Am
J
Otolaryngol
11:292-298,199O
Human Cytomegalovirus Labyrinthitis MELVIN STRAUSS, MD The purpose of this report is to present a review of the clinical and histopathologic findings in human cytomegalovirus (HCMV) labyrinthitis. Human cytomegalovirus infection was first recognized in its congenital disseminated form and, subsequently, in a subclinical form. In 1968, the first temporal bone demonstration of endolabyrinthine infection was reported. To date, there have been reports made on only nine temporal bones in infants with HCMV infection. The light microscopic findings, in general, have been similar, with cytomegalovirus inclusion-bearing cells involving the nonneuroepithelial cells in both the cochlear and vestibular endolabyrinth. A few cases have been studied with fluorescent antibody methods, but have resulted in conflicting reports regarding the presence of virus in cells of the organ of Corti and spiral ganglia. Subclinical congenital HCMV infection, including labyrinthine infection, has also been studied. Only two groups have performed major prospective studies along with systematic and careful periodic audiologic evaluation in cytomegalovirus congenitally infected children. Their findings demonstrated significant hearing loss in 33% to 48% of symptomatic patients and in 6.9% to 13% of asymptomatic children. There also was evidence of progressive hearing loss in some of these children. Vestibular dysfunction is less well characterized. The long-term sequelae may include Meniere’s syndrome. Acquired HCMV infection in children and adults has not been demonstrated histopathologically, although it has been associated with sudden deafness and acute labyrinthitis. This observation is based primarily on serologic data. AM J OTOLARYNGOL 11:292-298. 0 1990 by W.B. Saunders Company. Key words: cytomegalovirus, cytomegalic inclusion disease, labyrinthitis, viral-induced hearing loss.
range of 0.7% to 6% of women develop a primary CMV infection.5r” In addition, an even larger number of women experience secondary or recurrent infection during pregnancy.6,7 Both types of maternal infection may produce infection in the newborn. The vast majority of these maternal infections are unrecognized and result in similarly asymptomatic infections in the newborn. Transmission of the infection occurs by vertical and horizontal modes. Transplacental infection with CMV is the most common cause of congenital viral infections, with an incidence of 0.4% to 2.3%.” It appears to have significant potential injurious effects to the fetus and newborn, but under relatively specific conditions. Following primary maternal infections, transmission of the virus to the fetus occurs in 25% to 50% of cases, but only 10% to 14% of infected newborns suffer detectable consequences. The overall probability of being born with a congenital abnormality after primary infection is 5%. Serious sequelae in the newborn after secondary infection are rare and none have been recognized after perinatal infectiona*’ In most cases, recognition of congenital infection occurs only with special laboratory evaluations that demonstrate CMV. The remaining 5% to 10% of infected offspring have clinically apparent disease at birtha
In 1956, Smith and colleagues independently isolated human cytomegalovirus (HCMV).1*2 Several different strains of HCMV have since been isolated. Cytomegalovirus (CMV) is a member of the ubiquitous herpesvirus group. It is considered to be an ancient parasite of humans and has maintained itself by its low pathogenicity to the majority of those infected. This is manifested by its prevalence, as noted in serologic surveys, of large populations. Antibodies are noted in early childhood and present in 44% to 100% of study populations by adulthood.2-4 Cytomegalovirus infection is associated with distinct disease entities. The first significant forms of infection to be recognized were congenital, while the most recently recognized are serious opportunistic infections in the immunocompromised host. Studies have shown that during pregnancy, a Received May 16, 1990, from the department of Otolaryngology-Head and Neck Surgery, Case Western Reserve University School of Medicine, 2074 Abington Rd, Cleveland, OH 44106. Accepted for publication June 9, 1990. Address correspondence and reprint requests to Melvin Strauss, MD, Department of Otolaryngology, Head and Neck Sureerv. Case Western Reserve University School of Medicine, 2074 Abington Rd, Cleveland, OH 44106. 0 1990 by W.B. Saunders Company. 0196-0709/90/1105-0001$5.00/O Y
-.
292
293
STRAUSS
The congenital form of CMV infection that is symptomatic at birth is termed cytomegalic inclusion disease (CID). The first recognized cases of CID were severe and disseminated in nature.” Since their discovery, less severe but symptomatic forms of the disease have been recognized. Cytomegalic inclusion disease is characterized by low birthweight, prematurity, microcephaly, chorioretinitis, and psychomotor retardation. Other findings include jaundice, petechiae, purpura, and hepatosplenomegaly. In its most severe form, CID results in death in the perinatal period. It is estimated to occur in 1 in 3,000 live births and to be present in 1% to 2% of infants at autopsy.l”l’ Prospective studies have shown that approximately 1% of all newborns excrete CMV in their saliva or urine while being only mildly symptomatic or asymptomatic. Evidence from long-term follow-up of these infants indicates that subtle to obvious and serious sequelae may result. These abnormalities occur in up to 10% of these children and include varying degrees of perceptual, behavioral, psychomotor, and other neurologic abnormalities. CONGENITAL HUMAN CYTOMEGALOVIRUS-INDUCED DEAFNESS Deafness occurs in 20% to 65% of CID children, and is the most common sequela of asymptomatic infection.6~‘2‘15 Two groups have conducted major prospective studies of congenital CMV infection along with systematic and careful periodic auditory evaluation. Stagno et al detected sensorineural hearing loss in 17% of children with congenital CMV infecti0n.l” Significant hearing loss was detected in 48.4% of symptomatic children and in 6.9% of asymptomatic children. The defects were predominantly bilateral and were often of a severe to profound degree. There was no consistent pattern of loss, although high frequencies were generally involved. In an early report from this group, two children appeared to have their hearing deteriorate in a progressive fashion over a l- to z-year period. The most recent report from the same group indicates that in 25% of the children followed, hearing impairments have either developed or become more severe after the first year of life.16 Forty-one children with congenital CMV infection were detected prospectively and assessed audiometrically by Saigal et a1.15These investigators found sensorineural hearing loss in 33% of the children with symptomatic infection and in 13.1% of those with asymptomatic infection. No instances of progressive loss were detected.
In a study of six mildly symptomatic CID children, I documented a 50% incidence of moderate to severe hearing loss, as well as progressive hearing loss in those same children.17 CONGENITAL HUMAN CYTOMEGALOVIRUS-INDUCED VESTIBULAR DISEASE Evidence of vestibular problems in congenital CMV infections is much less well documented than hearing loss. Pappas reported vestibular findings in 19 children with documented congenital asymptomatic CMV infection.” He reported that vestibular impairment did not correlate with the presence of severity of associated hearing loss. My evaluation of six mildly symptomatic children showed a close correlation of vestibular dysfunction to hearing loss, such that only those children with hearing loss demonstrated caloric and/or electronystagmographic abnormalities and impairments of coordination.17 HUMAN CYTOMEGALOVIRUS TEMPORAL BONE PATHOLOGY In confirmation of the clinical data, CMV has been observed in the infant labyrinth. However, there have been only nine infant temporal bone reports of such cases.17,1g-24 Seven of these had CID and two had silent or apparently asymptomatic infection. Ages ranged from 3 weeks to 5 months (Table 1). Temporal bone findings in the CID group revealed endolabyrinthitis in four cases, infection in the endolabyrinth and spiral ganglia in two cases, no histologic evidence of infection in two cases, and CMV isolation by culture of the labyrinth in one case. In the two cases of asymptomatic infection, endolabyrinthitis was noted in one case and CMV isolation by culture in the other. Cochlear histopathology has shown inclusionbearing cells within the Scala media, involving Reissner’s membrane and the stria vascularis. The organ of Corti had been spared, that is, it has no demonstrable viral inclusions. Immunofluorescent methods were used to analyze four infant temporal bones. Stagno et al demonstrated evidence of viral antigen in two infants whose standard histopathology showed not only viral inclusions, but also showed evidence of CMV antigen in the organ of Corti, which did not reveal inclusions using standard methods.” Davis et al, however, could not demonstrate evidence of viral antigen in the organ of Corti in an infant they studied.23 Hydrops or collapse of Reissner’s membrane
3wk
Premature infant, hyaline membrane disease
Davis et alzl
3
Inclusion-bearing cells in the submandibular glands and kidneys, purulent meningitis, and bronchopneumonia CID with inclusionbearing cells in lung, liver, kidney, pancreas, thyroid and brain; pneumonia and bronchopulmonary dysplasia
6wk
Stiff neck and respiratory distress; CSF culture pneumococci
Davis”’
2
GENERAL HISTOPATHOLOGY
of Temporal
CID, exfoliative dermatitis, hepatosplenomegaly, and bronchopneumonia
Myers & Stool’9
1
AGE AT DEATH
Summary
3wk
Jaundice and hepatosplenomegaly
SOURCE
CASE
CLINICAL FINDINGS
TABLE I.
Studies
Reported
INNEREAR Endolabyrinth, only, involved
Endolabyrinth, only, involved
Endolabyrinth, only, involved
Formalin, celloidin, H&E with light microscopy
Formalin, paraffin, H&E with light microscopy, and transmission electron microscopy Glutaraldehyde, celloidin, H&E with light microscopy, and transmission electron microscopy; anticomplementary immunofluorescence assay for CMV
COCHLEAR
FINDINGS
VESTIBULAR FINDINGS
Human
Only nonneural and nonsensory epithelial cells contained CMV inclusions; associated serous labyrinthitis
Utricle and saccule with several inclusionbearing cells; could not be certain if any were sensory epithelial cells; inclusion-bearing cells in all semicircular canals but not within the planum semilunaturn or sensory epithelia of cristae [nclusion-bearing cells in epithelium of semicircular canals
in Cases of Congenital
One bone with most inclusion-bearing cells in the hook region, while the other showed involvement of all coils, including Reissner’s membrane and the stria vascularis; no inclusionbearing cells in the organ of Corti; hydrops of apical coil on one side and the entire Scala media of the other, cystic structure of stria vascularis Inclusion-bearing cells in Reissner’s membrane and stria vascularis; no inclusionbearing cells in organ of Corti; microcysts in stria vascularis Only nonneural and nonsensory epithelial cells contained CMV inclusions; all coils involved; extensive collapse of Reissner’s membrane
in the Literature Infection
HISTOPATHOLOGY
Cytomegalovirus TEMPORAL BONE HISTOPATHOLOGY: METHOD OF PREPARATION AND STUDY
Bone
No inclusion-bearing cells in ganglia or eighth nerve
No inclusion-bearing cells in spiral ganglia or eighth nerve
No inclusion-bearing cells in spiral or Scarpa’s ganglia
NEURAL FINDINGS
7
4-6
Severe CID
Lethargy and hypotonia, cutaneous petechial hemorrhage, jaundice, hepatosplenomegaly, no apparent response to sound, CID
Stagno et al*’
Davis et aP
3-
4wk
IO d
CMV inclusionbearing cells noted in kidney, lung, liver, and thyroid; CMV isolated by viral culture from lung, liver, and spleen; periventricular calcifications and severe bacterial pneumonia
None reported
Viral culture of aspirated nerilymph, 4% glutaraldehvde. osmium tetroxide, with microdissection, stereomicroscopy and phase microscopy; scanning and transmission electron microscopy; alcohol, paraffin, and H&E with light microscopy; indirect immunofluorescence antibody method; CMV isolated by culture
Ethanol, paraffin, H&E with light microscopy, and transmission electron microscopy; anticomplementary immunofluorescence assay for CMV
Endolabyrinth, only, involved; vestibular infection most extensive
Endolabyrinth and spiral ganglia involved in two cases; in the other, there was no evidence of CMV
In two cases, light Edema and inmicroscopy showed flammatory edema and inflamreaction with matory reaction with plasma and rare inclusion-bearing mononuclear cells involving Reisscells ner’s membrane and the stria vascularis; a third case demonstrated no viral inclusions; in two cases, immunofluorescence studies demonstrated the presence of CMV antigen in infected cells as rated bv light microscopy, asweil as in cells of the organ of Corti; the case in which standard examination showed no evidence for CMV also failed to show evidence by the immunofluorescence method Extensive numExtensive autolysis due bers of incluto delay in fixation; sion-bearing inclusion-bearing cells cells adjacent in basal-most portion to neuroepithewith Reissner’s hum in colmembrane and spiral lapsed saccule ligament involved; and utricle; collapse of ductus reupossible selecniens; no evidence for tive involveCMV antigen presence ment of dark by immunofluorescells; abnorotocence method in organ conia; few cells of Corti with inclusions involving semicircular canals Light microscopy and immunofluorescence method showed no evidence of CMV antigen by immunofluorescence method
Light microscopy showed no evidence of CMV presence, while immunofluorescence technique showed evidence of CMV antigen in the spiral ganglia in two cases where CMV was not present by standard methods
(XV-bearing inclusions noted in kidney and lung; prematurity, pseudomonas pneumonia and septicemia, cerebellar hemorrhage Atrophic, cystic brain: CMV cultures positive for kidney
GENERAL HISTOPATHOLOGY
stain; EM, electromicroscopy.
5mo
Microcephaly, seizure disorder, encephalitis
Davis et alz4
H&E, hematoxylin-eosin
I wk
Prematurity with 27-wk gestation and 1,050-g birth weight; apnea and bradycardia. CID
AGE AT DEATH
Strauss”
SOURCE
Abbreviations:
9
8
CASE
CLINICAL FINDINGS Formalin, celloidin, H&E with light microscopy; viral cultures for CMV from brain, salivary gland, kidney, and liver showed no evidence of CMV Viral culture of aspirated perilymph endolymph, 2.5% glutaraldehyde and buffered formalin; right temporal bone microdissected and studies by EM; lefi temporal bone celloidin, H&E light microscopy; CMV isolated by culture
TEMPORAL BONE HISTOPATHOLOGY: METHOD OF PREPARATION AND STUDY
TABLE 1.
Normal
Normal
No evidence of inclusionbearing cells or other abnormalities of the inner ear
COCHLEAR FINDINGS
No evidence of inclusionbearing cells or other abnormalities of the inner ear
INNER EAR HISTOPATHOLOGY
Continued
Exudate present in endolymph of saccule
Normal
VESTIBULAR FINDINGS
Normal
Normal
NEURAL FINDINGS
297
STRAUSS
was variably noted. Microcysts were noted in the stria vascularis in one case. The majority of cases showed no associated inflammation There was also no evidence of teratogenesis in any specimen. Vestibular histopathology demonstrated inclusion-bearing cells in the membranous labyrinth, involving the utricle, saccule, and semicircular canals. There was some variability in the extent of the involvement of cochlear versus vestibular sites, with cochlear involvement being predominant. However, some of the specimens did not include the pars superior. The vestibular neuroepithelium appeared to be spared from viral infection. There was variable collapse of the saccule and utricle in different specimens. Generally, there was no associated inflammation. No evidence of teratogenesis was noted. Neural histopathology, using standard methods, showed no evidence of viral inclusionbearing cells. Immunofluorescent methods were used by Stagno et al, who demonstrated viral antigen in the spiral ganglia while Davis et al could not. 22.23 ACQUIRED HUMAN CYTOMEGALOVIRUS LABYRINTHINE INFECTION Reports of labyrinthine involvement secondary to CMV infection in adults are sparse. Klemola and Kearianeu reported a patient with CMV mononucleosis and labyrinthitis.” This patient had both cochlear and vestibular abnormalities that resolved spontaneously (Klemola and Kearianeu, personal communication, September 1976). Weiss and Ronis reported a case of Meniere’s syndrome in a young adult who was previously diagnosed as having congenital deafness secondary to CID.26 This most likely represents a case of delayed endolymphatic hydrops secondary to the congenital infection.27 For the most part, CMV infection is acquired early in life and is relatively asymptomatic. However, it has been implicated as one of the etiologic agents of idiopathic sudden hearing loss in adults. This entity has an incidence of 7 to 15 per 100,000.Of the few convincing studies to demonstrate a viral etiology of sudden hearing loss, the most satisfactory is that of Wilson et al.” Sequential viral serologies were analyzed in 122 idiopathic sudden hearing loss patients and 95 control patients. Only fourfold viral titer changes were considered significant. Sixty-three per cent of patients with idiopathic sudden hearing loss versus 40% of control patients demonstrated significant titer changes. Nine of these 122 were pos-
itive for CMV, strongly implicating CMV as an etiologic factor in this disorder. Currently, CMV is one of the more common opportunistic viral agents affecting acquired immunodeficiency syndrome patients. Although sensorineural hearing loss is known to occur in acquired immunodeficiency syndrome patients, there is no evidence to date, of which I am aware, indicating that CMV is an etiologic factora2’ Standard or temporal bone studies have yet to be performed on older children or adults with proven CID or other forms of CMV labyrinthitis. It is assumed that once studies have been performed, the changes detected will be similar to those found in those cases of sudden hearing loss of unknown etiology that demonstrate organ of Corti atrophy. It would be important to determine CMV’s exact length of persistence in the labyrinth, its pathophysiologic mechanisms, and its possible role as an etiologic factor in producing endolymphatic hydrops. SUMMARY CMV is the most common cause of congenital viral-induced hearing loss. It is estimated to cause deafness in 2,000 to 4,000 new cases annually in the United States. Temporal bone studies of this entity have been limited. Additional human CMV-infected temporal bones need to be acquired and studied using standard and more sophisticated methods. Cytomegalovirus infection of the labyrinth, acquired at other times during life, may also produce sudden hearing loss and/or Meniere’s syndrome. Currently, the guinea pig model offers the best opportunity for investigating the pathophysiology of CMV labyrinthitis and developing methods for its prevention and control. References 1. Gold E, Nankaveris GA: Cytolomegalovirus, in Evans AS (ed): Viral Infections of Humans [ed 2). New York, NY, Plenum, 1982, pp 167-186 2. Rowe WP, Hartley JW, Waterman S, et al: Cytopathogenic agent resembling human salivary gland virus recovered from tissue culture of human adenoids. Proc Sot Exp Biol Med 1956; 92:418-424 3. Stern A, Elek SD: Incidence of infection with cytomegalovirus and normal population: Serologic study in greater London J Hyg (Lond) 1975; 63:79-87 4. Kreck U, Tobin J: Complement fixing antibodies against cytomegalovirus in different parts of the world. Bull WHO 1973; 49:103-106 5. Sever JL, Huebner RJ, Castellano GA, et al: Serologic diagnosis in “en masse” with multiple antigens. Ann Rev Respir Dis 1968; 88:342-359 (suppl) 6. Stagno S, Pass RF, Dworsky ME, et al: Maternal cytomegalovirus infection and perinatal transmission. Clin Obstet GynecollQ82; 25:563-576 7. Stagno S, Pass RF, Dworsky ME, et al: Congenital cytomegalovirus infection. The relative importance of primary and recurrent maternal infection. N Engl J Med 1982; 306:945-949
298 8. Peckham CS, Coleman JC, Hurley R, et al: Cytomegalovirus infection in pregnancy: Preliminary findings from a prospective study. Lancet 1983; 2135%1355 9. Ahlfors K, Harris S, Ivarsson S, et al: Secondary maternal infection causing symptomatic congenital infection. N Engl J Med 1981; 305:284 10. Wyatt JR, Saxon J, Lee R, et al: Generalized cytomegalic inclusion disease. J Pediatr 1950; 36:271-294 11. McCordock HH, Smith MG: Intranuclear inclusions. Am J Dis Child 1934; 47:771-779 12. McCracken GW, Jr, Scheinfield HR, Cobb K, et al: Congenital cytomegalic inclusion disease. Am J Dis Child 1969; 117:522-539 13. Pass RF, Sergio S, Myers GJ: Outcome of symptomatic congenital cytomegalic infection: Results of long-termlongitudinal follow-uo. Pediatrics 1980; 66:758-762 14. WIlliamson WD, Desmond MM, LaFevers N, et al: Symptomatic congenital cytomegalic virus disorders of language, learning and hearing. Am J Dis Child 1982; 136:903-905 15. Saigal S, Lunyk 0, Larke B, et al: The outcome in children with congenital cytomegalovirus infection. Am J Dis Child 1982; 136:896-901 16. Stagno S, Pass RF, Dworsky ME, et al: Congenital and perinatal cytomegalovirus infections. Semin Perinatol 1983; 7:31-42 17. Strauss M: A clinicopathologic study of hearing loss in congenital cytomegalovirus infection. Laryngoscope 1985; 95:951-962 18. Pappas DG: Hearing impairments and vestibular abnormalities among children with subclinical cytomegalovirus. Ann Otol Rhino1 Laryngol 1983; 92:552-557
HUMAN CYTOMEGALOVIRUS
LABYRINTHITIS
19. Myers EN, Stool S: Cytomegalic inclusion disease of the inner ear. Laryngoscope 1968; 78:1904-1915 20. Davis GL: Cytomegalovirus in the inner ear: Case report and electron microscopic study. Ann Otol 1969; 78:1178-1188 21. Davis GL, Spector J, Strauss M, et al: Cytomegalovirus endolabyrinthitis. Arch Path01 Lab Med 1977; 101:118-121 22. Stagno S, Reynolds DW, Amos CS, et al: Auditory and visual defects resulting from symptomatic and subclinical congenital cytomegaloviral and toxoplasma infections. Pediatrics 1977; 59:669-678 23. Davis LE, Johnson LG, Kornfeld M: Cytomegalovirus labvrinthitis in an infant: Moruholoaical. viroloaical. and immunofluorescent studies. J Paihol Exp Nkurol li81; 40:9-19 24. Davis LE, Rarey KE, Stewart JA, et al: Recovery and probable persistence of cytomegalovirus in human inner ear fluid without cochlear damage. Ann Otol Rhino1 Laryngol 1987; 96:380-383 25. Klemola E, Kearianeu L: Cytomegalovirus as a possible cause of a disease resembling infectious mononucleosis. Br J Med 1965; 2:1099-1192 26. Weiss JS, Ronis ML: Otologic manifestations of cytomegalic inclusion disease. A case report. Trans Pa Acad Ophthamol Otolaryngol 1977; 30:52-54 27. Shuknecht H: Delayed endolymphatic hydrops. Ann Otol Rhino1 Laryngol 1978; 87:743-748 28. Wilson WR, Veltri RW, Laird N, et al: Viral and epidemiologic studies of idiopathic sudden hearing loss. Otolaryna01 Head Neck Sum 1983: 91:653-658 29. Smith ME, Canalis RF: Otologic manifestations of AIDS: The otosyphilis connection. Laryngoscope 1989; 99:365-372