LUND UNIVERSITY

PO Box 117221 00 Lund+46 46-222 00 00

Optic nerve hypoplasia: Risk factors and epidemiology.

Tornqvist, Kristina; Ericsson, Anders; Källén, Bengt

Published in:Acta Ophthalmologica Scandinavica

DOI:10.1034/j.1600-0420.2002.800313.x

2002

Link to publication

Citation for published version (APA):Tornqvist, K., Ericsson, A., & Källén, B. (2002). Optic nerve hypoplasia: Risk factors and epidemiology. ActaOphthalmologica Scandinavica, 80(3), 300-304. https://doi.org/10.1034/j.1600-0420.2002.800313.x

General rightsUnless other specific re-use rights are stated the following general rights apply:Copyright and moral rights for the publications made accessible in the public portal are retained by the authorsand/or other copyright owners and it is a condition of accessing publications that users recognise and abide by thelegal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private studyor research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

Read more about Creative commons licenses: https://creativecommons.org/licenses/Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will removeaccess to the work immediately and investigate your claim.

A O S 2002

300

Optic nerve hypoplasia: Riskfactors and epidemiologyKristina Tornqvist1, Anders Ericsson2 and Bengt Källen3

1Department of Ophthalmology, University Hospital of Lund, Lund, Sweden2Epidemiological Centre, National Board of Health, Stockholm, Sweden3Tornblad Institute, University of Lund, Lund, Sweden

ABSTRACT.Objectives: To study the epidemiology of optic nerve hypoplasia.Design and Methods: Children with optic nerve hypoplasia and visual impair-ment were identified through the Swedish Register of Visually ImpairedChildren. Pre- and perinatal characteristics were obtained from the MedicalBirth Registry and by scrutinizing pregnancy and delivery records. Clinicalcharacteristics of children with optic nerve hypoplasia are described. The fol-lowing risk factors were studied: maternal age, parity, maternal smoking, ges-tational duration, birth weight, delivery method, Apgar score, maternal diseaseduring pregnancy, drugs used in early pregnancy.Results: Young maternal age, first parity, maternal smoking, preterm birth andfactors associated with preterm birth were risk factors for optic nerve hypo-plasia. There was an indicated association with the use of fertility drugs andantidepressant drugs.Conclusions: Optic nerve hypoplasia is apparently associated not only withother anomolies, notably of the central nervous system, but also with signs ofgeneral disturbance in fetal development.

Key words: optic nerve hypoplasia – epidemiology – risk factors – maternal age – smoking –drugs – prematurity

Acta Ophthalmol. Scand. 2002: 80: 300–304Copyright c Acta Ophthalmol Scand 2002. ISSN 1395-3907

Optic nerve hypoplasia is a non-pro-gressive congenital abnormality

associated with a reduced number ofaxons in one or both optic nerves (Mos-ier et al. 1978). It can be bilateral or uni-lateral and is clinically characterized by asmall optic disc, often surrounded by aperipapillary halo bordered by a darkring of pigment. In addition, increasedtortuosity of the retinal vessels as well asfew branching points can be seen(Hellström et al. 1999). The pathogenesisis not fully understood, but the disordercan result from a number of still un-identified adverse events occurring duringdevelopment of the visual pathways. Thecondition has been considered rare, butan increasing incidence has been reportedduring the last decades (Hoyt & Good

1992). The age-specific prevalence of vis-ual impairment due to optic nerve hypo-plasia in the Swedish population of 0–19years is 6.3/100000 (Blohme & Tornqvist1997c).

The clinical picture reflects varyingfunctional reduction ranging from verymild or no visual impairment to blind-ness (Björk et al. 1978; Lambert et al.1987; Zeki & Dutton 1990). Optic nervehypoplasia may occur as an isolatedevent or in association with other devel-opmental anomalies, most often cerebralmalformations (De Morsier 1956;Skarf & Hoyt 1984; Burke et al. 1991).

A number of aetiological factors suchas drugs (McKinna 1966; Chan et al.1978; Hoyt & Billson 1978; Good et al.1992), maternal diabetes (Patel et al.

1975; Peterson & Walton 1977; Donat1981; Kim et al. 1989; Landau et al.1998), and alcohol (Strömland 1987;Chan et al. 1991; Strömland & Pinazo-Duran 1994) have been associated withthe development of optic nerve hypo-plasia. Viral infections have also been im-plicated (Bistner et al. 1973).

Hitherto, however, no thorough inves-tigation concerning risk factors for opticnerve hypoplasia in a nationwide popula-tion has been performed. In this study,we have explored possible risk factors re-sponsible for the development of opticnerve hypoplasia in all Swedish childrenidentified with this diagnosis and with adisease severe enough to cause visual im-pairment.

Material and MethodsIndividuals with bilateral optic nerve hy-poplasia severe enough to cause visualimpairment were identified by inventoryof the Swedish Register of Visually Im-paired Children (Blohme & Tornqvist1997a). The analysis was restricted to in-dividuals born during 1979–97.

The Swedish Register of VisuallyImpaired Children

This database includes data on all visu-ally impaired children aged between 0and 19years in the country, with visualacuity (VA) of 0.3 and/or a simul-taneous visual field defect. The Registerwas originally established by collectingdata obtained from the medical recordsof low vision clinics and departments ofophthalmology throughout Sweden andis now continuously updated by reportingfrom these sources. Data obtained oneach child are recorded on a standardized

A O S 2002

301

form and subsequently entered into adatabase. Each record contains infor-mation on name, sex, date of birth, treat-ing ophthalmologist and low visionclinic, county, ophthalmological diag-nosis, systemic diagnosis, aetiologicalfactors, additional impairments, classifi-cation of visual impairment, visual fielddefects, refraction, near vision, and fam-ily history.

Suggested aetiological factors are di-vided into four main categories accordingto the system used in previous Nordicstudies (Rosenberg et al. 1992; Blohme &Tornqvist 1997b), as follows:

(1) prenatal;(2) peri/neonatal;(3) infantile/juvenile, and(4) unknown.

Classification of visual impairment ismade according to WHO (World HealthOrganisation) definitions. Ophthalmo-logical diagnoses are classified accordingto a Californian version of ICD-9 (Inter-national Classification of Diseases, 9thversion) (Riise et al. 1992; Hansen et al.1992; Blohme & Tornqvist 1997c).

On December 31, 1999 the database in-cluded data on 2774 individuals aged be-tween 0 and 19years. The total Swedishpopulation aged between 0 and 19yearswas then 2.25 million out of an entirepopulation of 8.86 million. Among in-fants with a diagnosis of optic nerve hy-poplasia, infants with Down syndromeand infants with incomplete personalidentification numbers were excluded.

The Medical Birth Registry

Cases with a complete personal identifi-cation number were matched with theMedical Birth Register for 1977–97. Thisregister was started in 1973 and containsdata on antenatal care, delivery, and thepaediatric examination of newbornchildren (Cnattingius et al. 1990). Since1982, it has been based on copies of theoriginal medical records, computerizedby the National Board of Health, Stock-holm.

Information on drug use in early preg-nancy was obtained from the records ofthe 100 women whose antenatal care rec-ords could be retrieved. The same formis used throughout Sweden. Pregnantwomen are interviewed by their attendingmidwife early in their antenatal care(usually in weeks 10–12), and are askedabout smoking habits and drug use dur-ing the pregnancy, among other issues.

Statistical analysis

Various risk factors were compared be-tween cases and between all infants born(nΩ2109316). Risk estimates were deter-mined as odds ratios (OR) using Mantel-Haenszel procedure after various strati-fications, and 95% confidence intervals(95% CI) were estimated using Mietti-nen’s test-based method. In order tostudy a putative trend in maternal agedistribution, a weighted linear regressionanalysis of the log (OR) was carried out.

The following risk factors werestudied:

(1) maternal age and parity;(2) maternal smoking in early preg-

nancy;(3) gestational duration;(4) birth weight and birth weight for

pregnancy week;(5) maternal diabetes;(6) maternal pre-eclampsia;(7) drug use in early pregnancy;(8) fetal presentation;(9) delivery method, and

(10) low Apgar score at 5min.

Birth weight for each pregnancy weekwas expressed as standard deviations(SD) of the expected birth weight at eachweek according to a sex and parity speci-fic normal growth curve (Källen 1995).Infants with a birth weight less than 2SDbelow the expected mean were regardedas small for gestational age (SGA).

No information on drug use amongwomen with normal infants was retrievedin the study. Comparisons were insteadmade with data from the Medical BirthRegistry, where records of drug use inearly pregnancy have been computerizedsince 1994. This information is based onthe same source as that used in the pres-ent study, and at least crude comparisonscan be made.

ResultsA total of 156 individuals (72 male and 84female; sex ratio 0.86, 95% CIΩ0.62–1.17) with optic nerve hypoplasia werefound in the database of visually impairedchildren. Among these, 63 (40%) had anisolated optic nerve hypoplasia withoutany systemic diagnosis. A total of 39 hadreported congenital cerebral malfor-mations, and of these 13 displayed septo-optic dysplasia and another six showedother mid-line deficiencies such as agen-

esis of the corpus callosum. Additionally28 had diagnoses such as encephalopathyor cerebral palsy, indicating associatedcerebral damage. Altogether, 43% of sub-jects showed signs of cerebral involvement.

As expected, prenatal aetiologies wereby far the most commonly suggested onesand occurred in 125 of the individuals. Ofthese, 104 were aetiologically classified’prenatal unspecified’. The aetiologicalclassification in 26 cases was peri/neo-natal, and in five cases it was unknown(Table1). Additional impairments oc-curred in 98 individuals (63%), for whoma combination of mental and motor im-pairment was the most frequently occur-ring (27%) (Table2). Reduction in visualacuity varied considerably (Table3). Atotal of 17 individuals (11%) had no lightperception, whereas 42% of cases fell intoWHO category 1 or had VA of 0.3. Theproportion of blindness according toWHO criteria was 32%.

Records for a total of 125 children withoptic nerve hypoplasia matched with dataobtained from the Medical Birth Registryand could be analysed. Matching provedimpossible in 23 cases because the sub-jects did not have complete identificationnumbers in the Registry of Visually Im-paired Children (due to registration rou-tines in one particular area of Sweden).The children in the remaining cases hadeither not been registered in the MedicalBirth Register (occurs in 1–2% of all in-fants born) or had not been born inSweden but had immigrated or had beenadopted from abroad.

Among the matched children, therewere three twins (1.6%) and thus 122singleton births.

Table4 shows ORs for maternal age,parity and maternal smoking, with eachfactor stratified for year of birth and theother two factors. There is an increasedrisk for low maternal age (independent ofparity) and a statistically significant de-clining trend with age (p 0.01) and forfirst parity (independent of age), com-pared with higher parity. There is also anincreased risk for parity 4π, althoughthis does not reach statistical significance.Maternal smoking in early pregnancy isa risk factor and a dose-dependency is in-dicated. The effects of smoking were ana-lysed in infants with a birth weight below2500g and infants with a birth weight of2500g or more. In the latter group, theeffect of smoking was further increased:ORΩ1.73 (95% CI 1.10–2.72) for anysmoking and ORΩ2.19 (95% CI 1.23–3.88) for smoking 10 or more cigarettes

A O S 2002

302

per day. No risk increase was noted forany smoking in low birth weight infants(ORΩ0.86, 95% CI 0.35–2.13).

Preterm birth (37 completed weeks)among singletons showed an OR of 3.47(95% CI 2.25–5.35) and low birth weight(2500g) an OR of 4.96 (95% CI 3.27–7.52). The OR for being SGA was 2.63(95% CI 1.42–4.86). All these ORs werestratified for year of birth, maternal age,parity, and smoking habits.

Only one woman was diagnosed withdiabetes. The expected number was 0.4.Four women reported a period of subfer-

Table1. Distribution of suggested aetiological factors among 156 children/adolescents with bilat-eral optic nerve hypoplasia.

Aetiological groups and subgroups Number

Prenatal aetiology 125Genetic without chromosomal cytogenetic abberation 5Genetic with chromosomal cytogenetic abberation 8Infectious disease, prenatal 4Intoxication 2Other prenatal influence, specified 2Other prenatal influence, unspecified 104

Peri/neonatal aetiology 26Maturity with asphyxia 10Maturity, other peri-neonatal complications 6Prematurity with dysoxygenation 5Prematurity with other complications 5

Unknown time and cause 5

Table2. Distribution of additional impairments among 156 Swedish children/adolescents withbilateral optic nerve hypoplasia.

Impairments Number Percentage

Mental (exclusively) 24 (15%)Mobility (exclusively) 12 (8%)Hearing (exclusively) 3 (2%)Mental and mobility 42 (27%)Mental and hearing 5 (3%)Mobility and hearing 1Mental, mobility and hearing 1Other impairments 10 (6%)Total additional impairments 98 (63%)

Table3. WHO classification of visual impairment and the distribution of visual acuities among156 Swedish children/adolescents with bilateral optic nerve hypoplasia.

Category Vision Number ofindividuals

1 Low vision 0.1–0.3 562 0.05–0.1 213 Blindness 0.02–0.05 54 0.02 285 No light perception 179 Unknown 19 Ω 0.3 10

tility before the pregnancy. Maternal pre-eclampsia occurred in only four womenwith singleton births.

The OR for having a caesarean section(based on 32 cases) versus non-instrumen-tal vaginal delivery in singleton deliverieswas 2.98 (95% CI 2.02–4.40), but afterstratification for gestational duration, theOR decreased to 2.55 (95% CI 1.63–3.99).Instrumental vaginal delivery versus non-instrumental vaginal delivery showed anOR of 1.00 (95% CI 0.43–2.34) after strati-fication for gestational week.

Among singleton vaginal deliveries,

only four were breech deliveries. Breechversus head presentation (stratifying forgestational week) showed an OR of 1.55(95% CI 0.54–4.45) (this analysis was re-stricted to the period 1982–97 for techni-cal reasons).

Low Apgar scores (7 at 5min) werenoted in eight singletons. The OR versusApgar score 7 was 3.90 (95% CI 1.96–7.75).

According to the antenatal records forthe 100 women whose records werescrutinized, drug use was reported by 37women, and included a total of 45 drugs(see Table5). Among these, the followingcan be commented upon.

One women used an antifungal drug(griseofulvin). Three women had becomepregnant after drug treatment for infer-tility: bromocriptine (1), cyklofenil (1),and clomifene (1). Four women useddrugs related to manic-depressive disease:lithium (1), clomipramine (2), and citalo-pram (1). All drugs used are accountedfor in Table5.

In the course of scrutiny of antenatalrecords, it was found that one infant hadtwo previous sibs with microcephaly andanother infant had a previous sib with amid-line defect.

DiscussionOur study identified a number of risk fac-tors for optic nerve hypoplasia. Low mat-

Table4. Odds ratio (OR) with 95% confidenceintervals (95% CI) for some maternal charac-teristics to have a child with optic nerve hypo-plasia. Each variable is stratified for year ofbirth and the other two variables.

Maternal OR 95% CIcharacteristic

Age19 1.46 0.70–3.0220–24 1.50 1.02–2.2125–29 0.88 0.60–1.2730–34 0.85 0.54–1.3535–39 0.49 0.21–1.1340– 0.99

Parity1 1.73 1.16–2.582 0.55 0.36–0.843 0.90 0.51–1.594 π 1.36 0.67–2.75

SmokingNone 1.00 reference 10 cigs/day 1.42 0.87–2.33 10 cigs/day 1.84 1.08–3.14All smoking 1.61 1.08–2.41

A O S 2002

303

ernal age (irrespective of parity) and lowparity (irrespective of age) were both riskfactors. Maternal smoking was an inde-pendent risk factor. Among the risk fac-tors reflected by infant characteristics arepreterm birth, low birth weight and beingsmall for gestational age. Analysis ofdrugs used during pregnancy indicatedrisk associated with fertility drugs andantidepressant drugs.

The characteristics of the infants withoptic nerve hypoplasia in the present ma-terial mainly agree with those found insmaller published studies. Our studyfound the condition to occur at a rate ofseven per 100000 births, higher than the1.8–6.3 per 100000 births described inthe literature (Jan et al. 1977; Blohme &Tornqvist 1997c). However, it has beensuggested pervasively that the diseasemay be more prevalent than previouslythought (Lambert et al. 1987). The actualoccurrence of the condition is certainlystill higher than that shown here, as ourstudy was restricted to serious cases in-volving visual impairment (Table3). In-fants with milder forms of optic nerve hy-poplasia may either not have any visualimpairment, or the condition may be uni-lateral.

Optic nerve hypoplasia is characterizedin the literature as a congenital malfor-mation of prenatal origin (Taylor & Stout1997). It is often seen together with septo-optic dysplasia and other cerebral mal-formations (De Morsier 1956; Skarf &Hoyt 1984; Burke et al. 1991). We founda high proportion of infants (63%) withadditional impairments, although this isnot surprising given the high rate of as-

Table5. Drug use recorded in antenatal medi-cal records for 100 women whose children de-veloped optic nerve hypoplasia.

Drug category Number Numberof drugs of women

Gastrointestinal drugs 3 3Insulin 2 1Antifungal drugs 1 1Bromocriptine 1 1Oral contraceptives 1 1Ovulation stimulator 2 2Antibiotics 7 6Muscle relaxants 1 1Analgesics 9 8Lithium 1 1Antidepressants 3 3Common cold drugs 5 3Anti-asthmatic drugs 7 4Anti-nausea drug 1 1Eye drops 1 1

sociated cerebral involvement. Burkeet al. (1991) found neuro-developmentalhandicaps in 32 of 46 children (70%) withoptic nerve hypoplasia, but describedstructural central nervous system abnor-malities in 90% of them.

Our study found a sex ratio of 0.87,thus a predominance of females. The sexratio does not differ significantly fromthat in the age-specific population (1.06),but it differs from that previously re-ported by Zion (1976), who found a sexratio of 1.5 in a review of the existing sig-nificant literature in English on the sub-ject.

As stated earlier, some aetiological fac-tors have been suggested in the literature,notably maternal diabetes and maternalalcoholism. Landau et al. (1998) foundthat 8.8% of children of mothers withdiabetes had optic nerve hypoplasia. Inour study, only one mother was diabetic.Children of diabetic mothers present su-perior segmental hypoplasia with normalor very limited decrease in visual function(Peterson & Walton 1977; Kim et al.1989) and the material we present con-sists of children with visual impairment.We have no data on maternal alcoholism,but none of the medical histories of the100 cases scrutinized contained notes onalcohol or drug abuse. Optic nerve hypo-plasia has been estimated to occur in asmany as 48% of infants with fetal alcoholsyndrome (FAS) (Strömland 1987).Given an incidence of FAS of 1/600(Olegård et al. 1979), this would meanthat the rate of optic nerve hypoplasia as-sociated with FAS would be 80 per 100000 births, a rate which does not seemrealistic even if mild forms are included.

We found an increased risk of opticnerve hypoplasia in infants born ofsmoking women. The association be-tween maternal smoking and infant con-genital malformation is much debatedbut, at least for some conditions, associ-ation exist which are probably causal(Källen 2001). The association betweenmaternal smoking and optic nerve hypo-plasia is dose-dependent and remainsafter stratification for birth weight, indi-cating a direct toxic effect.

Some authors have suggested that mat-ernal drug use can cause optic nerve hy-poplasia, generally based on single casereports. In our study, we found relativelyrare drugs occurring more than once fortwo groups: drugs used for treatment ofinfertility (three cases) and drugs used fortreatment of depressive disease (fourcases). These findings may be random,

but if not, they explain only a small num-ber of cases. During 1995–99, antide-pressant drug use in early pregnancy wasreported by seven per 1000 pregnantwomen; the finding that three mothers ofinfants with optic nerve hypoplasia usedantidepressants during pregnancy istherefore noteworthy, but more data areneeded in order to verify or reject an as-sociation. One woman reported the useof the antifungal drug griseofulvin, whichhas been associated with birth defects. Inthe 1995–99 Medical Birth Registry, onlyone woman reported use of griseofulvin.Table5 shows exposures for commonlyused drugs such as antibiotics, anti-asth-matics, and analgesics that mirror theiruse in the general population.

Other maternal characteristics foundto be risk factors for optic nerve hypo-plasia include young maternal age (ir-respective of parity) and low parity (ir-respective of age). Some previous studies,all of which were smaller than the presentstudy, have also observed that youngmaternal age and first parity may beover-represented (Elster & McAnarney1979; Purdy & Friend 1979; Margalithet al. 1984; Robinson & Conry 1986).Young maternal age is a risk factor for afew congenital malformations, the mostwell-known of which is probably gas-troschisis (Källen & Lindham 1982). Op-tic nerve hypoplasia may be a malfor-mation to be added to this list.

Preterm birth, low birth weight, SGA,low Apgar scores, and caesarean sectionare all associated with an increased riskfor optic nerve hypoplasia. Assumingthat the condition does not originate dur-ing birth, but is a congenital malfor-mation formed much earlier, these associ-ations could be due to unfavourable con-ditions during development, affectingfetal growth, general fetal developmentand the risk of malformation. Early fetalmaldevelopment might be associatedwith an increased risk of peri-neonatalcomplication.

Another possibility is that optic nervehypoplasia actually results from pretermbirth and the complications associatedwith such births. Optic nerve atrophymight be more commonly expected, as aresult of neonatal ischaemia. It is possiblethat some misclassification has occurred,but it can hardly be significant enough toaccount for the results seen here.

Apart from causing visual impairment,optic nerve hypoplasia can be associatedwith endocrine disturbances (Hoyt et al.1970; Skarf & Hoyt 1984; Brodsky &

A O S 2002

304

Glasier 1993). Sudden death in childrenwith septo-optic dysplasia has been re-ported (Brodsky et al. 1997) and it is thusof vital importance that optic nerve hy-poplasia is recognized and that appropri-ate further investigation is performed.The increasing incidence and the in-triguing pathogenesis illuminates the im-portance of epidemiological studies andthe evaluation of possible risk factors aspart of preventive work.

AcknowledgementsThis study was supported by the Foundationfor the Visually Impaired in the former coun-try of Malmöhs and a donation from MalinMårtensson.

ReferencesBistner S, Rubin L & Aguine G (1973): Devel-

opment of the bovine eye. Am J Vet Res 34:7–12.

Björk Å, Laurell CG & Laurell U (1978): Bilat-eral optic nerve hypoplasia with normal vis-ual acuity. Am J Ophthalmol 86: 524–529.

Blohme J & Tornqvist K (1997a): Visual im-pairment in Swedish children. I. Registerand prevalence data. Acta OphthalmolScand 75: 194–198.

Blohme J & Tornqvist K (1997b): Visual im-pairment in Swedish children. II. Aetiolo-gical factors. Acta Ophthalmol Scand 75:199–205.

Blohme J & Tornqvist K (1997c): Visual im-pairment in Swedish children. III. Diag-noses. Acta Ophthalmol Scand 75: 681–687.

Brodsky MC, Conte FA, Taylor D, Hoyt CS &Mrak RE (1997): Sudden death in septo-op-tic dysplasia. Report of 5 cases. ArchOphthalmol 115: 66–70.

Brodsky MC & Glasier CM (1993): Opticnerve hypoplasia. Clinical significance of as-sociated central nervous system abnormali-ties on magnetic resonance imaging (pub-lished erratum appears in Arch Ophthalmol111: 491). Arch Ophththalmol 111: 66–74.

Burke JP, O’Keefe M & Bowell R (1991): Opticnerve hypoplasia, encephalopathy, and neur-odevelopmental handicap. Br J Ophthalmol75: 236–239.

Chan T, Bowell R, O’Keefe M & Lanigan B(1991): Ocular manifestations in fetal alco-hol syndrome. Br J Ophthalmol 75: 524–526.

Chan CC, Fishman M & Egbert PR (1978):Multiple ocular anomalies associated withmaternal LSD ingestion. Arch Ophthalmol96: 282–284.

Cnattingius S, Ericson A, Gunnarskog J &Källen B (1990): A quality study of a medi-cal birth registry. Scand J Soc Med 18: 143–148.

De Morsier G (1956): Agenesie du septum lu-

cidum avec malformation du tractus op-tique. La dysplasie septo-optique. SchweizArch Neurol Neurochir Psychiatr 77: 267–292.

Donat JFG (1981): Septo-optic dysplasia in aninfant of a diabetic mother. Arch Neurol 38:580–591.

Elster AB & McAnarney ER (1979): Maternalage re: septo-optic dysplasia. J Pediatr 94:162.

Good WV, Ferriero DM, Golabi M & KoboriJB (1992): Abnormalities of the visual sys-tem in infants exposed to cocaine. Ophthal-mol 99: 341–346.

Hansen E, Flage T, Rosenberg T, Rudanko SL,Viggosson G & Riise R (1992): Visual im-pairment in Nordic children. III. Diagnoses.Acta Ophthalmol 70: 597–604.

Hellström A, Wiklund L, Svensson E, Alberts-son-Wikland K & Strömland K (1999): Op-tic nerve hypoplasia with isolated tortuosityof the retinal veins. Arch Ophthalmol 117:880–884.

Hoyt CS & Billson FA (1978): Maternal anti-convulsants and optic nerve hypoplasia. BrJ Ophthalmol 62: 3–6.

Hoyt C & Good W (1992): Do we really under-stand the difference between optic nerve hy-poplasia and atrophy. Eye 6: 201–204.

Hoyt WF, Kaplan SL, Grumbach MM & Glas-er J (1970): Septo-optic dysplasia and pitu-itary dwarfism. Letter. Lancet 2: 893–894.

Jan JE, Robinson GC, Kinnis C & MacLeodPJM (1977): Blindness due to optic-nerveatrophy and hypoplasia in children: an epi-demiological study. Dev Med Child Neurol19: 353–363.

Källen B (1995): A birth weight for gestationalage standard based on data in the SwedishMedical Birth Registry 1985–89 Europ J Ep-idemiol 11: 601–606.

Källen K (2001): Maternal smoking and con-genital malformations. Fetal Maternal MedRev 13: 63–86.

Källen B & Lindham S (1982): A women’sbirth cohort effect on malformation rates.Int J Epidemiol 11: 398–401.

Kim RY, Hoyt WE, Lessell S & Narahara MH(1989): Superior segmental optic nerve hy-poplasia: a sign of maternal diabetes. ArchOphthalmol 107: 1312–1315.

Lambert SR, Hoyt CS & Narahara MH(1987): Optic nerve hypoplasia. SurvOphthalmol 32: 1–9.

Landau K, Djahanschahi Bajka J & Kirchsch-loger BM (1998): Topless optic disks inchildren of mothers with type 1 diabetesmellitus. Am J Ophthalmol 125: 605–611.

Margalith D, Jan JE, McCormick AQ, TzeWJ & Lapointe J (1984): Clinical spectrumof optic nerve hypoplasia: a review of 51 pa-tients. Dev Med Child Neurol 26: 311–322.

McKinna AJ (1966): Quinine induced hypo-plasia of the optic nerve. Can J Ophthalmol1: 261–265.

Mosier MA, Lieberman MF, Green WR &Knox DL (1978): Hypoplasia of the opticnerve. Arch Ophthalmol 96: 1437–1442.

Olegård R, Sabel KG, Aronsson M et al.(1979): Effects on the child of alcohol abuseduring pregnancy - retrospective and pros-pective studies. Acta Paediatr Scand Suppl275: 112–121.

Patel H, Tze WJ, Crichton JU, McCormickAQ, Robinson GC & Dolman CL (1975):Optic nerve hypoplasia with hypopituitar-ism. Am J Dis Child 129: 175–180.

Peterson RA & Walton DS (1977): Optic nervehypoplasia with good visual acuity and vis-ual field defects: a study of infants of dia-betic mothers. Arch Ophthalmol 95: 254–258.

Purdy F & Friend JCM (1979): Maternal fac-tors in septo-optic dysplasia. J Pediatr 95:661.

Riise R, Flage T, Hansen E, Rosenberg T, Ru-danko SL, Viggosson G & Warburg M(1992): Visual impairment in Nordicchildren. I. Nordic registers and prevalencedata. Acta Ophthamol 70: 145–154.

Robinson GC & Conry RF (1986): Maternalage and congenital optic nerve hypoplasia: apossible clue to aetiology. Dev Med ChildNeurol 28: 294–298.

Rosenberg T, Flage T, Hansen E, Rudanko S-L, Viggosson G & Riise R (1992): Visual im-pairment in Nordic children. II. Aetiologicalfactors. Acta Ophthalmol 70: 155–164.

Skarf B & Hoyt CS (1984): Optic nerve hypo-plasia in children. An association withanomalies of the endocrine and CNS. ArchOphthalmol 102: 62–67.

Strömland K (1987): Ocular involvement in thefetal alcohol syndrome. Surv Ophthalmol31: 277–284.

Strömland K & Pinazo-Duran MD (1994):Optic nerve hypoplasia: Comparative effectsin children and rats exposed to alcohol dur-ing pregnancy. Teratology 50: 100–111.

Taylor D & Stout A (1997): Optic nerve: con-genital anomalies. In: Taylor, D (ed). Pedi-atric Ophthalmology, 2nd edn. BlackwellScience Ltd, London. 660–700.

Zeki SM & Dutton GN (1990): Optic nervehypoplasia in children. Br J Ophthalmol 74:300–304.

Zion V (1976): Optic nerve hypoplasia.Ophthalm Seminars 1: 171–196.

Received on March 19th, 2001.Accepted on February 26th, 2002.

Correspondence:Kristina Tornqvist, MDDepartment of OphthalmologyUniversity Hospital of LundSE-221 85 LundSwedenTel: π46 46 17 20 74Fax: π46 46 21 15 074e-mail: kristina.tornqvist/oft.lu.se