide. Statistics, 5th ed. Cary: SAS Institute, 1985;470-6. [16] Bratlid D, Cashore W J, Oh W. Effect of serum hyperosmolality on opening of blood-brain barrier for bilirubin in rat brain. Pediatrics 1983;71:909-12. [17] Bratlid D, Cashore WJ, Oh W. Effect of acidosis on bilirubin deposition in rat brain. Pediatrics 1984;73:431-4. [18] Shapiro SM. Acute brainstem auditory evoked potential abnormalities in jaundiced Gunn rats given sulfonamide. Pediatr Res 1988; 23:306-10. [191 Karplus M, Lee C, Cashore WJ, Oh W. The effect of brain bilirubin deposition on auditory brain stem evoked response in rats. Early Hum Dev 1988;16:185-94.

Kuriyama et al: ABR in Hyperbilirubinemic Rats

379