Aim: There is much debate as to whether the prevalence rates of cryptorchidism and hypospadias are increasing. To address this issue we investigated the birth prevalence of cryptorchidism and hypospadias in the northern region of England during the period 1993–2000.
Methods: Cases of cryptorchidism and hypospadias were identified from northern region hospital episodes statistics (HES). Trends in birth prevalence, based on the number of male live births, were assessed using linear regression.
Results: Prevalence was 7.6 per 1000 male live births for cryptorchidism and 3.1 per 1000 male live births for hypospadias. The orchidopexy rates for 0–4 year olds and 5–14 year olds were 1.8 and 0.8 per 1000 male population, respectively. The rates for hypospadias repair for 0–4 year olds and 5–14 year olds were 0.6 and 0.1 per 1000 male population, respectively. There was a statistically significant decreasing temporal trend for the corrective procedure in cryptorchidism of 0.1 per 1000 male population aged under 5 years per annum (95% confidence interval: −0.01 to −0.05, p<0.001), but no temporal change for the corrective procedure in hypospadias (p = 0.60).
Conclusion: HES data were of high quality for the study period. There was no significant change in the prevalence of surgically corrected hypospadias. However, there was an apparent decline in the prevalence of surgically corrected cryptorchidism that may reflect a decrease in the prevalence of the condition or may be due to a decrease in the rate of surgical intervention.
- birth prevalence
Statistics from Altmetric.com
There is controversy as to whether there is an increasing birth prevalence of cryptorchidism and hypospadias,1–3 which together with testicular cancer are components of the postulated testicular dysgenesis syndrome.4 Evidence suggests that there is geographical and temporal variation in the birth prevalence of these anomalies.5 The cause of this variation is unknown, although the speed of the reported increase in birth prevalence over time has been suggested to implicate environmental factors.6 Exposure to synthetic chemicals acting as endocrine disruptors in utero has been hypothesised as a possible explanation.7 Studies in animals have found effects of endocrine disruptors on subsequent male reproductive health, but there is a lack of direct evidence for a similar effect in humans.5 Hypospadias and cryptorchidism have also been linked with low birth weight,8,9 suggesting that fetal androgen dysfunction plays a role in aetiology.
Complete datasets of these anomalies are rare because their construction is difficult, with reported trends often hard to interpret due to incomplete ascertainment and varying inclusion criteria. Under-reporting of cases at birth leads to errors in ascertainment. Undescended testis at birth may descend naturally in the months after birth and ascertainment is further complicated by difficulty in distinguishing between undescended testes, ascending testis and retractile testis. Hypospadias may not be obvious in young babies. Consequently, reported temporal trends and comparisons between congenital malformations registries may to some extent reflect differences in the methods of case ascertainment and reporting. Hence, the absolute magnitude of any change in risk in male genital congenital anomalies is uncertain, as is the extent to which any such change in risk is a global or local phenomenon.
We report analyses of prevalence and temporal trends of cryptorchidism and hypospadias in northern England in patients diagnosed during the period 1993–2000. We have also recently reported an increasing trend in testicular cancer in young men, aged less than 25 years, from northern England.10
The study was based in the northern region of England, which is a mixture of several heavily populated urban areas and widespread rural communities. It has a predominantly Caucasian population of 3.1 million (ethnic minorities account for under 2%), with approximately 37 000 live births per year.
Males diagnosed with cryptorchidism or hypospadias whilst resident in the northern region of England were eligible for inclusion in this study. Cases were eligible for inclusion in this study only if the child had a diagnosis of either hypospadias or cryptorchidism and were included in the hospital episodes statistics (HES) for the north of England which were accessed via the North East Public Health Observatory (NEPHO). HES comprise clinical details provided by National Health Service (NHS) hospitals in England and for NHS hospital patients treated elsewhere. The quality and timeliness of the HES data were poor at the beginning of data collection but gradually improved. In 1987–1988, the first year that the system was in operation, coverage was only 88%, with only 75% of records coded for diagnosis. By the year 1993–4, coverage was 98% with 96% of records containing coded details of operative procedures.11
Cases of cryptorchidism and hypospadias were identified from these data by primary ICD codes (cryptorchidism 7525, Q530, Q531, Q532, Q538 and Q539; hypospadias 7526, Q540, Q541, Q542, Q543, Q548 and Q549).12,13 Details of corrective procedures were used to validate the diagnoses recorded in the HES data. Due to the limitations of ICD coding, it was not possible to clearly categorise the severity of these conditions based on ICD code only. However, it was possible to assess clinically significant cases using surgical procedure codes.14 Additional information on date of birth, maternal residence at the time of admission, and on dates and types of operative procedures were also abstracted from the HES data. These data were used to eliminate multiple entries on the dataset, which is episode rather than person based, and hence a person-based data set was constructed.
Episodes identified at birth were excluded under the assumption that if these cases were true undescended testis or clinically significant cases of hypospadias, they would be subsequently admitted for surgical corrective procedures. The quality of the data was assessed by examining missing values and inconsistent ICD codes in relation to surgical codes.
Birth prevalence per 1000 male live births was calculated based on birth estimates for the study area obtained from the regional Office for National Statistics birth tapes for the period 1993–2000. Only the operated cases and cases who were less than 15 years old on the day of admission were used for the calculation of birth prevalence. Data from 1985–1992 and 2001–2002 were excluded from this study population due to high levels of incomplete ascertainment. The rates of surgical procedures were calculated based on the estimates of mid-year populations, which were also obtained from the Office for National Statistics. The rate for all procedures (ie, orchidopexy and orchidectomy for cryptorchidism and all procedures under surgical codes M731, M736, M738, M739, M812 and M813 for hypospadias) was calculated. The orchidopexy rate was calculated for the purpose of comparison with previously published papers. The rate of operative procedures gave estimates for true undescended testis and also for the clinically significant form of hypospadias. Temporal trends in birth prevalence and the rate of surgical corrective procedure were assessed using linear regression. The statistical software package Stata, version 8 (Stata, College Station, TX, USA) was used for all statistical analyses.
There were minimal errors in recorded diagnoses as counter checked by relevant operative procedures. The operation procedures were mostly consistent with the ICD codes except for two cases of hypospadias which were operated under repair of epispadias. About 8% of cases of cryptorchidism and 7% of cases of hypospadias had missing or inaccurate postcodes. Less than 0.5% of cases of cryptorchidism and 8% of cases of hypospadias had missing operation dates.
A total of 1259 individual cases of cryptorchidism and 588 cases of hypospadias were identified from HES from 1993 to 2000. Only 104 (8%) cases of cryptorchidism and 103 (18%) cases of hypospadias had no corrective procedures coded.
The overall birth prevalence was 7.6 per 1000 male live births for cryptorchidism and 3.1 per 1000 male live births for hypospadias.
The orchidopexy rate (N08 and N09) was 1.8 per 1000 male population aged less than 5 years old and 0.8 per 1000 male population aged 5–14 years old. For all males aged 0–14 years, the orchidopexy (N08 and N09) rate was 1.1 per 1000 male population.
The rate of specific hypospadias corrective repair (M731) was 0.6 per 1000 male population aged less than 5 years old and 0.1 per 1000 male population aged 5–14 years. For all males aged 0–14 years, the rate for hypospadias repair (M731) was 0.3 cases per 1000 male population.
The exact age when cases were first diagnosed was not available. Age at the first operation was taken as a proxy for the date of the diagnosis being confirmed. The median age for the corrective procedures for cryptorchidism was 3.1 years and for hypospadias it was 2.3 years.
The estimated birth prevalence for cryptorchidism for the years 1993–2000 is shown in fig 1. Assuming a constant linear trend, the observed birth prevalence of cryptorchidism decreased at a rate of 0.08 cases per 1000 male live births per year across the study period (p<0.001).
The temporal trend in the birth prevalence of hypospadias is shown in fig 2. The birth prevalence of hypospadias increased at a rate of 0.05 cases per 1000 male live births per year (95% CI: 0.009 to 0.09, p = 0.02).
The trend in the orchidopexy rate over the 8-year study period is shown in fig 3. The orchidopexy rate decreased at a rate of 0.08 per 1000 per year for males aged under 5 years (p<0.001) and at a rate of 0.07 per 1000 for males aged 5–14 years (95% CI: −0.09 to −0.06, p<0.01).
The rates of hypospadias repair (M731) did not show a trend over the 8-year period for either age group (p>0.05) (fig 4).
Due to the dynamics of testicular descent, the true birth prevalence of cryptorchidism can only be known after the first year of life. Previous epidemiological studies have reported widely varying estimates of birth prevalence,1,15,16 probably reflecting challenges in case ascertainment. However, it is difficult to compare rates due to the inconsistency in methodology and inclusion criteria used both nationally and internationally. Within the north of England, there was no existing structured data capture system for cryptorchidism and hypospadias. However, we constructed such a dataset from the routinely collected HES database.
Cases of true undescended testis and hypospadias were estimated by assuming that only true undescended testis and the clinically significant cases of hypospadias would be operated on. Close examination of the dataset showed that there were minimal missing values and contradicting variables, and so it appeared to be of high quality.
The birth prevalence for cryptorchidism seen in this investigation is higher than the rates previously reported in England and Wales between 1981 and 198315 and is also higher than that in the recent report by Carbone and colleagues in Italy.16 The orchidopexy rate was 1.1 per 1000 male population aged 0–14 years. This is slightly lower than previously published data from Great Britain between 1992 and 1998. Toledano et al had reported the rate in the northern and Yorkshire region as 1.4 per 1000 male population aged 0–14 years in 1998.17 The rate of surgical corrective procedures for cryptorchidism was therefore decreasing in trend. As suggested by a previous report, as orchidopexy rate is a marker of the actual prevalence of undescended testis,17 this may indicate a real decrease in cryptorchidism. However, it must be acknowledged that the apparent decline in prevalence could be due to a trend towards less intervention in equivocal cases of undescended testes.
The birth prevalence for hypospadias (3.1 per 1000 male live births) was lower than that reported by Ahmed and colleagues.18 The Northern Congenital Anomaly Survey considered a combination of hypospadias and epispadias and also used total births as denominator. Between 1992 and 1997, for both conditions the Northern Congenital Anomaly Survey reported a rate of 0.8 per 1000 live and still births.19 However, the comparability of findings is made difficult due to inconsistencies in the denominator data used. In addition, most of the previous estimates have been based on total births rather than male live births. There are also differences in inclusion criteria across many registries for hypospadias as reported by the European Surveillance of Congenital Anomalies (EUROCAT).20 It is difficult to determine whether or not the milder form of hypospadias was included in the registries and the prevalence estimates.
The trend of corrective procedures for hypospadias appears not to show any temporal change across the entire study period. Assuming that mostly clinically significant forms of hypospadias will have been operated on, and that the indications for surgery have remained unchanged over the study period, there is no evidence of an increase in the birth prevalence of clinically significant cases of hypospadias in the region. It has been previously been reported from earlier studies in different parts of Europe that there were increasing trends for both cryptorchidism and hypospadias. However, it is possible that much of this increase can be explained by improved ascertainment in recent years.1
The aetiology of these conditions remains poorly understood, although a number of previous studies have examined potential aetiological factors. Exposure to environmental endocrine disruptors is the most popular theory; an oestrogenic effect is suggested to be involved as it interferes with the feed-back mechanism of sex hormones in the pituitary gonadal axis during the fetal period. Mothers may be exposed to endocrine disruptors through inhalation, through the skin, or ingestion via food and water. They may also come into contact with these chemicals in the workplace.21
Low birth weight has also been linked with the occurrence of cryptorchidism and hypospadias.22–26 There is also strong evidence of hereditary or familial factors being involved. The risk of siblings also having hypospadias is about 4.2%.27
In conclusion, the HES data were of high quality for the study period. Data on correctional procedures were used to assess prevalence trends. There did not appear to be any change in the prevalence of hypospadias. There was a possible decline in the prevalence of cryptorchidism which may be due to a real decrease in disease prevalence or a decrease in interventions in equivocal cases during the study period. Further research should include a prospective study to confirm the continuing high quality of the HES data.
What is already known on this topic
There is controversy as to whether there is an increasing birth prevalence of cryptorchidism and hypospadias.
Complete datasets of these anomalies are rare because their construction is difficult, with reported trends often hard to interpret due to incomplete ascertainment and varying inclusion criteria.
Comparison of rates between studies is made difficult by inconsistent choice of denominators.
What this study adds
Taking correctional procedures into account there did not appear to be a significant change in the prevalence of hypospadias.
There was an apparent decline in prevalence of cryptorchidism which may be real or may be due to a decrease in interventions.
We would like to thank the North East Public Health Observatory and the Regional Maternity Survey Office for provision of data, and especially Dr John E Scott and Dr Judith Rankin for their guidance in the early stages of this study. We would also like to thank Mr Neil McKnight, Senior Information Manager, North East Public Health Observatory, for assisting us in data abstraction.
Published Online First 6 December 2007
This study was funded by the Birth Defects Foundation.
Competing interests: None.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.