- Email: [email protected]
Effects of cochlear ablation on local cerebral glucose utilization in fetal sheep
Thurnau, Kemp, and Jarvis
3. Pritehard JA. The use 01' the magnesium ion in the management of eclamptogenic toxemias. Surg Gyneeol Obstet 1955;100:13I. 4. Oppelt WW. Magnesium exchange between blood and cerebrospinal fluid. Am J Physiol 1963;205:959-62. 5. Somjen G. Effeets of magnesium and calcium on neurons in the eentral nervous system. Brain Res 1968;9: 161-4. 6. Borges LF. Effeet of magnesium on epileptie foei. Epilepsia 1978;19:81-9I. 7. Berkowitz RL. The management of hypertensive erisis during pregnaney. In: Critieal eare of the obstetric patient. New York: Churehill Livingstone, Inc, 1983:322.
December 1987 Am J Obstet Gynecol
8. Berkowitz RL. Neurologie emergeneies during pregnaney. In: Critieal eare of the obstetrie patient. New York: Churchill Livingstone, Inc, 1983:372. 9. PritehardJA, MaeDonald PC, Gant NF. Hypertensive disorders in pregnaney. In: Williams' Obstetries. 17th ed. Norwalk, Corm.: Appleton-Century-Crofts, 1985:525. 10. Connerty HV, Lau HSC, Briggs AR. Speetrophotometrie determination of magnesium by use of methylthymol blue. Clin Chem 1971; 17:661-2. 11. Chesley LC, Tepper I. Plasma levels of magnesium attained in MgSO, therapy for preeclampsia and eclampsia. Surg Clin North Am I957;April:353-67.
Effects of cochlear ablation on local cerebral glucose utilization in fetal sheep Rohert M. Abrams, Ph.D., Alastair A. Hutehison, M.B., Ch.B., Michael J. McTiernan, B.S., and Gerald E. Merwin, M.D. Gainesville, Florida Local cerebral glucose utilization was measured by the ['4C]-deoxyglucose method in five near-term fetal sheep in whom bilateral ablation of the cochleae had been accomplished aseptically 5 to 8 days earlier. The tympanie membrane and ossicles were removed and all turns of each cochlea were unroofed with destruction carried to the modiolus. Mean local cerebral glucose utilization of 33 of 34 gray matter structures and four of four white matter structures in operated animals were significantly lower (p < 0.05) than that in unoperated control fetuses. The depression in local cerebral glucose utilization was greatest (p < 0.002) In brain stern auditory nuclei, in which the mean rate of glucose utilization was approximately 25% of the levels in unoperated fetuses. The pattern of glucose utilization in these structures was clearly altered, with areversal of the normal distribution in density of the inferior colliculus. Tonotopic bands of high local cerebral glucose utilization frequently seen in autoradiographs of inferior colliculus in unoperated fetuses were not observed in operated fetuses. These resutts show that the glucose utilization of the brain, and by implication the normal growth and maturation of the brain, depends on an intact auditory system during prenatal Iife. (AM J OSSTET GVNECOL 1987;157:1438-42.)
Key words: Fetal sheep, local cerebral glucose utilization, brain stern, auditory nuclei, cochlear ablation Environmental sounds are known to penetrate the abdominal wall and uterus of pregnant women. Specific sounds can induce fetal movementsI." and cardioacceleration," reduce fetal breathing movernents,' and evoke fetal electroencephalographic responses." There have been speculations about the importance of sound in prenatal nervous system maturation,? irnprinting," and early neonatal speech recognition." Fetal cochlear From the Departments of Obstetrics and Gynecology, Pediatrics, and Otolaryngology, College of Mediane, University of Florida. This study was aided by National Institutes of Health Grant HD-20084 and Reproductive Hazards in the Workplace Research Grant No. 15-76 from the March of Dimes Birth Defects Foundation. Sponsored by the Society[or Gynecologic1nvestigation. Reprint requests:Robert M. Abrams, Ph.D., Department ofObstetrics & Gynecology, BoxJ-294,JHMHC, Gainesville, FL 326/0.
1438
damage may result if noise is excessive. 10 Thus prenatal sound may have far-reaching implications for the developing child. The constraints placed on human experimentation force us to turn to appropriate animals to further our understanding in this area of developmentalbioacoustics. As a model for human characteristics, the pregnant sheep has two advantages over common Iaboratory animals: (l) the physical dirnensions (and thus the acoustic properties) of the ewe and her products of conception, including fetal fluids, are similar to those in human pregnancy, and (2) the cochlea and central auditory connections are anatomically completed and presumed functional by the third trimester of gestation in both species.": 12 An important first step in our understanding of this model, the characterization of the fetal sound
Fetal cersoral glucose utilization
Volume 157 Number 6
1439
Table I. Data from experiments to measu re the effect o f bilateral coc h lear destruction on local cerebra l glucose u tilization in feta l sheep Cochlear destru ction
126 Gestational age (days) T ime after operation (days) Fetal weight (kg) Glucose, fetus (mg/mi) Temperature of ewe (0 Cl Fetal arterial pH* Fetal Paco 2 (111m Hg)* Fetal Pao. pressure (mm Hg)* Fetal hernatocri t Rapid eye movement sleept (%) Non-rapid eye rnovement sleept (%)
o
142 3
I
122 141 3
I W-82 I 135 8
Y,8 1
134 5
Contr ol
I
0-9
137 5
0 0~28
135 4
1
0' 17
I
135 4
214 138 7
I
0·14
137 8
I
0-2
135 5
3.7 0.27 39.4 7.41 1 7.43 1 63.5 5903 10.9
3.8 0.56 39.4 7.377 7.380 47.2 48 .8 15.2
3.3 0.27 39.4 7.370 7.364 60.3 57.3 23. 1
2.4 0.17 39.7 7.384 7.384 43.5 44.6 16.3
0.14
3.5 0.2 1
0. 14
4.4 0.27
3.5 0.16
2.5 0. 13
7.393 7.472 33.2 40. 2 16.6
7.380 7.368 42.8 41.7 20.6
7.376 7.349 50.6 54.6 17.8
7.361 7.377 36.4 34.2 16.4
7.370 7.364 37.2 44.4 21.3
7.316 7.330 40 .9 40.5 17.5
16.2 39
15.0 40
28.7 37
18.4 34 43
18.7 30 29
19.5 37 77
18.8 32 16
20.7 28
22.2 35 31
21.0 43 34
53
55
18
54
57
69
56
*Control and + 20 minute s. tWeighted for fraction of free deoxyglucose concentration in the tissue integrated with rcspect
enviroriment within the pregnant ewe, has a lready bee;] made.": " Thus a 2 to 5 d B enhancement within the ewe fo r fr eq uencies <0.25 kH z was fou nd ; fr eq uencies > 0.25 kH z were progressively atten uated up to 20 dB at 4.0 kH z; externa l speech and music were easily recognizable, a lbeit rnuffled, once the noise fioor (50 dB spectrum le vel ) was ex ceeded . While these res u lts p roved that the fe ta l so und environ ment is ric her a nd mo re varied than he retofore be lieved," the specific effects of so u nd on the fet us are less accessible to q uan tita tive assessmen t, and are therefore sometimes inconsistent.' : 7 One consistent finding, however, is the high rate of glucose utilization along the central auditory pathways of the fetus in u tero.!" T he origins of th is high metabolic rate for glucose, wh ich is also p resent in an imals after birth,"?" has no t been establish ed . T he p re se n t study was unde r ta ke n to de termine the im portance of the cochlea in sustaining th is h ig h rate of loca l cerebral glucose utilization in the fetal sheep. Methods
Ten g ra d e Western ewes with kno wn dates of insemination were fasted for 24 hours, anesthetized with ha lo th a ne by mask, intubated, and ca rried throughout an aseptic surgery with halothane and oxygen. The fetal head was located through the uterine wall and moved to a re latively avascular area on the greater curvature, With the head held firmly by one hand, the uterin e serosa, but not u nderlying tissues, was incised over 4 in ches. Small openings in the myometrium and
tu
time.
membranes were made, and these tissues were then stretched to the extent of the serosal incision . This procedure resu lted in little bleed ing. The feta l head and one fo re leg were de livered. Polyviny l catheters were placed in th e fetal brachiocephalic artery and superior vena cava through incisions in the axillary artery and a superficial shou lder vein . In five experimental animals, we used an operating mic roscope to re move the tym pa nic me m brane and ossieles , wo r king thro u gh a posta uricular approach. Oval and round windows were identified. With the aid of round cutting burs, all turns of the cochlea were unroofed, with destruction carried to rhe mod iolus . After cornple tion of three ablation experirnents, it was brought to o u r attention that fe ta l cerebral oxygen co ns umption rates and local cerebral glucose uti liza tio n rares were co rrelated with th e be ha vioral state of the fetus at the time of the experiment.": n T hus in five co ntrol animals and two of five experimental animals, stainless steel screws (0-80 ; l/8 in) were placed bilate ra lly in parietal bone I to 2 cm on ei ther side of the midline cranial suture. Additional screws (0-80; 1/4 in) were placed through pred ri lled holes in the roof of the orbit. Stra nd ed tin ned co p per lea d wires, previo usly silver soldered to the screws, were led out through the scalp and skin incisions, Wire eleetrodes (Cooner Wire Cornpany, Chatsworth, CaIif.) placed on a neck strap muscle completed the instrumentation required for asscssmeni: of fetal behavioral state, All fetal skin incisions were cIosed. Ampicillin, 0.5 grn, was introd uced in the arnniotic flu id. The feta l
1440
Abrams et aI.
December 1987 Am.J Obstet GynecoI
Table 11. Local cerebral glucose utilization (!LmoIlIOO gm/min) Cochlear destruction
126 Motor, sensory cortex Visual cortex Auditory cortex Cingulate gyrus Thalamus Medial geniculate Lateral geniculate Hypothalamus Mamillary body Hippocampus Amygdala Septum Caudate Putamen Subthalamic nuclei Vestibular nuclei Cochlear nuclei Superior olives Inferior olives Inferior colliculus Superior colIiculus (st. super.) Superior colliculus (st. profund.) Habenula Preoptic area Spinal cord, ventral horns Reticular formation Lateral lemniscus Periaqueductal gray Nuclei of oculomotor nerves Red nuclei Vermis Cerebellar cortex Cerebellar nuclei Flococulus Spinal cord, lateral columns Internal capsule Corona radiata Optic chiasm
I
122
I W-B2
j
Y-Bl
I 0-9
0-28
30 24 22 33 35 35 45 22 57 31 30 31 30 27 44 40 29 30 46 33 43
35 30 27 36 45 53 57 30 59 31 32 45 25 27 69 61 50 68 52 82 38
I
Controls
0-14
I 214)
0-2
p*
29 29 22 41 42 62 55 20 51 50 30 32 31 35 56 76 57 86 75 105 56
54 46 42 66 68 103 98 37 111 44 64 74 47 48 124 115 92 105 122 178 77
20 21 17 24 29 41 35 18 41 26 29 20 22 24 34 41 38 44 39 73 30
48 51 47 70 69 96 78 39 112 60 57 54 50 52 95 100 72 100 91 126 101
0.029 0.025 0.044 0.034 0.009 0.014 0.025 0.021 0.050 0.020 0.014 0.027 0.036 0.015 0.022 0.007 0.002 0.0007 0.016 0.002 0.058
0-17
I
20 23 17 27 20 24 37 14 52 27 15 19 18 18 25 20 13 16 24 24 28
18 20 20 22 18 19 28 17 31 18 12 14 14 18 19 21 18 19 23 28 30
15 7 18 10 10 9 8 10 21 17 8 10 15 21 15
15 15 13 21 21 21 35 13 39 21 15 17 20 17 38 38 19 19 35 34 34
16
16
10
19
25
29
30
45
18
65
0.045
22 14 13 16 18 13 22 19 19
18 14 17 14 19 16 18 18 20 12 26 19 11
10 9 8 10 10 7 13 12 10 8 16 14 5 7 3 3
2'.) 13 21 17 20 12 29 30 16 14 32 22
44 20 29 27 27 21 34 40 30 21 52 40 15 24 9 12
46 27 39 35 50 24 34 44 36 25 50 35 19 20 11 14
48 22 48 31 66 18 49
85 38
25 19 27 17 39 15 35 25 23 16 36 25 16 14 12 16
97 44 43 47 71 35 72 70 50 38 82 62 25 30 13 28
0.036 0.014 0.006 0.018 0.001 0.041 0.012 0.039 0.010 0.017 0.024 0.026 0.006 0.046 0.010 0.014
II
24 20 8 14 9 10
IO
8 14
9 10 7 12 10
10
12 7 11
51 22 61 48 29 23 22 18
48 89 35 68 64 63 36 100 56 34 16 25
*Destruction versus controls.
head and foreleg were repositioned in the uterus and the uterus was c1osed. A second dose of 0.5 gm ampicillin was placed in the peritoneal cavity of the ewe be fore the anterior abdominal wall was c1osed. Leads were tunneled subcutaneously and exited on the maternal flank, where they were stored in a pouch sutured to the skin. The skin was closed with clips. Finally, maternal femoral arterial and venous catheters were placed. Anesthetic levels were reduced during the c1osing procedure so that the ewe stood within 5 minutes of extubation. Ewes were placed in carts daily to flush catheters. A cart containing a companion ewe was positioned nearby. On the day of the local cerebral glucose utilization experiment (3 to 8 days after surgery), the fetal behavioral state was assessed during a 2-hour control period (Table I).
Details of the method used to measure cerebral glucose utilization have been published." Briefly, 100 !LCi [I4C]-2-deoxyglucose (American Radiolabeled Chernicals, St. Louis, Mo.) per kilogram of estimated fetal weight were given over 30 seconds. The histories of the arterial plasma levels of [I'C]-2-deoxyglucose and glucose were determined by frequent blood sampling for 30 to 40 minutes. Fetuses and ewes were given intravenous sodium pentobarbital solution followed immediately by intravenous KCI to induce cardiac arrest. The fetus was delivered and weighed and the brain was removed and frozen for later sectioning and quantitative autoradiographic examination.." Local cerebral glucose utilization was caIculated for 38 cerebral structures. Means and SEM of values for experimental and control groups were caIculated across structures. Comparisons between groups were by
Fetal cerebral glucose utilization
Volurne 157
1441
Nurnber 6
paired Student's t test with significance taken as
p < 0.05. The average rate of glucose utilization for the entire brain was not measured. Results
Gross inspection ofthe fetal temporal bones removed after the study revealed that the cochleae were destroyed bilaterally to the level of the modiolus. Table I summarizes data for both experimental and control fetuses. Fetal weights, pH, Paco., Pao., and glucose levels were within the range of those characteristic of our Hock." Local cerebral glucose utilization in experimental fetuses was uniformly depressed, with 33 of 34 gray matter structures and foul' of foul' white matter structures differing significantly (p < 0.05) from respective structures in control fetuses (Table 11). Rednerion in local cerebral glucose utilization in the four brain stem auditory nuelei (cochlear nuelei, superior olives, lateral lemniscus, and inferior colliculus) reached even higher levels of significance (p < 0.002). The pattern of glucose utilization in the inferior colliculus as represented pictorially in autoradiographs was elearly altered. with the appearance of a very pale central nueleus and a denser periphery. Behavioral state cyeling during the 30 to 40 minutes required to measure local cerebral glucose utilization appeared to be normal in two operated fetuses, and the quality of the electrocorticogram, electrooculogram, and neck eIectromyographic tracings was good. Comment
The [IIGJ-2-deoxyglucose method of Sokoloff et al." offers a powerful method of metabolic encephalography. As seen consistently on autoradiographs, the brain structures with the highest concentrations of [IIGJ (and thus with the highest glucose utilization rares) are the cochlear nuclei, superior olives, lateral lernnisci, and inferior colliculi-all structures in the central auditory pathway. Audiographs of the inferior colliculi are characterized by differing densities, with the darkest area occupying part 01' all of the central nueleus. This dark region, whose metabolic rate normally exceeds that of all other loci in the brain, commonly takes on the shape of an ovoid 01' band. Other bands, presumably representative of a tonotopic, spacial organization of frequency coding,"'?" are occasionally seen. These autoradiographic features have been observed by us in all normal fetal sheep. We inferred that the high rares of glucose utilization in central auditory structures could be at least partly relared to environmental sound that is continuously present.":" The sound originates outside the ewe as well as within the ewe, through intestinal, cardiovascular, and respiratory activities and through vocalization. Intra-amniotic pressure changes resulting from pulsation of the umbilical artery could be an additional source of sound. Because
of the well-known link between local functional activity and local metabolie rate in the brain," ambient noise intense enough to evoke fetal cardioacceleration and movement would be expected to activate the metabolism along theauditory pathways. The specificity of this metabolic response, depending as it must on the rnaturation of the central auditory pathways and connections, would be more advanced in sheep fetuses than in term fetuses of less precocious species. Bilateral destruction of the fetal cochleae resulted in adramatic reduction in glucose utilization in the central auditory nuclei. This phenomenon has been weil documented postnatally in animals.": 24. 25 We did not explore the effects of unilateral destruction, but this procedure has been used to judge the levels and extent of crossing of afferent auditory fibers." A decrease in glucose utilization in nonauditory structures in operated fetuses was not anticipated, and has not been mentioned previously in descriptions of cochlear ablation in animals postnataIly. Virtually all gray and white matter structures were affected to a significant extent. In earlier experiments in fetal sheep, glucose utilization was found to be relatively high in brain stem auditory structures at the beginning of the third trirnester." Thus central auditory projection to nonauditory structures, including the reticular formation, would appeal' early and could dominate the representation of other sensory modalities that have yet to mature to the same degree. Continuing with this reasoning, removal of the cochleae could result in a diffuse inactivation of the central nervous systern, which would be reflected in its lowered metabolie rate. The reduction could be linked to the elimination of any transduction of acoustic energy by the destroyed cochleae as weIl as to the removal of spontaneous, resting neural discharge of auditory neurons." The possibility must also be considered that the overall reduction in cerebral glucose utilization may have resulted from the generalized trauma associated with the destructive lesioning. However, the technique used in bilateral cochlear destruction was entirely peripheral, without violation of dura 01' intracranial vasculature 01' contents. While microcontamination of cerebrospinal fluid via a patent cochlear aqueduct is possible, the effects of such blood contarnination must be considered minimal. Whatever the cause of the lowered cerebral metabolie rate, it may have some consequences for normal maturation of the fetal brain. We thank Drs. K. J. Gerhardt and Jack Johnson for helpful comments on the manuscript, the Lesly Martin and Sue Laing for manuscript preparation.
REFERENCES J. Gagnon R, Runse C, Carmiehael L, Fellows F, Patriek J. Effeets of vibratory aeoustie stimulation on human fetal
1442 Abrams et al.
2. 3. 4. 5. 6. 7. 8. 9. 10.
11. 12. 13. 14. 15.
breat hing a nd g ross feta l bod y movc rnents near term. AM j O BSTET GYNECOL 1986 ; 155: 1227-30. Oelman SR, Wood S, Sp elIacy WM , Abrams RM. Fetal movernents in respon se to sou nd stimulation . AM j OBSTET GYNECOL 1982; 143:484-5. Murphy KP, Smyth CN . Respo nse of foetus to a uditory stimulation . Lan cet 1962;5:972-3. Gri mwade j C , Walke r DW, Wood C . Sensory stirnu lation of th e h uman feru s. Aust j Ment Res 197 0 ;2:63-4. Serafini P, Lind say MBj , Nagey DA, et al. Anteparturn fe tal heart rate resp onse to sou nd stim ulation : th e acoustic sti mula tio n test. AMj OBSTET GYN ECOL 1984 ; 148:4 1-5. Sa kabe N, Ara ya ma T , Suzuki T. Human feta l evok ed response to aco usti c stim ula tion . Acta Otolary ngol (su p pl) 1969 ;252:29-36. Goo dli n RC, Schmidt W. H uman fetal a ro usal levels as ind icated by heart rate reco rdin gs. AMj OBSTET GYNECOL 1972;114 :613-21. Salk L. Mothers' heartbeat as a n imprinting stimulus. T rans NY Acad Sei 1962 ;24:753-63. Kolat a G. Studying learnin g in th e womb. Science 1984 ; 225 :302 -3. Prenatal e ffects of ex pos ure to high-l evel noise . Repo rt o f Wo rk ing Grou p 85, Co mmi ttee o n Hea ring, Bioacoustics, a nd Biorn echanics, Assernb ly of Behavioral an d Social Sciences , National Resea rch Co u ncil. Washi ng to n, DC: National Academy Press, 198 2. Br edb erg G. The h u ma n cochlea d uring de velo pme n t a nd ag ing. j La ryn gol Ot o l 1967 ;8 1:739-58. Be rnhard GG, Kaise r IH , Kalom odi n GM . O n the d evelo p mem ofcortical activity in feta l sheep. Acta Ph ysio l Sca nd 1959;4 7:333-49 . Ar mitage SE, Bald win BA, Vince MA. T he fe tal sou nd env iro nment of sheep. Scien ce 1980;208 : I 173-4. Vin ce MA, Billing AE , Bald win BA, T oner j H , Weller C. Mat emal voca Iizations a nd othe r so unds in th e fe tal Iarnb 's sou nd enviro n rne nt . Earl y H uman Dev 1985; I I: 179-90. Gerhard t KH, Abrams RM. T he sound envi ron me nt of the fet al sheep in ut ero . Fed Proc ' 1987;46:358,
Dece m ber 1987 Am J Obs tet Gynecol
16. Abrams RM, 110 M, Fris inger JE, Pa tlak CS, Pettig re w KD, Ken ned y C. Local cer ebral gluc ose utilizat io n in fe tal a nd neona tal sheep. Am ] Ph ysiol 1984 ;246 :R608 - 18. 17. Sok o loff L. Relat io n between ph ysio log ical fu nc tio n and energy me tabolism in the ce nt ra l nervous system. ] Ne urochern 19 77 ;29 ;13-26. 18. Webs te r WR, Servie re ] , Batini C , LaPl ante S. Autoradiogra phie demonstration with 2-[liC]deoxyglucose of frequ ency selec tivity in the a uditory syste m of ca ts under co nditio ns of fu nctional activity. Ne urosei Lett 1978; 10 :43-8. 19. Ryan AF, Woolf NK , Sharp FR. T onotopic orga nization in the centra l aud ito r y path way of the mon golian gerbil: A 2-d eox yglucose study. j Comp Neurol 1982 ;207 : 369-80. 20 . Woo lf NK, Sharp FR, Davidson T M, Ryan AF. Cochlear and middle ear effects o n metab olism in the central auditory pathway during sile nce: a 2-deoxyglucose study. Brain Res 1983;274:1I9-27. 21 . Rich ardson BS, Patrick ] E, Abd uljabbar H . CerebraI ox id ati ve metabolism in th e fe tal lamb : Relationship to electroco rtical state. AM] O BSTET GYNECOL 1985; 153:42 6-3 I . 22 . Abram s R, Hu tch ison AA,]ay TM , Kenned y C . Cerebr a! glucose ut ilizat ion incr eased d u rin g 'ra pid eye movement slee p. Ann Ne u ro l 1986;20:40 1. 23 . Sok oloff" L, Reivich M, Ke nn ed y C, et al. The[" C) deox yglucos e meth od for the measurement of Ioeal cerebral gl ucose utilizatio n : th eory, proced u re, a nd no rmal values in the consc ious and anesthetized albino rat. J Ne uroc hern 19 77 ;28 :89 7-91 6. 24. Sasa ki CT, Kaver .IS, Bab itz L. Diffe re nti al [" C] 2-deoxyglucose u ptake after deafferentation of the mam malian au ditory pathw a y-a model for e xa mining tinniWs . Brain Res 1980 ;194 :5 11-6. 25. T a nag uchi 1. Changes in metabolic ac tivity in the in feri o r co llicu lus fo llo wing removal of au di tory inpu t. Biomed Res 1980; 1:5 10-6. '
3. Pritehard JA. The use 01' the magnesium ion in the management of eclamptogenic toxemias. Surg Gyneeol Obstet 1955;100:13I. 4. Oppelt WW. Magnesium exchange between blood and cerebrospinal fluid. Am J Physiol 1963;205:959-62. 5. Somjen G. Effeets of magnesium and calcium on neurons in the eentral nervous system. Brain Res 1968;9: 161-4. 6. Borges LF. Effeet of magnesium on epileptie foei. Epilepsia 1978;19:81-9I. 7. Berkowitz RL. The management of hypertensive erisis during pregnaney. In: Critieal eare of the obstetric patient. New York: Churehill Livingstone, Inc, 1983:322.
December 1987 Am J Obstet Gynecol
8. Berkowitz RL. Neurologie emergeneies during pregnaney. In: Critieal eare of the obstetrie patient. New York: Churchill Livingstone, Inc, 1983:372. 9. PritehardJA, MaeDonald PC, Gant NF. Hypertensive disorders in pregnaney. In: Williams' Obstetries. 17th ed. Norwalk, Corm.: Appleton-Century-Crofts, 1985:525. 10. Connerty HV, Lau HSC, Briggs AR. Speetrophotometrie determination of magnesium by use of methylthymol blue. Clin Chem 1971; 17:661-2. 11. Chesley LC, Tepper I. Plasma levels of magnesium attained in MgSO, therapy for preeclampsia and eclampsia. Surg Clin North Am I957;April:353-67.
Effects of cochlear ablation on local cerebral glucose utilization in fetal sheep Rohert M. Abrams, Ph.D., Alastair A. Hutehison, M.B., Ch.B., Michael J. McTiernan, B.S., and Gerald E. Merwin, M.D. Gainesville, Florida Local cerebral glucose utilization was measured by the ['4C]-deoxyglucose method in five near-term fetal sheep in whom bilateral ablation of the cochleae had been accomplished aseptically 5 to 8 days earlier. The tympanie membrane and ossicles were removed and all turns of each cochlea were unroofed with destruction carried to the modiolus. Mean local cerebral glucose utilization of 33 of 34 gray matter structures and four of four white matter structures in operated animals were significantly lower (p < 0.05) than that in unoperated control fetuses. The depression in local cerebral glucose utilization was greatest (p < 0.002) In brain stern auditory nuclei, in which the mean rate of glucose utilization was approximately 25% of the levels in unoperated fetuses. The pattern of glucose utilization in these structures was clearly altered, with areversal of the normal distribution in density of the inferior colliculus. Tonotopic bands of high local cerebral glucose utilization frequently seen in autoradiographs of inferior colliculus in unoperated fetuses were not observed in operated fetuses. These resutts show that the glucose utilization of the brain, and by implication the normal growth and maturation of the brain, depends on an intact auditory system during prenatal Iife. (AM J OSSTET GVNECOL 1987;157:1438-42.)
Key words: Fetal sheep, local cerebral glucose utilization, brain stern, auditory nuclei, cochlear ablation Environmental sounds are known to penetrate the abdominal wall and uterus of pregnant women. Specific sounds can induce fetal movementsI." and cardioacceleration," reduce fetal breathing movernents,' and evoke fetal electroencephalographic responses." There have been speculations about the importance of sound in prenatal nervous system maturation,? irnprinting," and early neonatal speech recognition." Fetal cochlear From the Departments of Obstetrics and Gynecology, Pediatrics, and Otolaryngology, College of Mediane, University of Florida. This study was aided by National Institutes of Health Grant HD-20084 and Reproductive Hazards in the Workplace Research Grant No. 15-76 from the March of Dimes Birth Defects Foundation. Sponsored by the Society[or Gynecologic1nvestigation. Reprint requests:Robert M. Abrams, Ph.D., Department ofObstetrics & Gynecology, BoxJ-294,JHMHC, Gainesville, FL 326/0.
1438
damage may result if noise is excessive. 10 Thus prenatal sound may have far-reaching implications for the developing child. The constraints placed on human experimentation force us to turn to appropriate animals to further our understanding in this area of developmentalbioacoustics. As a model for human characteristics, the pregnant sheep has two advantages over common Iaboratory animals: (l) the physical dirnensions (and thus the acoustic properties) of the ewe and her products of conception, including fetal fluids, are similar to those in human pregnancy, and (2) the cochlea and central auditory connections are anatomically completed and presumed functional by the third trimester of gestation in both species.": 12 An important first step in our understanding of this model, the characterization of the fetal sound
Fetal cersoral glucose utilization
Volume 157 Number 6
1439
Table I. Data from experiments to measu re the effect o f bilateral coc h lear destruction on local cerebra l glucose u tilization in feta l sheep Cochlear destru ction
126 Gestational age (days) T ime after operation (days) Fetal weight (kg) Glucose, fetus (mg/mi) Temperature of ewe (0 Cl Fetal arterial pH* Fetal Paco 2 (111m Hg)* Fetal Pao. pressure (mm Hg)* Fetal hernatocri t Rapid eye movement sleept (%) Non-rapid eye rnovement sleept (%)
o
142 3
I
122 141 3
I W-82 I 135 8
Y,8 1
134 5
Contr ol
I
0-9
137 5
0 0~28
135 4
1
0' 17
I
135 4
214 138 7
I
0·14
137 8
I
0-2
135 5
3.7 0.27 39.4 7.41 1 7.43 1 63.5 5903 10.9
3.8 0.56 39.4 7.377 7.380 47.2 48 .8 15.2
3.3 0.27 39.4 7.370 7.364 60.3 57.3 23. 1
2.4 0.17 39.7 7.384 7.384 43.5 44.6 16.3
0.14
3.5 0.2 1
0. 14
4.4 0.27
3.5 0.16
2.5 0. 13
7.393 7.472 33.2 40. 2 16.6
7.380 7.368 42.8 41.7 20.6
7.376 7.349 50.6 54.6 17.8
7.361 7.377 36.4 34.2 16.4
7.370 7.364 37.2 44.4 21.3
7.316 7.330 40 .9 40.5 17.5
16.2 39
15.0 40
28.7 37
18.4 34 43
18.7 30 29
19.5 37 77
18.8 32 16
20.7 28
22.2 35 31
21.0 43 34
53
55
18
54
57
69
56
*Control and + 20 minute s. tWeighted for fraction of free deoxyglucose concentration in the tissue integrated with rcspect
enviroriment within the pregnant ewe, has a lready bee;] made.": " Thus a 2 to 5 d B enhancement within the ewe fo r fr eq uencies <0.25 kH z was fou nd ; fr eq uencies > 0.25 kH z were progressively atten uated up to 20 dB at 4.0 kH z; externa l speech and music were easily recognizable, a lbeit rnuffled, once the noise fioor (50 dB spectrum le vel ) was ex ceeded . While these res u lts p roved that the fe ta l so und environ ment is ric her a nd mo re varied than he retofore be lieved," the specific effects of so u nd on the fet us are less accessible to q uan tita tive assessmen t, and are therefore sometimes inconsistent.' : 7 One consistent finding, however, is the high rate of glucose utilization along the central auditory pathways of the fetus in u tero.!" T he origins of th is high metabolic rate for glucose, wh ich is also p resent in an imals after birth,"?" has no t been establish ed . T he p re se n t study was unde r ta ke n to de termine the im portance of the cochlea in sustaining th is h ig h rate of loca l cerebral glucose utilization in the fetal sheep. Methods
Ten g ra d e Western ewes with kno wn dates of insemination were fasted for 24 hours, anesthetized with ha lo th a ne by mask, intubated, and ca rried throughout an aseptic surgery with halothane and oxygen. The fetal head was located through the uterine wall and moved to a re latively avascular area on the greater curvature, With the head held firmly by one hand, the uterin e serosa, but not u nderlying tissues, was incised over 4 in ches. Small openings in the myometrium and
tu
time.
membranes were made, and these tissues were then stretched to the extent of the serosal incision . This procedure resu lted in little bleed ing. The feta l head and one fo re leg were de livered. Polyviny l catheters were placed in th e fetal brachiocephalic artery and superior vena cava through incisions in the axillary artery and a superficial shou lder vein . In five experimental animals, we used an operating mic roscope to re move the tym pa nic me m brane and ossieles , wo r king thro u gh a posta uricular approach. Oval and round windows were identified. With the aid of round cutting burs, all turns of the cochlea were unroofed, with destruction carried to rhe mod iolus . After cornple tion of three ablation experirnents, it was brought to o u r attention that fe ta l cerebral oxygen co ns umption rates and local cerebral glucose uti liza tio n rares were co rrelated with th e be ha vioral state of the fetus at the time of the experiment.": n T hus in five co ntrol animals and two of five experimental animals, stainless steel screws (0-80 ; l/8 in) were placed bilate ra lly in parietal bone I to 2 cm on ei ther side of the midline cranial suture. Additional screws (0-80; 1/4 in) were placed through pred ri lled holes in the roof of the orbit. Stra nd ed tin ned co p per lea d wires, previo usly silver soldered to the screws, were led out through the scalp and skin incisions, Wire eleetrodes (Cooner Wire Cornpany, Chatsworth, CaIif.) placed on a neck strap muscle completed the instrumentation required for asscssmeni: of fetal behavioral state, All fetal skin incisions were cIosed. Ampicillin, 0.5 grn, was introd uced in the arnniotic flu id. The feta l
1440
Abrams et aI.
December 1987 Am.J Obstet GynecoI
Table 11. Local cerebral glucose utilization (!LmoIlIOO gm/min) Cochlear destruction
126 Motor, sensory cortex Visual cortex Auditory cortex Cingulate gyrus Thalamus Medial geniculate Lateral geniculate Hypothalamus Mamillary body Hippocampus Amygdala Septum Caudate Putamen Subthalamic nuclei Vestibular nuclei Cochlear nuclei Superior olives Inferior olives Inferior colliculus Superior colIiculus (st. super.) Superior colliculus (st. profund.) Habenula Preoptic area Spinal cord, ventral horns Reticular formation Lateral lemniscus Periaqueductal gray Nuclei of oculomotor nerves Red nuclei Vermis Cerebellar cortex Cerebellar nuclei Flococulus Spinal cord, lateral columns Internal capsule Corona radiata Optic chiasm
I
122
I W-B2
j
Y-Bl
I 0-9
0-28
30 24 22 33 35 35 45 22 57 31 30 31 30 27 44 40 29 30 46 33 43
35 30 27 36 45 53 57 30 59 31 32 45 25 27 69 61 50 68 52 82 38
I
Controls
0-14
I 214)
0-2
p*
29 29 22 41 42 62 55 20 51 50 30 32 31 35 56 76 57 86 75 105 56
54 46 42 66 68 103 98 37 111 44 64 74 47 48 124 115 92 105 122 178 77
20 21 17 24 29 41 35 18 41 26 29 20 22 24 34 41 38 44 39 73 30
48 51 47 70 69 96 78 39 112 60 57 54 50 52 95 100 72 100 91 126 101
0.029 0.025 0.044 0.034 0.009 0.014 0.025 0.021 0.050 0.020 0.014 0.027 0.036 0.015 0.022 0.007 0.002 0.0007 0.016 0.002 0.058
0-17
I
20 23 17 27 20 24 37 14 52 27 15 19 18 18 25 20 13 16 24 24 28
18 20 20 22 18 19 28 17 31 18 12 14 14 18 19 21 18 19 23 28 30
15 7 18 10 10 9 8 10 21 17 8 10 15 21 15
15 15 13 21 21 21 35 13 39 21 15 17 20 17 38 38 19 19 35 34 34
16
16
10
19
25
29
30
45
18
65
0.045
22 14 13 16 18 13 22 19 19
18 14 17 14 19 16 18 18 20 12 26 19 11
10 9 8 10 10 7 13 12 10 8 16 14 5 7 3 3
2'.) 13 21 17 20 12 29 30 16 14 32 22
44 20 29 27 27 21 34 40 30 21 52 40 15 24 9 12
46 27 39 35 50 24 34 44 36 25 50 35 19 20 11 14
48 22 48 31 66 18 49
85 38
25 19 27 17 39 15 35 25 23 16 36 25 16 14 12 16
97 44 43 47 71 35 72 70 50 38 82 62 25 30 13 28
0.036 0.014 0.006 0.018 0.001 0.041 0.012 0.039 0.010 0.017 0.024 0.026 0.006 0.046 0.010 0.014
II
24 20 8 14 9 10
IO
8 14
9 10 7 12 10
10
12 7 11
51 22 61 48 29 23 22 18
48 89 35 68 64 63 36 100 56 34 16 25
*Destruction versus controls.
head and foreleg were repositioned in the uterus and the uterus was c1osed. A second dose of 0.5 gm ampicillin was placed in the peritoneal cavity of the ewe be fore the anterior abdominal wall was c1osed. Leads were tunneled subcutaneously and exited on the maternal flank, where they were stored in a pouch sutured to the skin. The skin was closed with clips. Finally, maternal femoral arterial and venous catheters were placed. Anesthetic levels were reduced during the c1osing procedure so that the ewe stood within 5 minutes of extubation. Ewes were placed in carts daily to flush catheters. A cart containing a companion ewe was positioned nearby. On the day of the local cerebral glucose utilization experiment (3 to 8 days after surgery), the fetal behavioral state was assessed during a 2-hour control period (Table I).
Details of the method used to measure cerebral glucose utilization have been published." Briefly, 100 !LCi [I4C]-2-deoxyglucose (American Radiolabeled Chernicals, St. Louis, Mo.) per kilogram of estimated fetal weight were given over 30 seconds. The histories of the arterial plasma levels of [I'C]-2-deoxyglucose and glucose were determined by frequent blood sampling for 30 to 40 minutes. Fetuses and ewes were given intravenous sodium pentobarbital solution followed immediately by intravenous KCI to induce cardiac arrest. The fetus was delivered and weighed and the brain was removed and frozen for later sectioning and quantitative autoradiographic examination.." Local cerebral glucose utilization was caIculated for 38 cerebral structures. Means and SEM of values for experimental and control groups were caIculated across structures. Comparisons between groups were by
Fetal cerebral glucose utilization
Volurne 157
1441
Nurnber 6
paired Student's t test with significance taken as
p < 0.05. The average rate of glucose utilization for the entire brain was not measured. Results
Gross inspection ofthe fetal temporal bones removed after the study revealed that the cochleae were destroyed bilaterally to the level of the modiolus. Table I summarizes data for both experimental and control fetuses. Fetal weights, pH, Paco., Pao., and glucose levels were within the range of those characteristic of our Hock." Local cerebral glucose utilization in experimental fetuses was uniformly depressed, with 33 of 34 gray matter structures and foul' of foul' white matter structures differing significantly (p < 0.05) from respective structures in control fetuses (Table 11). Rednerion in local cerebral glucose utilization in the four brain stem auditory nuelei (cochlear nuelei, superior olives, lateral lemniscus, and inferior colliculus) reached even higher levels of significance (p < 0.002). The pattern of glucose utilization in the inferior colliculus as represented pictorially in autoradiographs was elearly altered. with the appearance of a very pale central nueleus and a denser periphery. Behavioral state cyeling during the 30 to 40 minutes required to measure local cerebral glucose utilization appeared to be normal in two operated fetuses, and the quality of the electrocorticogram, electrooculogram, and neck eIectromyographic tracings was good. Comment
The [IIGJ-2-deoxyglucose method of Sokoloff et al." offers a powerful method of metabolic encephalography. As seen consistently on autoradiographs, the brain structures with the highest concentrations of [IIGJ (and thus with the highest glucose utilization rares) are the cochlear nuclei, superior olives, lateral lernnisci, and inferior colliculi-all structures in the central auditory pathway. Audiographs of the inferior colliculi are characterized by differing densities, with the darkest area occupying part 01' all of the central nueleus. This dark region, whose metabolic rate normally exceeds that of all other loci in the brain, commonly takes on the shape of an ovoid 01' band. Other bands, presumably representative of a tonotopic, spacial organization of frequency coding,"'?" are occasionally seen. These autoradiographic features have been observed by us in all normal fetal sheep. We inferred that the high rares of glucose utilization in central auditory structures could be at least partly relared to environmental sound that is continuously present.":" The sound originates outside the ewe as well as within the ewe, through intestinal, cardiovascular, and respiratory activities and through vocalization. Intra-amniotic pressure changes resulting from pulsation of the umbilical artery could be an additional source of sound. Because
of the well-known link between local functional activity and local metabolie rate in the brain," ambient noise intense enough to evoke fetal cardioacceleration and movement would be expected to activate the metabolism along theauditory pathways. The specificity of this metabolic response, depending as it must on the rnaturation of the central auditory pathways and connections, would be more advanced in sheep fetuses than in term fetuses of less precocious species. Bilateral destruction of the fetal cochleae resulted in adramatic reduction in glucose utilization in the central auditory nuclei. This phenomenon has been weil documented postnatally in animals.": 24. 25 We did not explore the effects of unilateral destruction, but this procedure has been used to judge the levels and extent of crossing of afferent auditory fibers." A decrease in glucose utilization in nonauditory structures in operated fetuses was not anticipated, and has not been mentioned previously in descriptions of cochlear ablation in animals postnataIly. Virtually all gray and white matter structures were affected to a significant extent. In earlier experiments in fetal sheep, glucose utilization was found to be relatively high in brain stem auditory structures at the beginning of the third trirnester." Thus central auditory projection to nonauditory structures, including the reticular formation, would appeal' early and could dominate the representation of other sensory modalities that have yet to mature to the same degree. Continuing with this reasoning, removal of the cochleae could result in a diffuse inactivation of the central nervous systern, which would be reflected in its lowered metabolie rate. The reduction could be linked to the elimination of any transduction of acoustic energy by the destroyed cochleae as weIl as to the removal of spontaneous, resting neural discharge of auditory neurons." The possibility must also be considered that the overall reduction in cerebral glucose utilization may have resulted from the generalized trauma associated with the destructive lesioning. However, the technique used in bilateral cochlear destruction was entirely peripheral, without violation of dura 01' intracranial vasculature 01' contents. While microcontamination of cerebrospinal fluid via a patent cochlear aqueduct is possible, the effects of such blood contarnination must be considered minimal. Whatever the cause of the lowered cerebral metabolie rate, it may have some consequences for normal maturation of the fetal brain. We thank Drs. K. J. Gerhardt and Jack Johnson for helpful comments on the manuscript, the Lesly Martin and Sue Laing for manuscript preparation.
REFERENCES J. Gagnon R, Runse C, Carmiehael L, Fellows F, Patriek J. Effeets of vibratory aeoustie stimulation on human fetal
1442 Abrams et al.
2. 3. 4. 5. 6. 7. 8. 9. 10.
11. 12. 13. 14. 15.
breat hing a nd g ross feta l bod y movc rnents near term. AM j O BSTET GYNECOL 1986 ; 155: 1227-30. Oelman SR, Wood S, Sp elIacy WM , Abrams RM. Fetal movernents in respon se to sou nd stimulation . AM j OBSTET GYNECOL 1982; 143:484-5. Murphy KP, Smyth CN . Respo nse of foetus to a uditory stimulation . Lan cet 1962;5:972-3. Gri mwade j C , Walke r DW, Wood C . Sensory stirnu lation of th e h uman feru s. Aust j Ment Res 197 0 ;2:63-4. Serafini P, Lind say MBj , Nagey DA, et al. Anteparturn fe tal heart rate resp onse to sou nd stim ulation : th e acoustic sti mula tio n test. AMj OBSTET GYN ECOL 1984 ; 148:4 1-5. Sa kabe N, Ara ya ma T , Suzuki T. Human feta l evok ed response to aco usti c stim ula tion . Acta Otolary ngol (su p pl) 1969 ;252:29-36. Goo dli n RC, Schmidt W. H uman fetal a ro usal levels as ind icated by heart rate reco rdin gs. AMj OBSTET GYNECOL 1972;114 :613-21. Salk L. Mothers' heartbeat as a n imprinting stimulus. T rans NY Acad Sei 1962 ;24:753-63. Kolat a G. Studying learnin g in th e womb. Science 1984 ; 225 :302 -3. Prenatal e ffects of ex pos ure to high-l evel noise . Repo rt o f Wo rk ing Grou p 85, Co mmi ttee o n Hea ring, Bioacoustics, a nd Biorn echanics, Assernb ly of Behavioral an d Social Sciences , National Resea rch Co u ncil. Washi ng to n, DC: National Academy Press, 198 2. Br edb erg G. The h u ma n cochlea d uring de velo pme n t a nd ag ing. j La ryn gol Ot o l 1967 ;8 1:739-58. Be rnhard GG, Kaise r IH , Kalom odi n GM . O n the d evelo p mem ofcortical activity in feta l sheep. Acta Ph ysio l Sca nd 1959;4 7:333-49 . Ar mitage SE, Bald win BA, Vince MA. T he fe tal sou nd env iro nment of sheep. Scien ce 1980;208 : I 173-4. Vin ce MA, Billing AE , Bald win BA, T oner j H , Weller C. Mat emal voca Iizations a nd othe r so unds in th e fe tal Iarnb 's sou nd enviro n rne nt . Earl y H uman Dev 1985; I I: 179-90. Gerhard t KH, Abrams RM. T he sound envi ron me nt of the fet al sheep in ut ero . Fed Proc ' 1987;46:358,
Dece m ber 1987 Am J Obs tet Gynecol
16. Abrams RM, 110 M, Fris inger JE, Pa tlak CS, Pettig re w KD, Ken ned y C. Local cer ebral gluc ose utilizat io n in fe tal a nd neona tal sheep. Am ] Ph ysiol 1984 ;246 :R608 - 18. 17. Sok o loff L. Relat io n between ph ysio log ical fu nc tio n and energy me tabolism in the ce nt ra l nervous system. ] Ne urochern 19 77 ;29 ;13-26. 18. Webs te r WR, Servie re ] , Batini C , LaPl ante S. Autoradiogra phie demonstration with 2-[liC]deoxyglucose of frequ ency selec tivity in the a uditory syste m of ca ts under co nditio ns of fu nctional activity. Ne urosei Lett 1978; 10 :43-8. 19. Ryan AF, Woolf NK , Sharp FR. T onotopic orga nization in the centra l aud ito r y path way of the mon golian gerbil: A 2-d eox yglucose study. j Comp Neurol 1982 ;207 : 369-80. 20 . Woo lf NK, Sharp FR, Davidson T M, Ryan AF. Cochlear and middle ear effects o n metab olism in the central auditory pathway during sile nce: a 2-deoxyglucose study. Brain Res 1983;274:1I9-27. 21 . Rich ardson BS, Patrick ] E, Abd uljabbar H . CerebraI ox id ati ve metabolism in th e fe tal lamb : Relationship to electroco rtical state. AM] O BSTET GYNECOL 1985; 153:42 6-3 I . 22 . Abram s R, Hu tch ison AA,]ay TM , Kenned y C . Cerebr a! glucose ut ilizat ion incr eased d u rin g 'ra pid eye movement slee p. Ann Ne u ro l 1986;20:40 1. 23 . Sok oloff" L, Reivich M, Ke nn ed y C, et al. The[" C) deox yglucos e meth od for the measurement of Ioeal cerebral gl ucose utilizatio n : th eory, proced u re, a nd no rmal values in the consc ious and anesthetized albino rat. J Ne uroc hern 19 77 ;28 :89 7-91 6. 24. Sasa ki CT, Kaver .IS, Bab itz L. Diffe re nti al [" C] 2-deoxyglucose u ptake after deafferentation of the mam malian au ditory pathw a y-a model for e xa mining tinniWs . Brain Res 1980 ;194 :5 11-6. 25. T a nag uchi 1. Changes in metabolic ac tivity in the in feri o r co llicu lus fo llo wing removal of au di tory inpu t. Biomed Res 1980; 1:5 10-6. '