Articles International Trends in Rates of Hypospadias and Cryptorchidism Leonard J. Paulozzi National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA 30341-3724 USA Res s fm sev E ios and te United Stawt pu repors of icreasi rates of hy diss dngdid e .0, 9`70s, and' 1980s. ts of i rae of cryp- torcidim hve cme rimrilyfro Enland Inrecnt yw, hes reprtshav bcm one focus ofthedebateove endocrine disption.Thisstudy examinesmorerecent datafro alrg number ~~~~~~~~~~ibe Bir lii+ whthr hee regogapicpaten to any obevd incess Th ICDS hedqatran individual system provd de d. Systems were c i ugrs domestic product in H i ewe mtm i tw and inSadTe iease ld iaa 198 ere not seen i less ffluen na ion.Cytrhds`rtswr vial fo 1sytm.cl:ear. increase in thisanomalywereseenintw U.& systems-andin th e SouthAmericansystem,but ot elsewhr e ~~~~~~~~~~~~~~~~~~Q .E.i. M . ._ i. *~~~~~~~~~~- Pnw.M. -o.. treds n hyopdaI osbe""cue icudedmogahccags n.norn disrup- tion amog ohers Keywri sbnrmalty ryptrchiism endcrin, gnita, hyospdias,tsi Within the past 5 years, researchers have hypothesized that some natural or manufac- tured agents are disrupting normal endocrine function in humans and animals, with particular emphasis on male reproduc- tive effects (1-8). This hypothesis attempts to unify and explain worrisome trends in measures of male reproductive health as the effects of estrogenic or antiandrogenic chemicals. Among the most frequently cited trends, along with trends in sperm count and testicular cancer, are increases in the male genital birth defects of hypospadias and cryptorchidism (3-5). These defects represent mild degrees of femininization. Hypospadias occurs when the urethral opening is displaced toward the scrotum. Cryptorchidism is a condi- tion in which one or both testicles do not descend into the scrotum. Increasing rates of these two anomalies have been reported within the past 25 years by a number of authors (9-19) (see Table 1). The increases that have been cited in support of the endocrine disrup- tion hypothesis occurred for the most part in the 1960s and 1970s. They also derive from a small number of countries in North America and Europe. I have compiled information on worldwide trends in these anomalies to provide more contemporary and complete information for the ongoing debate on endocrine disruption. Table 1. English-language published reports of upward trends in the prevalence of hypospadias and cryp- torchidism Year Location Birth years covered Reference Hypospadias 1975 Norway 1967-1973 Bjerkedal and Bakketeig (9) 1979 Sweden 1965-1977 Kallen and Winberg (10) 1982 Sweden 1965-1979 Kall6n and Winberg (11) 1985 England and Wales 1964-1983 Matlai and Beral (12) 1985 Hungary 1971-1983 Czeizel (13) 1986 Denmark 1970-1981 Kill6n et al. (14) 1991 United States 1974-1988 ICBDMS (15) 1991 Strasbourg 1982-1988 ICBDMS (15) 1991 Italy 1981-1988 ICBDMS (15) 1997 United States 1970-1993 Paulozzi et al. (16) 1997 Atlanta 1968-1995 Paulozzi et al. (16) Cryptorchidism 1984 England and Wales 1952-1977 Chilvers et al. (17) 1985 England and Wales 1969-1983 Matlai and Beral (12) 1986 England (Oxford) 1950s vs. 1984-1985 JRHCSG (18) 1992 England (Oxford) 1950s vs. 1984-1988 JRHCSG (19) Abbreviations: ICBDMS, International Clearinghouse for Birth Defects Monitoring Systems; JRHCSG, John Radcliffe Hospital Cryptorchidism Study Group. Methods The International Clearinghouse for Birth Defects Monitoring Systems (ICBDMS), a nongovernmental organization of the World Health Organization, collects rates of select- ed birth defects from member programs. To be a part of the ICBDMS, member pro- grams must be actively engaged in the sys- tematic and continuous collection of birth defects cases. The ICBDMS does not accept data from programs that only passively receive and report health statistics data from administrative sources. The Appendix lists the size and base (population or hospital) of each program as of 1990 (15). Hospital- based systems calculated their rates based on all births occurring in participating hospi- tals. Population-based systems used all births in a given geographic area. In late 1997, the ICBDMS headquarters in Rome provided the latest hypospadias and cryptorchidism birth prevalence rates per 10,000 total births from 29 birth defects registries in 21 countries for all years avail- able. In many cases, I received information for additional years via personal communi- cation with the registrars. Registrars also provided information useful in the interpre- tation of changes in their rates over time. The ICBDMS has defined both hypospa- dias and cryptorchidism, but the extent of adherence to those definitions by participat- ing programs is not known. Almost all coun- tries included defects among stillbirths, but the definition of a stillbirth varied. Prenatal diagnoses that were followed by pregnancy termination were not counted in the rates. Both isolated defects and defects found as part of syndromes are included. The number of cases per year of hypospadias ranged from 12 in the Northern Netherlands system to nearly a thousand in England and Wales. Address correspondence to LJ. Paulozzi, Division of Birth Defects and Pediatric Genetics, National Center for Environmental Health, Centers for Disease Control and Prevention, Mailstop F-45, 4770 Buford Highway NE, Atlanta, GA 30341-3724 USA. I thank the ICBDMS scientists making individual contributions to the data and its interpretation: P. Lancaster, S.W. Wen, B. Sibbald, W. Yanqiao, Z. Jun, A. Sipek, K. Helweg-Larsen, B. Botting, A. Ritvanen, E. Robert, J. Goujard, C. Siffel, Z. Johnson, P. Merlog, G. Cocchi, P. Mastroiacovo, R Tenconi, Y. Sumiyoshi, 0. Mutchinick, B. Borman, M. Cornel, A. Daltveit, M. da Graqa Dutra, M.L. Martinez-Frias, B. Kalldn, and C. Torfs. Received 29 October 1998; accepted 29 December 1998. Environmental Health Perspectives * Volume 107, Number 4, April 1999 297 Articles * Paulozzi 4 0 --- -- -- -- - -- -- -.'.e.ij'.' , .',: ili ... i.. ' . 3Atl Aanta !(severe} _ ll| l1 G l! l m 35 California isevere) Stts group. 30 S ___ S2 5 ............... 20iW0E |i f 4 65 6. 6 67.H 68 69 70j: 712^ s 72 73 74 757 77 9 al .82. 83 84 858 8 8*9 1 2939 5 69 IN ?'State-s3 group.4 40 35 o I S iK b. m, CD A= 30 25 20 15 10 5 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 Year of birth Figure 2. Hypospadias rates by system and year of birth (1974-1996), 3-year moving averages, Commonwealth group. 40...... 35 -U Finiand 25 -' =10 64865 66867 8869 7071 72 7374 75 7677 7879 90 81 8268384 85 8887 8889 9091 92W934 95 96 97 Year of birth Figure 3. Hypospadias rates by system and year of birth (1967-1996), 3-year moving averages, Scandinavia group. Three ICBDMS members, the California Birth Defects Monitoring Program, the Italy-Northeast registry, and the English reg- istry (after 1989), collected data only on the more severe types of hypospadias (i.e., penile, scrotal, or perineal), sometimes known as second or third-degree hypospa- dias, wherein the meatal opening is proximal to the glans of the penis. These types of hypospadias will be referred to as `severe` hypospadias in this paper. To determine whether trends depended on the degree of industrialization and to facilitate presentation, registries were grouped into categories loosely based on their country's gross domestic product (GDP) in 1984, chosen as the middle year of the bulk of the data. GDP was obtained from a University of Toronto online data- base (20). The U.S. group consisted of three American registries and had the high- est GDP ($15,900 U.S.) in 1984. The Commonwealth group consisted of the most affluent nations from English-speaking countries (Australia, Canada, New Zealand), all of which had GDPs in excess of $12,000 U.S. The Scandinavian group induded the four Scandinavian countries, whose GDPs were similar to those in the Commonwealth nations. The Northern Europe and Japan group (England, France, Japan, The Netherlands) had GDPs slightly lower than Scandinavia. The Mediterranean and Ireland group (Ireland, Israel, Italy, Spain) had yet lower GDPs, in the range of $5,500-8,000 U.S. The least affluent nations group (GDP less than $5,500 U.S.) included systems from two Eastern European nations, Hungary and Czechoslovakia, and the Chinese and Latin American systems. I calculated 3-year moving averages to smooth the trends. An "increase" in a trend in this report is based only on visual inspection, and not on statistical testing. Results The highest and lowest rates reported by individual systems during the period under study sometimes varied by a factor of three or more for both hypospadias and cryp- torchidism. However, few systems showed monotonic, unbroken upward or down- ward trends; trends typically reversed direc- tion at least once during the period of years under study. Hypospadias. Nationwide data from the Birth Defects Monitoring Program of the Centers for Disease Control and Prevention (CDC) showed an upward trend in hypospadias beginning in 1970 (Fig. 1). A more discontinuous upward trend began in 1968 in the CDC's Atlanta, Georgia, sur- veillance system. Severe hypospadias in the Atlanta system increased from 1982 to Volume 107, Number 4, April 1999 . Environmental Health Perspectives 298 Articles * Trends in rates of hypospadias and cryptorchidism 1985 and then leveled off. Rates from the California Birth Defects Monitoring Program for severe hypospadias showed no upward trend. In the Commonwealth group (Fig. 2), each system showed both short upward and short downward excursions and very little net change. Increases in the Australian and Canadian rates and rates in the Canadian province of Alberta were restricted to the late 1980s, whereas rates in the province of Ontario and in New Zealand were down or unchanged through the 1980s and early 1990s. Scandinavian countries, with the excep- tion of Sweden, show overall upward trends, with rates approximately doubling in Norway and Denmark during the 1970s and 1980s (Fig. 3). Norwegian rates declined somewhat in the 1990s. The Finnish registry reports that their increase was restricted to the mild form of hypospadias (first degree). In the northern Europe and Japan group (Fig. 4), all but the northern Netherlands registry showed some net increase. Rates dropped sharply in the northern Netherlands system during the 1980s. The Mediterranean and Ireland systems (Fig. 5) include only one system with an increase, the Italian Multicentric Register of Congenital Malformations (IPIMC). The IPIMC suggests that its increase may have been secondary to a special case-control study of hypospadias launched during this interval. The Italy-northeast system, which records only severe hypospadias, showed a decrease. Rates in the Israeli system made wide upward and downward excursions. Among the least affluent nations (Fig. 6), rates were generally stable since 1980, with the exception of the Czechoslovakian system, which registered an increase. Cryptorchidism. Few countries in the ICBDMS had data on cryptorchidism. Among U.S. and Commonwealth systems combined (Fig. 7), the U.S. national rates increased during the 1970s and 1980s, whereas the U.S.-Atlanta system began to increase in 1970, rose sharply in 1985, and dedined equally sharply by 1994. This peak corresponds to a 10-year period during which a more inclusive case definition was in effect in Atlanta. Rates in the Canadian national system increased until about 1980 and then stabilized. The Canadian provin- cial systems of Alberta and Ontario report- ed declines, at least since 1985. The Norwegian system, the only Scandinavian system collecting cryptorchidism rates, shows no consistent trend between 1974 and 1996 (Fig. 8). The same can be said of the data from the France-Paris system (Fig. 9), whereas English rates dropped sharply around 1990, contemporaneous with the introduction 40 35 % 30 .,0 - 25 1 20 d C 15 = 10 5 B4 85 88 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 8283 84 85 88 87 89 90 91 92 93 94 95 9 97 Year of birth Figure 4. Hypospadias rates by system and year of birth (1964-1996), 3-year moving averages, Northern Europe and Japan group. Abbreviations: CE, central east; N, north. 50 45 40 ca 35 15 :a E 25 Cd CDC 20 S m 10 5 0 64 65 66 67 68 89 70 71 72 73 74 75 76 77 78 79 80 81 8283 84 85 " 87 88 89 90 91 92 93 94 95 8 97 Year of birth Figure 5. Hypospadias rates by system and year of birth (1974-1997), 3-year moving averages, Mediterranean and Ireland group. Abbreviations: IPIMC, Italian Multicentric Register of Congenital Malformations; IMER, Emilia-Romagna Registry of Congenital Malformations; NE, northeast. of an "exclusion list." The Hungarian system rates have dedined from an early peak, while the South American rates increased overall and since 1985 (Fig. 10). Discussion Review of data from 29 registries that monitor a total of 4 million births per year around the world reveals wide intercountry variation in rates of hypospadias and cryp- torchidism. Given differences in registry methods, genetic variation, and other fac- tors, the rates themselves are not directly comparable. The primary value of this data is what it shows about changes within sys- tems in recent decades. The data suggests an increase in report- ed rates of hypospadias during the 1970s and 1980s in two United States systems and in Scandinavia and Japan. Rates from other nations increased only in one Italian system (IPIMC), where an artifact is sus- pected, and in the Israeli system, which is the smallest system and the system showing the most unstable rates of hypospadias over time. The absence of an increase is perhaps most notable in Canada, whose society is similar to that of the United States. Among all systems showing an increase, rates tend- ed to level offafter 1985. There is no indication of a generalized increase in cryptorchidism rates over time Environmental Health Perspectives * Volume 107, Number 4, April 1999 299 Articles * Paulozzi JU * 25 20 _ CL 15 0 64 85 66 67 68 69 70 71 72 7374 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 Year of birth Figure 6. Hypospadias rates by system and year of birth (1970-1996), 3-year moving averages, least afflu- ent nations group. -U- Canada (Alberta) 40 .--- Canada (Ontario) 35 * 20 64 65 66 67 66977172 73 74175 16 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 9697 Yeur of birth Figure 7. Cryptorchidism rates by system and year of birth (1968-1997), 3-year moving averages, United States and Commonwealth groups. since 1970, although data on this defect is much more limited. Two U.S. systems show marked increases, but the data from the Atlanta system is difficult to interpret because of coding changes. Since 1985, rates in most systems have actually declined. A number of factors may account for reported changes in these rates. Chief among them are artifacts. One possible explanation is that the definition of hypospadias may have changed over time to include more minor degrees of deviation from the normal position of the urethral opening on the tip of the penis. There is no anatomical marker that defines when nor- mal variation stops and first-degree hypospa- dias begins. Slight degrees of deviation are much more common than more proximal meatal positions (21), and a subtle change in the case definition could have produced a large change in overall rates. There is conflicting evidence on whether the case definition of hypospadias has indeed loosened to include more of the milder, first-degree cases. Previously pub- lished data from the Atlanta registry indi- cated that the percent of first-degree cases did not increase over time (16). In con- trast, the Finnish registry communicated that the percent of more serious degrees of hypospadias declined as overall rates increased. Moreover, the California and the northeast Italy programs have shown no increase in rates of severe hypospadias. Severe hypospadias is much less likely to be affected by changes in definition because it has clearer anatomical boundaries. Another possible explanation for the increase is gradual improvement over time in physician documentation of hypospadias. Because the foreskin is used in some surgical procedures to repair hypospadias and circum- cision must be deferred if hypospadias is pre- sent, medicolegal considerations may increas- ingly cause physicians who perform circumci- sion to examine the penis carefully. They may therefore be referring more boys to urologists. Increasing numbers of such referrals may increase the number and/or prominence of diagnoses of hypospadias in medical records, thereby improving the chances of detection by a surveillance program. The same artifacts could explain the increases in cryptorchidism rates noted in some of the systems. In particular, cryp- torchidism may be sought more aggressive- ly now because of the strong evidence accu- mulated over the past 20 years that unde- scended testicles are likely to become can- cerous (22) and because of the standard practice of removing them early in life in hopes of reducing this risk. A second hospi- talization for orchidopexy in infancy may double the chances of the anomaly being registered in a surveillance system. For both hypospadias and cryp- torchidism, it is also conceivable that rates after 1985 were affected by literature pub- lished during the 1980s describing increas- es in these anomalies. Perhaps criticism of the reports led to a tightening of case defin- itions in some systems. It is noteworthy that of the five registries publishing hypospadias increases by 1986 (Table 1), only Denmark reported any further increase in subsequent birth years. Other, nonartifactual explanations have been proposed to explain the increasing hypospadias rates in Europe reported earli- er. Initially, it was hypothesized that increasing use of steroid-containing med- ications by pregnant women might be responsible (23). However, the consensus now seems to be that the risk from such preparations is exaggerated, and the preva- lence of their use is not great enough to account for the increase (12,24,25). Alternatively, evidence of increased risk of hypospadias among couples of reduced fertility has produced speculation that an increasing proportion of such couples among all parents could account for an increasing trend in this anomaly (25,26). However, the magnitude of the risk for relatively infertile couples (26), combined with their low preva- lence among the population of all parents, does not seem sufficient to account for the large observed increases in some registries. Volume 107, Number 4, April 1999 * Environmental Health Perspectives 300 Articles * Trends in rates of hypospadias and cryptorchidism 25m * 64 65 6667 68 69 707172 737475 7677 78 798081 8283848a58687 88 89909s192 9394959697 Year of birda FigureS8. Cryptorchidism rates by year of birth (1974-1996), 3-year moving averages, Scandinavia group. 25 15~_ 64 65 6B 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 Year of birth Figure 9. Cryptorchidism rates by sse n year of birth (1974-1996), 3-year moving averages, Sadnavagorther Europe andEJapandgop 2o -UM- France (Paris) aL. S 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 Year of birth Figure 9. Cryptorchidism rates by system and year of birth (1979-1995), 3-year moving averages, northern Europe and Japan group. -*-Hungary 20 U- South America Cd 16 .5 0 84 85 66 67868 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 948596 97 Year of birdh Figure 10. Cryptorchidism rates by system and year of birth (1967-1996), 3-year moving averages, least affluent nations group. Conclusion There is some evidence for an increase in hypospadias rates concentrated in more affluent nations. That increase may have ended in the mid-1980s. More registries that experienced increasing trends in hypospadias should report how their per- centage of severe cases has changed over time. If an increase in all degrees of hypospadias is reported in more surveil- lance systems, more in-depth investigation will be warranted. Assuming these upward trends are real and assuming exogenous agents are responsi- ble, the relevant exposures may be more common in highly industrialized countries. Those exposures (or their body burdens) that may have stabilized since 1985 might also be the most logical ones to pursue among all potential environmental exposures. Although it is important to examine these trends broadly, it is unlikely that fur- ther inspection of international trends alone will shed additional light on the question of endocrine disruption as a cause of birth defects. Such descriptive analysis is provocative, but more sophisticated study designs should be sought. REFERENCES AND NOTES 1. Carlsen E, Giwercman A, Keiding N, Skakkebaek NE. Declining semen quality and increasing incidence of testicular cancer: is there a common cause? Environ Health Perspect 103(suppl 7):137-139 (1995). 2. Sharpe R, Skakkebaek NE. Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet 341:1392-1395 (1993). .Appendi Characteritcs of Inernationl Clearinghoue for Birth Defects. M oiniS ICBDMMS). sincuded in dais ;studyas of 19 Annual ICBDMSprogram births(thousands) Base Australia ...250 P .Can da 215 P * Ca.nada (Alberta } .38 P 'Canada (Ontrio) 146 P China 496 P China. (Szeahuan) 65 H Czea.Xch Repubi:.l .i i .c . ...;.130 P *Denmark : 60 P England and Wales 700 P Finland 64 P France (central, east) 90 P ; France (Paris 40 P Hungary 125 P Ireland 19 P Israel 18 H Itay (IMER) 23 P Italy (IPIMC) 140 :H Italy (no 57 P 'Japan 115 H Mexico 50 H * :Neotherlands (north) 19 P NZean 55 P Norway ~ 55 P South America 215 H Spain 60 H Sweden 120 P United States 575 H United States (Atlanta, GA);: 40 P United States (Califomia) 269 P Total 4,249 Abbreviations: 1, hospital-basd; P, population-based; IMER, Emilia-Romagna Registry of Con:genital Malformations; IPiMC, tliain Multicentric Register of Congenial Malformations. 3. Giwercman A, Carisen E, Keiding N, Skakkebaek NE. Evidence for increasing incidence of abnormalities of the human testis: a review. Environ Health Perspect 101(suppl 21:65-71 (1993). 4. Giwercman A. Declining semen quality and increas- ing incidence of abnormalities in male reproductive organs-fact or fiction? Hum Reprod 10suppI 1):158-164 (1995). 5. Jensen TK, Toppari J, Keiding N, Skakkebaek NE. Do environmental estrogens contribute to the decline in male reproductive health? Clin Chem 41:1896-1901 (1995). 6. Guillette U Jr. Endocrine disrupting environmental contaminants and developmental abnormalities in embryos. Hum Ecol Risk Assess 1:25-36(1995). 7. Guillette U Jr. Endocrine-disrupting environmental contaminants and reproduction: lessons from the study of wildlife. In: Women's Health Today: Perspectives on Current Research and Clinical Practice (Popkin DR, Peddle LJ, eds). New York: Parthenon Publishing Group, 1994;201-207. 8. Facemire CF, Gross TS, Guillette U Jr. Reproductive impairment in the Florida panther: nature or nurture? Environ Health Perspect 103(suppl 4):79-86 (1995). 9. Bjerkedal T, Bakketeig LS. Surveillance of congenital malformations and other conditions of the newborn. Int J Epidemiol 4:31-36 (1975). 10. Killen B, Winberg J. Dealing with suspicions of mal- formation frequency increase. Acta Paediatr Scand Suppl 275:66-74 (1979). 11. Killen B, Winberg J. An epidemiological study of hypospadias in Sweden. Acta Paediatr Scand Suppl 293:3-21 (1982). Environmental Health Perspectives * Volume 107, Number 4, April 1999 301 Articles * Paulozzi 12. Matlai P, Beral V. Trends in congenital malforma- tions of external genitalia. Lancet 1:108 (1985). 13. Czeizel A. Increasing trends in congenital malforma- tions of male external genitalia. Lancet 1:462-463 (1985). 14. Kallen B, Bertollini E, Castilla A, Czeizel A, Knudsen ML, Martinez-Frias LM, Mastroiacovo P, Mutchinick 0. A joint international study on the epidemiology of hypospadias. Acta Paediatr Scand Suppl 324:1-52 (1986). 15. Congenital malformations worldwide: a report from the International Clearinghouse for Birth Defects Monitoring Systems. New York:Elsevier Science Publishers, 1991. 16. Paulozzi LJ, Erickson JD, Jackson RJ. Hypospadias trends in two US surveillance systems. Pediatrics 100:831-834 (1997). 17. Chilvers C, Pike MC, Forman D, Fogelman K, Wadsworth MEJ. Apparent doubling of frequency of undescended testis in England and Wales in 1962-81. Lancet 1:330-333 (1984). 18. Cryptorchidism: an apparent substantial increase since 1960. John Radcliffe Hospital Cryptorchidism Study Group. Br Med J 293:1401-1404 (1986). 19. Chilvers CED. Cryptorchidism: a prospective study of 7,500 consecutive male births, 1984-8. Arch Dis Child 67:892-899 (1992). 20. CHASS Data Centre at the University of Toronto. Penn World Tables, Real GDP per capita. Available: http://datacentre.chass.utoronto.ca [cited 10 March 1998]. 21. Fichtner J, Filipas D, Mottrie AM, Voges GE, Hohenfellner R. Analysis of meatal location in 500 men: wide variation questions need for meatal advancement in all pediatric anterior hypospadias cases. J Urol 154:833-834 (1995). 22. Schottenfeld D. Testicular cancer. In: Cancer Epidemiology and Prevention (Schottenfeld D, Fraumeni JF, eds). New York:Oxford University Press, 1996;1207-1219. 23. Aarskog D. Maternal progestins as a possible cause of hypospadias. N EngI J Med 300:75-78 (1979). 24. Kallen B, Martinez-Frias ML, Castilla EE, Robert E, Lancaster PAL, Kringelbach M, Mutchinick OM, Mastroiacovo P. Hormone therapy during pregnancy and isolated hypospadias: an international case-control study. Int J Risk Safety Med 3:183-198 (1992). 25. Czeizel A, Toth J. Correlation between the birth prevalence of isolated hypospadias and parental subfertility. Teratology 41:167-172 (1990). 26. Kallen B, Castilla EE, Kringelback M, Lancaster PAL, Martinez-Frias ML, Mastroiacovo P, Mutchinick 0, Robert E. Parental fertility and infant hypospadias: an international case-control study. Teratology 44:629-634 (1991). 302 Volume 107, Number 4, April 1999 * Environmental Health Perspectives