Diagnosis and mortality in 47,XYY persons: a registry study
© Stochholm et al; licensee BioMed Central Ltd. 2010
Received: 4 December 2009
Accepted: 29 May 2010
Published: 29 May 2010
Sex chromosomal abnormalities are relatively common, yet many aspects of these syndromes remain unexplored. For instance epidemiological data in 47,XYY persons are still limited.
Using a national Danish registry, we identified 208 persons with 47,XYY or a compatible karyotype, whereof 36 were deceased; all were diagnosed from 1968 to 2008. For further analyses, we identified age matched controls from the male background population (n = 20,078) in Statistics Denmark. We report nationwide prevalence data, data regarding age at diagnosis, as well as total and cause specific mortality data in these persons.
The average prevalence was 14.2 47,XYY persons per 100,000, which is reduced compared to the expected 98 per 100,000. Their median age at diagnosis was 17.1 years. We found a significantly decreased lifespan from 77.9 years (controls) to 67.5 years (47,XYY persons). Total mortality was significantly increased compared to controls, with a hazard ratio of 3.6 (2.6-5.1). Dividing the causes of deaths according to the International Classification of Diseases, we identified an increased hazard ratio in all informative chapters, with a significantly increased ratio in cancer, pulmonary, neurological and unspecified diseases, and trauma.
We here present national epidemiological data regarding 47,XYY syndrome, including prevalence and mortality data, showing a significantly delay to diagnosis, reduced life expectancy and an increased total and cause specific mortality.
One of the first descriptions of 47,XYY is from 1965 by Jacobs et al . Here a chromosome survey of male patients at the State Hospital in Carstairs was conducted. The hypothesis being that 47,XYY was particularly frequent among inmates in penal institutions. Later, other studies took place in hospitals among consecutively born babies [2–7] using techniques enabling the identification of extra Y chromosome material. These studies identified highly variable number of 47,XYY persons, ranging in liveborn from 26 per 100,000  to 375 per 100,000 . We calculated the prevalence of 47,XYY at birth by pooling data from the surveys in consecutively liveborn babies in various countries and estimated the prevalence of 47,XYY to 98 per 100,000 (95% CI: 73-129) (51 47,XYY persons out of 52,004 liveborn boys). A few earlier chromosomal studies only used identification of Barr bodies, for instance from smears from the oral mucosa, hence no Y chromosome defects were identified [8, 9]. By comparison, it is estimated that sex chromosomal abnormalities occur in 1 per 400 births .
Only limited data regarding age at diagnosis in 47,XYY syndrome in a relatively unselected population are available . To date, not much is generally known regarding clinical phenotype of individuals with 47,XYY, except tall stature , possibly due to the expression of three copies of the short statue homeobox-containing gene (SHOX), which is located on the distal ends of Xp and Yp in the pseudoautosomal region 1 (PAR1) .
One British study  identified a significantly increased mortality in total and in diseases of the respiratory system in a 47,XYY population compared to the background population. However, to our knowledge nothing is known concerning prevalence and mortality in a nationwide identified 47,XYY population. We therefore undertook the present Danish nationwide study, investigating age at diagnosis, prevalence and mortality in a cohort of 47, XYY persons and comparing this to a large background population.
Identification of 47,XYY persons
Details regarding the karyotypes
Identification of controls
Statistics Denmark was founded in 1850, and contains numerous statistics regarding Denmark and Danes. Using the ID-number, Statistics Denmark http://www.dst.dk identified for each index-person up to 100 controls from the male background population, matched on age (year and month of birth). All controls were alive and living in Denmark on the date their index-person was diagnosed with 47,XYY. For two index-persons the matching was not undertaken for unknown reasons, and mortality data are therefore on 206 47,XYY persons. Furthermore, we retrieved data regarding date of emigration, date of death, and up to three causes of death on these men. None were lost to follow-up.
All controls were alive and registered in Denmark on the day the index-person was diagnosed with 47,XYY. Mortality data was updated with 31st of December 2008 being the last date of death registered, whereas causes of death were updated until 31st of December 2006. Thus, all who were deceased in 2007 and 2008 were registered with date of death, but without registration of causes of death.
The causes of death were given in International Classification of Diseases (ICD) 8th edition until 1993, and in ICD-10 and onwards. We translated ICD-8 diagnoses to ICD-10, and divided all deaths into 19 chapters according to ICD-10 for analysis of cause-specific mortality. Cause-specific mortality hazard ratios (HR) were calculated for each of the 19 chapters in ICD-10, as well as in total.
The calculation of the expected prevalence was performed by adding all 47,XYY persons or variants identified in screening studies as well as the number of liveborn boys investigated. The confidence intervals were calculated using the poisson distribution.
To compare median age at diagnosis, median date of birth and median date of diagnosis in the three subgroups, we used the Kruskal-Wallis test. The time trend in age at diagnosis was analyzed using linear regression.
The prevalence was calculated as number of diagnoses per 100 000 liveborn boys in the background population per year of diagnosis. Confidence intervals were estimated using an approximation to the Poisson distribution. The specific numbers of liveborn boys in the background population were obtained using Statistics Denmark. To identify changes in prevalence per year of diagnosis we used Poisson regression. The average prevalence was calculated as average number of incident 47,XYY persons during 1970-2008 divided by the average number of liveborn boys in the background population during the same period. This study period was applied to ensure enough run-in time in the Danish Cytogenetic Registry from time of establishment.
Mortality was described with Kaplan-Meier survival estimates constructed using date of birth as entry. Date of emigration, date of death or 31st of December 2008, whichever came first, were used as date of exit. For comparison log-rank analysis was applied.
HRs were calculated using Cox regression analysis with stratification, using each person and his matched controls as a stratum. Hereby, comparisons were adjusted for age and calendar time, and calculation of expected number of deaths was possible. Time at risk was calculated as time from date of diagnosis until date of emigration, date of death or 31st of December 2008, whichever came first, this applied to both index-persons and controls. All results are shown with 95% confidence intervals, or with range if relevant, and p < 0.05 was considered statistically significant. We made no formal corrections for multiple comparisons. We used Stata 10.0 (Stata Corp. College Station,TX, USA) for all calculations.
Details regarding 47,XYY persons.
Number of persons
Median age at diagnosis (range)
Median year of birth (range)
Median year of diagnosis (range)
Nationwide we have identified all males diagnosed with a diagnosis compatible with 47,XYY and identified a significantly increased total mortality. Generally, cause specific mortality was increased compared to age and gender matched controls. Mortality data have to our knowledge not before been reported in a nationwide cohort with this specific karyotype. Our finding of a total mortality ratio of 3.6 is in comparison or even more pronounced than findings in Turner Syndrome [15, 16] or Klinefelter Syndrome  or 47,XXX  and also higher than found in the only other study in a 47,XYY cohort by Swerdlow et al  with a relative risk of 1.9 (1.20-2.85), although with overlapping confidence interval. The reduction of lifespan of 10.3 years is even more than the expected loss of median lifespan in heavy smokers compared to non-smokers in Denmark . In addition, we demonstrate a considerable delay in diagnosis and a low prevalence of 47,XYY.
In all informative chapters cause specific mortality was generally increased. Analyzing the chapters separately, we identified a significantly increased mortality in cancer, neurological, and pulmonary diseases, trauma and the chapter of unspecified diseases. Further, we identified a statistically significant HR in the chapters concerning the skin, urological diseases, and chromosomal disorders. These analyses are compromised by a small number of deaths among the 47,XYY persons (n = 1), and we consider the significant HR in these three chapters as a chance finding, however in line with the generally increased mortality. The British data showed a significantly increased mortality in pulmonary diseases only . Time at risk is comparable in the British study (3 174 years) and the present one (3 373 years), but the distribution of age at diagnosis cannot readily be extracted from the former study. However, our exact matching of index-persons and controls is methodologically superior. If a similar approach was possible in the British study, probably further significant chapters would have been identified.
We have no obvious explanation for the finding of significantly increased mortality in the various ICD-10 chapters. However, it is noteworthy that the HR of the cardiovascular diseases reached 2.2 (0.99-5.04), p = 0.05, and was as such not significant. We presume that an increased number of index-persons or increased time of observation would have identified a significant HR here as well. It is important to note the total number of deaths in the chapters and not only the significance level. When few deaths among 47,XYY persons are registered, estimates of hazard ratio have a wide confidence interval and hence are relatively imprecise. Due to the limited number of deaths, we have not undertaken analyses regarding eventual sub grouping in any of the chapters, apart from the trauma chapter. Due to the early reports regarding 47,XYY males being overrepresented in prisons [20, 21] we scrutinized data in the trauma chapter. Here, a total of five 47,XYY persons and 176 controls died. The deaths in the 47,XYY persons were due to suicide in an oligophrenic person (n = 1), traffic accident (a pedestrian addicted to drugs) (n = 1), various somatic lesions due to foreign body in the gastro-intestinal system (n = 1), due to overdose with tricyclic antidepressive (n = 1) and a subarachnoid hemorrhage whilst abusing heroin (n = 1). It is important to note that three of these deaths include either an overdose or drug abuse. We recommend that the possibility of a diagnosis of 47,XYY is considered in drug addicted men who are tall [22, 23] or have other stigmata.
The average prevalence of 14.1 per 100 000 is lower than the expected prevalence of 98 in 100 000, however higher than the finding by Abramsky et al in a smaller study from three English laboratories . Here 5.3 percent of the estimated number of 47,XYY persons were identified. In the study by Abramsky at al., six 47,XYY persons were younger and five older than 20 years at diagnosis, which we consider comparable with our finding of a median age at diagnosis of 17.1 years. Only very large scale studies of huge populations will be sufficient to fully establish the exact prevalence of 47,XYY.
The reduction of age at diagnosis during the study period may signal an increased awareness of 47,XYY among physicians. Males identified in a clinical setting are not comparable to those identified in surveys, basically due to the presumed highly variable phenotype among 47,XYY persons spanning from a normal phenotype to a clearly abnormal phenotype. Thus, males identified using our approach of focusing only on all persons diagnosed naturally bias the results. However, for the time being there is no other possible way of identifying the remaining undiagnosed males nationwide, and, more importantly, it is the persons diagnosed that we see in the daily clinic, or who themselves know that they have this chromosomal abnormality. It is important to emphasize that this report only includes males diagnosed with this karyotype. Thus, we consider this cohort as representative of the 47,XYY males being seen by clinicians at the current point of time. To which degree the fact that only a limited percentage of the 47,XYY persons are identified, influences the increased mortality is not known. However, we expect that inclusion of more 47,XYY persons, possibly less stigmatized, will tend to reduce the increased mortality found in the present study.
In conclusion, in this first nationwide study in diagnosed 47,XYY persons we have identified an average prevalence of 14.1 per 100 000, which is lower than the expected of 98 per 100 000. The 47,XYY persons are diagnosed relatively late with a median age at diagnosis of 17.1 years. Their total mortality is significantly increased compared to age and gender matched controls from the background population. Furthermore, this increased mortality is present in all informative chapters according to the ICD-10 and significantly increased in the following: cancer, neurological, and pulmonary diseases, trauma and the chapter of unspecified diseases. Much more needs to be learned about this syndrome and clinical studies should be conducted in order to identify clinical problems enabling future decrease in the increased risk of death.
Kirstine Stochholm was supported by a research grant from Central Denmark Region and from Danish Ministry of Science, Technology and Innovation. Jan Hansen from The Danish Cytogenetic Central Office is thanked for the identification of 47,XYY persons. Statistics Denmark supplied the data regarding controls.
- Jacobs P, Brunton M, Melville M, Brittain R, McClemont W: Aggressive behavior, mental sub-normality and the XYY male. Nature. 1965, 208: 1351-1352. 10.1038/2081351a0.View ArticlePubMedGoogle Scholar
- Goad W, Robinson A, Puck T: Incidence of aneuploidy in a human population. The American Journal of Human Genetics. 1976, 28: 62-68.PubMedGoogle Scholar
- Nielsen J, Wohlert M: Sex chromosome abnormalities found among 34,910 newborn children: results from a 13-year incidence study in Arhus, Denmark. Birth Defects: Original Article Series. 1991, 26: 209-223.Google Scholar
- Sergovich F, Valentine GH, Chen AT, Kinch RA, Smout MS: Chromosome aberrations in 2159 consecutive newborn babies. New England Journal of Medicine. 1969, 280: 851-855.View ArticlePubMedGoogle Scholar
- Maeda T, Ohno M, Matsunobu A, Yoshihara K, Yabe N: A cytogenetic survey of 14,835 consecutive liveborns. Jinrui Idengaku Zasshi. 1991, 36: 117-129.View ArticlePubMedGoogle Scholar
- Ratcliffe SH: Development of children with sex chromosome abnormalities. Proc R Soc Med. 1976, 69: 189-191.PubMed CentralPubMedGoogle Scholar
- Hamerton JL, Canning N, Ray M, Smith S: A cytogenetic survey of 14,069 newborn infants. I. Incidence of chromosome abnormalities. Clin Genet. 1975, 8: 223-243.View ArticlePubMedGoogle Scholar
- Baikie AG, Garson OM, Weste SM, Ferguson J: Numerical abnormalities of the X chromosome. Frequency among inpatients of a general hospital and in a general population. Lancet. 1966, 1: 398-400. 10.1016/S0140-6736(66)91393-6.View ArticlePubMedGoogle Scholar
- MacLean N, Harnden DG, Brown WM, Bond J, Mantle DJ: Sex-chromosome abnormalities in newborn babies. Lancet. 1964, 1: 286-290. 10.1016/S0140-6736(64)92405-5.View ArticlePubMedGoogle Scholar
- Linden MC, Bender BG, Robinson A: Sex Chromosome Tetrasomy and Pentasomy. Pediatrics. 1995, 96: 672-682.PubMedGoogle Scholar
- Abramsky LF, Chapple J: 47,XXY (Klinefelter syndrome) and 47,XYY: estimated rates of and indication for postnatal diagnosis with implications for prenatal counselling. Prenat Diagn. 1997, 17: 363-368. 10.1002/(SICI)1097-0223(199704)17:4<363::AID-PD79>3.0.CO;2-O.View ArticlePubMedGoogle Scholar
- Ratcliffe SG: The effect of chromosome abnormalities on human growth. British Medical Bulletin. 1981, 37: 291-295.PubMedGoogle Scholar
- Vorona E, Zitzmann M, Gromoll J, Schuring AN, Nieschlag E: Clinical, Endocrinological, and Epigenetic Features of the 46,XX Male Syndrome, Compared with 47,XXY Klinefelter Patients. J Clin Endocrinol Metab. 2007, 92: 3458-3465. 10.1210/jc.2007-0447.View ArticlePubMedGoogle Scholar
- Swerdlow AJ, Hermon C, Jacobs PA, Alberman E, Beral V, Daker M, Fordyce A, Youings S: Mortality and cancer incidence in persons with numerical sex chromosome abnormalities: a cohort study. Ann Hum Genet. 2001, 65: 177-188. 10.1046/j.1469-1809.2001.6520177.x.View ArticlePubMedGoogle Scholar
- Schoemaker MJ, Swerdlow AJ, Higgins CD, Wright AF, Jacobs PA: Mortality in women with Turner syndrome in Great Britain: a national cohort study. J Clin Endocrinol Metab. 2008, 93: 4735-4742. 10.1210/jc.2008-1049.View ArticlePubMedGoogle Scholar
- Stochholm K, Juul S, Juel K, Naeraa RW, Gravholt CH: Prevalence, incidence, diagnostic delay, and mortality in Turner syndrome. J Clin Endocrinol Metab. 2006, 91: 3897-3902. 10.1210/jc.2006-0558.View ArticlePubMedGoogle Scholar
- Bojesen A, Juul S, Birkebaek N, Gravholt CH: Increased Mortality in Klinefelter Syndrome. J Clin Endocrinol Metab. 2004, 89: 3830-3834. 10.1210/jc.2004-0777.View ArticlePubMedGoogle Scholar
- Stochholm K, Juul S, Gravholt CH: Mortality and incidence in 47,XXX and variants. American Journal of Medical Genetics. 2010Google Scholar
- Juel K, Sørensen J, Brønnum-Hansen H: Risikofatorer og folkesundhed i Danmark. Statens Institut for Folkesundhed. 2008Google Scholar
- Finley W, McDanal CJ, Finley S, Rosecrans C: Prison survey for the XYY karyotype in tall inmates. Behavior Genetics. 1973, 3: 97-100. 10.1007/BF01067694.View ArticlePubMedGoogle Scholar
- Price W, Whatmore P: Behaviour disorders and pattern of crime among XYY males identified at a maximum security hospital. British Medical Journal. 1967, 1: 533-536. 10.1136/bmj.1.5539.533.PubMed CentralView ArticlePubMedGoogle Scholar
- Witkin HA, Mednick SA, Schulsinger F, Bakkestrom E, Christiansen KO, Goodenough DR, Hirschhorn K, Lundsteen C, Owen DR, Philip J: Criminality in XYY and XXY men. Science. 1976, 193: 547-555. 10.1126/science.959813.View ArticlePubMedGoogle Scholar
- Ottesen AM, Aksglaede L, Garn I, Tartaglia N, Tassone F, Gravholt CH, Bojesen A, Sorensen K, Jorgensen N, Rajpert-De Meyts E: Increased number of sex chromosomes affects height in a nonlinear fashion: a study of 305 patients with sex chromosome aneuploidy. American Journal of Medical Genetics. 2010, 152A (5): 1206-1212. 10.1002/ajmg.a.33334.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.