Higher than expected prevalence of congenital cryptorchidism in Lithuania: a study of 1204 boys at birth and 1 year follow-up

Abstract

BACKGROUND: Cryptorchidism at birth is one of the symptoms of testicular dysgenesis syndrome (TDS). The aim of the study was to detect prevalence of cryptorchidism in Lithuanian newborn boys. METHODS: A total of 1204 consecutively born boys were examined within the first days after birth in one regional hospital. Boys cryptorchid at birth were reexamined 1 year later. RESULTS: The prevalence of cryptorchidism at birth was 5.7% (69 cases). Cryptorchidism was associated with low birth weight (P<0.0001), preterm delivery (P<0.0001), small gestational weight (P=0.03) and other congenital abnormalities of genitalia (P=0.0001). No correlation between cryptorchidism at birth and maternal age, birth order or mode of delivery was demonstrated in this study, but paternal body mass index <20 kg/m² was found to be a significant risk factor (P=0.001). The prevalence of congenital cryptorchidism at 1 year of age was 1.4%. CONCLUSIONS: We detected lower frequency of cryptorchidism at birth in Lithuanian boys than in Danes (9.0%), but higher than in Finns (2.4%). We had expected the frequencies in Lithuania and Finland to be relatively similar because the other symptoms of TDS (incidence of testicular cancer and semen quality) are close in these countries.

Introduction

Several epidemiological studies have demonstrated reduced semen quality, a high incidence of testicular cancer and congenital abnormalities of the human male reproductive system such as cryptorchidism and hypospadias in certain geographical areas (Scorer, 1964; Chilvers et al., 1984; Carlsen et al., 1992; John Radcliffe Hospital Cryptorchidism Study Group, 1992; Adami et al., 1994; Auger et al., 1995; Irvine et al., 1996; Andersen et al., 2002; Jorgensen et al., 2002; Punab et al., 2002; Boisen et al., 2004). The causes of these disorders are unknown, possibly involving both genetic and environmental factors (Norgil Damgaard et al., 2002; Ivell and Hartung, 2003; Kaleva and Toppari, 2003). However, it has been suggested that these abnormalities of the male reproductive system may be symptoms of a testicular dysgenesis syndrome (TDS) (Skakkebaek et al., 2001; Asklund et al., 2004). In particular, the association of cryptorchidism with testicular cancer and impaired fertility is well established (Berthelsen and Skakkebaek, 1983; Giwercman et al., 1988; Huff et al., 1991; Giwercman et al., 1993).

The highest frequency of men with symptoms of TDS has been found in Danish men, whereas there seems to be a lower frequency among Finnish men (Adami et al., 1994; Andersen et al., 2002; Slama et al., 2002; Boisen et al., 2004). Recently, the prevalence of cryptorchidism in Danish newborn boys has been shown to be almost four-fold higher than that in Finnish newborns. More especially, the mild forms appeared more frequently among the Danish boys (Boisen et al., 2004).

Some preliminary data on male reproductive health in Lithuania, which is geographically close to Nordic countries, are also available. A recent study of semen quality of young men from the general population in Lithuania demonstrated almost as good semen quality in these men as in Finnish men (Punab et al., 2002), and the incidence of testicular cancer in Lithuania is one of the lowest in Europe (Adami et al., 1994). Therefore, we undertook a study of Lithuanian newborn boys to further characterize the occurrence of TDS symptoms among Lithuanians and also to investigate whether they have a frequency of cryptorchidism at the same level as Finnish boys.

Materials and methods

This prospective cohort study was carried out at the Panevęžys City Hospital in Lithuania during the period October 1996 to November 1997. The Panevęžys county is in the northern part of the country. Some 320 800 inhabitants live in the county, and 119 000 of these live in Panevęžys city. Panevęžys is the fifth largest city in Lithuania and the county is one of the 10 similar administrative units of the country. Almost all child deliveries in the county take place in one of five hospitals. Eighty per cent take place in Panevęžys City Hospital and the remaining 20% are shared between four smaller hospitals in the county. We included boys born in the Panevęžys City Hospital during the study period. Except for three fathers born in Estonia, Latvia and Russia, all parents were born in Lithuania. This cohort was rather homogenous ethnically, because 96.6% of population in Panevęžys county, according the Statistics Department data (www.std.lt), are Lithuanians.

A total of 1226 boys were delivered at the Panevęžys City Hospital during the study period, and 1204 (98%) were examined by one investigator (R.T.P.) 1 or 2 days after birth. Twenty-two of the 1226 boys were not included in the study either due to their immediate referral to the university hospital for medical reasons (two cases), due to early discharge from the hospital or beacause the parents did not agree to participate in the study.

We used the same examination technique and definition of cryptorchidism as in the study on cryptorchidism in Denmark and Finland (Boisen et al., 2004): the examination took place in the postnatal wards, at normal room temperature (20–24 °C), with a boy in supine position. The position of the testis was recorded after its manipulation to the most distal position along the pathway of the normal anatomical descent without the forced traction. The position of each testis was classified as normal (normal scrotal and normal retractile testes are considered to be attributed to this definition) or undescended (high scrotal and suprascrotal, combined to one group, called prescrotal, inguinal and non-palpable testes).

If the boys were born before 37 completed weeks of gestation they were defined as preterm. Gestational age was assessed by antenatal ultrasonographic examination in pregnancy week 18–20 (80% of cases). In the remaining cases the data on the last menstrual period and, in addition, Dubowitz score postnatally (Dubowitz et al., 1970) were used to determine the gestational age. An abnormal pregnancy was recorded if threatened spontaneous abortion, gestational toxicosis, fetal hypoxia, multiple gestation or a combination of these had appeared. A low body weight (LBW) boy was defined by birth weight <2500 g. A sex-differentiated fourth degree polynomia (Marsal et al., 1996; Larsen, 2001)– as in the Nordic study of cryptorchidism (Boisen et al., 2004)– was used for calculation of the predicted weight according to the days of gestation. Small for gestational age (SGA) was defined if the body weight for gestational age was less than −24% from that predicted. Additional information was obtained from a standardized questionnaire that the mother of the newborn filled out by herself. The questionnaire included information on the working conditions (heat, vibration, any chemicals, car or truck driving), tobacco smoking and alcohol consumption, and previous or current diseases.

All boys who were cryptorchid at birth, except for one who died due to prematurity and multiple congenital abnormalities, were reexamined at 1 year of age under the same conditions as the newborns. Their body weight and body length in supine position was also measured.

Statistical analysis

The software packets STATISTICA 5.5 and R version 1.60 were used for statistical analysis of the data. χ²-tests and Fisher's exact tests were used for testing differences in proportions in tables. The effect of risk factors for congenital cryptorchidism is expressed by odds ratio (OR), which, along with a 95% confidence interval (CI), is estimated by conditional maximum likelihood (Agresti, 1990). P-values <0.05 were considered to be significant.

Results

Newborn boys

Of all 1204 examined newborns, 1135 (94.3%) had fully descended testes while 69 (5.7%) had one or both testes undescended. The incidence is higher than in Finland (2.4%; P=0.0001), but lower than in Denmark (9.0%; P=0.003) (Boisen et al., 2004). Unilateral cryptorchidism was detected in 37 (3.07%) and bilateral in 32 (2.66%) cases. A total of 101 undescended testis were classified according to their position: 71 (69.3%) were prescrotal, 11 (12.9%) inguinal and 19 (17.8%) non-palpable.

An increased frequency of cryptorchidism was detected in boys with LBW and preterm boys (both P<0.0001) (Tables I and II).

The prevalence of bilateral cryptorchidism was significantly higher in newborns with LBW (70.6%) than in boys with normal birth weight (30.5%) (OR 3.8; 95% CI 1.04–15.8; P=0.03). Difference between unilateral versus bilateral cryptorchidism was not dependent on gestational weeks and weight for gestational age. However, overall being born SGA was a significant risk factor for cryptorchidism (OR 3.46; 95% CI 1.00–9.63; P=0.03) (Table III).

In all, 21.7% (15 of the 69) of the cryptorchid newborns had associated abnormalities of genitalia: scrotal hypoplasia, hypospadias, micropenis, hydrocele or inguinal hernia (P<0.0001) (Table IV). None of these 15 boys had more than one associated abnormality.

The prevalence of cryptorchidism (8.8%) was significantly higher in newborns of mothers who had abnormalities during the pregnancy (threatened spontaneous abortion, gestational toxicosis, fetal hypoxia, multiple gestation or a combination of these) (OR 2.13; 95% CI 1.26–3.58; P=0.003) than in case of normal pregnancy (4.3%). Additionally, body mass index (BMI) of the father of <20 kg/m² was a significant risk factor for cryptorchidism (OR 5.99; 95% CI 1.63–18.3; P=0.001). Mode of delivery, birth order and other parental constitutional (age, height, weight, maternal BMI) factors were not found to be associated with the prevalence of cryptorchidism.

Prevalence of cryptorchidism in the sons of tobacco smoking mothers (7.8%) was not significantly higher than in the sons of those not smoking (5.1%) (OR 1.58; 95% CI 0.94–2.65; P=0.08). For mothers reported any harmful professional factor in her job (heat, vibration, any chemicals) associated with tobacco smoking, cryptorchidism prevalence in sons was 7.13% (OR 1.49; 95% CI 0.91–2.43; P=0.11); prevalence of cryptorchidism was 4.90% if no professional putative risk factors were reported. Concerning any harmful paternal professional factor on the risk of congenital cryptorchidism in their sons (heat, car or truck driving, any chemicals), no significant risk was detected if reported (OR 1.18; 95% CI 0.63–2.21; P=0.60).

One-year-old boys

When reexamined at 1 year of age, overall 51 (75%) of the 68 children with cryptorchidism at birth had achieved full testicular descent. LBW, preterm and SGA infants, and infants from an abnormal pregnancy appeared to have a higher chance of achieving spontaneous testicular descent during the first year of life than those who had not experienced the problems, but the difference was not statistically significant (Table V). If boys with cryptorchidism were small at 1 year of age, less than the 10th percentile of the normal weight and height, they had higher ORs of 3.17 (95% CI 0.85–12.0; P=0.06) and 4.75 (95% CI 0.88–28.0; P=0.04), respectively, of not achieving full testicular descent at 1 year of age.

Assuming that none of the ‘non-cryptorchid’ boys at birth developed cryptorchidism (ascensus testis) during their first year of life, the overall rate of cryptorchidism during the first year of life decreased form 5.7% at birth to 1.4%. Fourteen (1.2%) infants had unilateral cryptorchidism and only three (0.2%) bilateral. Cryptorchidism rate at 1 year of age was higher in LBW (OR 3.69; 95% CI 0.66–13.7; P=0.066), preterm (OR 2.61; 95% CI 0.47–9.59; P=0.14) and SGA (OR 2.49; 95% CI 0.05–17.1; P=0.35) boys, as compared with boys with normal weight, but the difference was not significant (Table VI).

Discussion

This prospective cohort study demonstrated a 5.7% prevalence of cryptorchidism in male infants born in Panevęžys City Hospital (Lithuania) during a study period of 13 months. The area and the population in the county, from which women come for delivery to the city hospital, are in all aspects similar to the rest of the country (climatic conditions, proportion of people employed in the industry and agriculture, etc.). Therefore, it is most likely that the population investigated in this hospital is representative for the whole country and the obtained results can be extrapolated to the whole population (www.std.lt).

The study design was comparable to that used in the recent studies in Denmark and Finland (Boisen et al., 2004). Highly variable prevalence of cryptorchidism in Western countries was reported on the basis of registry studies (Paulozzi, 1999; Toppari and Kaleva, 1999; Toppari et al., 2001). However, the studies have many methodological limitations. The most reliable data on the prevalence of cryptorchidism originate from the cohort studies in which the case definitions and examination techniques have been clearly described (Paulozzi, 1999; Toppari and Kaleva, 1999; Toppari et al., 2001). The study on the prevalence of cryptorchidism in Lithuania, as well as the study in Finland and Denmark, could be considered to be such investigations. Therefore, the difference in the prevalence of cryptorchidism between these countries, located closely to each other around the Baltic Sea, seems to be valid and real.

There were some differences in the study in the follow-up of the boys compared with Danish study (Boisen et al., 2004). Since our study was hospital based, no strict criteria for ethnic origin were used and all newborns born during the period October 1996 to November 1997 were included. In Denmark and Finland, only Danes and Finns were included in the study. Only those who were cryptorchid at birth were reexamined at 1 year of age, whereas in Denmark all boys, except those lost at follow-up, were reexamined at 3 months and at 18 months of age, and the overall prevalence of cryptorchidism assessed. In Finland, all boys were reexamined at 3 months, but at 18 months only cases and selected controls were reexamined. Thus, in our study only data on the prevalence of congenital cryptorchidism at 1 year of age are available. A direct comparison of the prevalence data at 3–18 months after birth in these three countries is complicated, but the numbers are rather close. The prevalence of congenital cryptorchidism 1 year after birth was 1.4% in our study, the 3 months prevalence of cryptorchidism in Denmark, assuming 100% testicular descent among cryptorchid boys lost at follow-up, was 1.8%, and in Finland 1.0%.

The cause of cryptorchidism in most cases is unknown, but preterm birth, low birth weight, especially if a child is small for gestational age, and other congenital malformations are known to be associated with cryptorchidism (Berkowitz et al., 1995; Akre et al., 1999; Weidner et al., 1999). Our data also confirm this association. Some epidemiological studies have found that maternal age, parity and mode of delivery could also be risk factors for cryptorchidism (Berkowitz et al., 1995; Akre et al., 1999). We detected a significant increase in cryptorchidism rate only in cases of abnormal pregnancy (owing to the small number of cases in each subgroup, a statistical analysis of these could not contribute with further information; data not shown). However, this hospital does not have a higher than average percentage of abnormal pregnancies and deliveries, and therefore these could not be the cause of the high prevalence of cryptorchidism in our study. The obstetric care in the hospital corresponds to the standards of second level non-university county hospital (internet site for the Lithuanian Ministry of Health: www.sam.lt). The prevalence of congenital cryptorchidism was higher in the cases when the father had a low BMI (<20 kg/m²). This finding is interesting, but should be evaluated cautiously; the number of cases was rather low and only larger studies will be able to fully elucidate such an association. However, new data on the reduced semen quality among under- and overweight men should be taken into account (Jensen et al., 2004). If underweight men more often have poor semen quality, one of the symptoms of TDS, there also might be some genetic link with the increased prevalence of cryptorchidism in their sons.

No factors important for cryptorchidism rate at birth (gestational age, birth weight, weight according to the gestational age, pregnancy abnormalities) were significantly related to the spontaneous descent of the testes. However, there was a tendency for a better chance of spontaneous testicular descent during the first year of life in boys with LBW, preterm, SGA infants and boys from abnormal pregnancy. The only estimated significant association was between weight of the child at 1 year of age and testicular descent rate; those children who reached normal weight at 1 year of age had a significantly higher chance for spontaneous testes descent than those who did not catch up weight.

The estimated prevalence of cryptorchidism at birth in Lithuania (5.7%) is higher than in Finland (2.4%), but lower than in Denmark (9.0%), (Boisen et al., 2004). As cryptorchidism is considered to be one of the symptoms of TDS (Skakkebaek et al., 2001; Asklund et al., 2004), these figures are not completely consistent with the recently available data on semen quality and testicular cancer in the Nordic–Baltic region (Jorgensen et al., 2002; Punab et al., 2002). Based on these data, it could be expected that the frequency of cryptorchidism in Lithuanian boys would be similar to that of Finnish boys. The reasons for the different frequencies of congenital cryptorchidism between the countries are unknown. As in the Danish and Finnish studies, the population investigated in our study was genetically homogenous. Thus, the genetic factors for which a causal relationship with cryptorchidism was suggested recently (Ivell and Hartung, 2003; Kaleva and Toppari, 2003) cannot be excluded. The different genetic factors in complex with environmental ones might modify the manifestation of TDS in different ethnic groups.

Recent data show that the potential semen quality and testicular cancer development is probably determined during the fetal period (Jorgensen et al., 1995; Moller and Skakkebaek, 1997), but clinically this is usually not detected before the third decade of life. Thus, the recent findings of good semen quality and low frequency of testicular cancer in Lithuania may reflect the intrauterine environmental situation more than 20 years ago. Cryptorchidism and hypospadias, the other two symptoms of TDS, can, however, be detected immediately after the birth, and therefore the prevalence of congenital cryptorchidism detected recently may better reflect the current environmental situation than semen quality and the incidence rate of testicular cancer. If the TDS hypothesis holds true, the question arises as to whether adverse changes of the environmental factors affecting male reproductive health have taken place in Lithuania. Further follow-up of male reproductive health in the populations in the countries surrounding the Baltic Sea region should be undertaken, and the role of environmental factors, including endocrine disrupters, as well as the role of genetic, lifestyle and other factors, should be further explored.

This work was supported by The Lithuanian State Science and Study Foundation, The Danish Research Council (grant No. 22-03-0198) and the European Commission (grant No. QLK4-CT-2002-00603).

References

Author notes

1Institute of Endocrinology, Kaunas University of Medicine, LT-50009 Kaunas, Lithuania, 2Department of Growth and Reproduction, Rigshospitalet, DK-2100 Copenhagen, 3Department of Biostatistics, University of Copenhagen, DK-2200 Copenhagen, Denmark and 4Departments of Physiology and Paediatrics, University of Turku, FI-20520 Turku, Finland