Abstract
Aims
In this study, we aimed to evaluate incidence trends and potential risk factors associated with Perthes’ disease in Denmark, using publicly available data.
Methods
Our population-based case-control study used data from the Danish National Patient Register and Danish Civil Registration System, accessed through the publicly available Danish Biobank Register, to identify 1,924,292 infants born between 1985 and 2016. We estimated age-specific incidence rates for four birth periods of equal duration (1985 to 1992, 1993 to 2000, 2001 to 2008, and 2009 to 2016) and investigated associations with perinatal conditions, congenital malformations, coagulation defects, autism spectrum disorders (ASD), and attention deficit hyperactivity disorders (ADHD).
Results
We identified 2,374 (81.6% male) diagnosed with Perthes’ disease aged between two and 12 years, corresponding to an overall incidence of 12.1 per 100,000 live births relative to the year of birth. The incidence declined across all four birth periods, irrespective of sex or age at diagnosis. Several perinatal conditions were associated with higher Perthes’ disease risk. Children with reported birth injuries (vs no reported injuries) exhibited the highest risk (relative risk (RR) 7.48 (95% CI 3.37 to 16.63)) followed by those with versus without coagulation defects (RR 4.77 (95% CI 1.79 to 12.69)). Children diagnosed with syndromic (RR 2.90 (95% CI 2.08 to 4.04)) or non-syndromic major congenital malformations (RR 1.86 (95% CI 1.55 to 2.23)) (vs those with no malformation diagnosis) were also associated with higher Perthes’ disease risk. The development of Perthes’ disease was positively associated with several ASD and ADHD diagnoses. However, once adjusting for the sex of the child and period of birth, the associations with ASD and ADHD were no longer significant.
Conclusion
Using publicly available data, we observed a declining incidence of Perthes’ disease in Denmark over a 32-year study period. Our findings also confirm positive associations between Perthes’ disease and various perinatal conditions, coagulation defects, and congenital malformations, highlighting potential aetiological pathways for further investigation.
Cite this article: Bone Joint J 2025;107-B(2):268–276.
Take home message
A statistically significant decline in the incidence of Perthes’ disease was observed in Denmark over the study period, suggesting that possible shifts in environmental factors, lifestyle, and/or healthcare factors are associated with disease development.
Positive associations between Perthes’ disease and various perinatal conditions, coagulation defects, and congenital malformations were identified, suggesting potential pathological mechanisms for further investigation to improve preventive strategies and early intervention for this disease.
Introduction
Perthes’ disease is a rare childhood hip disorder, with estimated annual incidence ranging from 0.2 to 29 per 100,000 children aged under 15 years.1-4 Perthes’ disease is characterized by a disruption to the blood supply to the femoral head, resulting in avascular necrosis and leading to potential deformity and limited range of hip motion.5 Four phases characterize the progression of Perthes’ disease: necrosis, fragmentation, reossification, and remodelling.6 It is noteworthy that the process of reossification and remodelling occurs as part of the pathophysiology of the disease, i.e. in the absence of treatment.7 Despite the absence of randomized controlled trials to guide treatment decisions, clinical management of Perthes’ disease encompasses a spectrum of interventions ranging from observation to surgical procedures.8 Age at presentation and development of joint stiffness are considered crucial factors influencing the decision to perform surgery.9 Supportive evidence suggests that corrective interventions may be necessary to optimize outcomes, with patients with aspherical hips reporting poorer functional capacity, diminished quality of life, and increased pain scores.10
Perthes’ disease is more common in boys and typically occurs between the ages of three and 12 years, with a peak onset between five and seven years of age.5 It is associated with impaired growth and exhibits variation based on race and geography, with two systematic reviews showing higher incidence estimates among people of European descent and those living at higher latitudes.2,3 More recent data suggest a declining incidence of Perthes’ disease in some regions, such as Northern Ireland.11
The aetiology of Perthes’ disease remains elusive, but likely involves a combination of mechanical and biological factors, e.g. mechanical overload and nutritional status.4,5 Several risk factors, including acetabular retroversion, obesity, high latitude, hyperactivity, and coagulopathy, have been reported,5,12 along with perinatal factors including maternal smoking, very low birth weight, caesarean section, and breech birth.13,14 Furthermore, certain thrombophilic conditions, specifically the factor V Leiden mutation and high levels of anticardiolipin antibodies, may also contribute to the risk of developing Perthes’ disease.15 Further research is needed to fully elucidate the pathophysiology for Perthes’ disease.
To this end, we examined the Danish national registry data to explore the incidence of Perthes’ disease diagnoses in children born from 1985 to 2016 and aged two to 12 years at diagnosis. In order to provide a more detailed assessment of potential changes in incidence, we estimated incidence for four distinct birth periods of equal duration (1985 to 1992, 1993 to 2000, 2001 to 2008, and 2009 to 2016) and explored the risk of developing Perthes’ disease in children with selected perinatal conditions, congenital malformations, coagulation defects, autism spectrum disorders (ASD), and attention deficit hyperactivity disorders (ADHD).
Methods
Data source and cohort definition
Our population-based case-control study used the Danish Biobank Register, which integrates data from the Danish Civil Registration System (demographic information) and the Danish National Patient Register (diagnostic codes and dates).16 We accessed the publicly available online interface to identify all children born between 1 January 1985 and 31 December 2016, using data from individuals sampled for neonatal screening. The Danish National Patient Register used ICD-8 codes from 1977 to 1993 and transitioned to ICD-10 beginning in 1994; ICD-9 codes were never used in Denmark.17 Consequently, we selected patients diagnosed at two to 12 years of age, using the following diagnostic codes (ICD-8: 722.10, 722.11, 722.19 between 1985 and 1993; ICD-10-DK: DM911, DM912, DM913 between 1994 and 2016). Our cohort represents a nationwide sample of infants alive at the time of screening (days 5 to 7 in 1994 to 2008, days 2 to 3 in 2009 to the present).18
Ethical approval and consent to participate
According to Danish law and regulations, no formal approval or review of ethics were required for our study as data were retrieved from publicly available sources, and individual patient data were not included.
Exposure data
We examined potential risk factors including perinatal conditions (ICD-8: 760–779; ICD-10-DK: DP00–DP96), congenital malformations (Supplementary Table i), coagulation defects (ICD-10-DK: DD68), and ASD and ADHD (Supplementary Table ii). Perinatal conditions were examined for the years 1985 to 2016, and consequently both ICD-8 and ICD-10 codes were used.17 However, to ensure diagnostic accuracy in identifying patients with congenital malformations, coagulation disorders, ASD, and ADHD, we examined these conditions in a subset of the population born between 1994 and 2011. This subset captures patients, born after the implementation of ICD-10-DK (1994), allowing for greater precision, and who have passed their twelfth birthday (2011). This reflects the youngest possible subset of the population and as such is more reflective of the population under current care. Frequencies and proportions were calculated for each study population’s descriptive characteristics. Additionally, we assessed the co-occurrence of Perthes’ disease with developmental dysplasia of the hip (DDH) (ICD-10-DK: DQ658D, DQ658E, DQ658F, DQ659) to evaluate potential confounding.
Statistical analysis
Incidence was estimated as the number of diagnoses per 100,000 children, per birth year. Relative risks (RRs) and 95% CIs were estimated using an exact method based on binomial distribution to explore the relationship between Perthes’ disease and perinatal conditions, congenital malformations (analyzed hierarchically), and coagulation defects. Poisson regression was used to assess the age-specific trends for the four birth periods, chosen to create equal time periods within the study period (1985 to 1992, 1993 to 2000, 2001 to 2008, 2009 to 2016), and the relationship between ASD and ADHD (analyzed hierarchically) with Perthes’ disease. Adjusted relative risks (aRRs) and 95% CIs were estimated, temporal trends were adjusted for sex and age at diagnosis, and ASD and ADHD models were adjusted for sex and period of birth. All analyses were performed using R version 4.3.2 (R Foundation for Statistical Computing, Austria).
Results
Incidence by birth period
We identified 2,374 individuals diagnosed with Perthes’ disease among 1,924,292 children born in Denmark between 1985 and 2016, corresponding to an overall incidence of 12.1 per 100,000 children (4.6 per 100,000 girls and 19.2 per 100,000 boys). Notably, the incidence of Perthes’ disease declined across the study period, although the initial years (1985 to 1991) showed an increase, reaching a peak in 1989 for boys and 1991 for girls. Following the implementation of ICD-10-DK in 1994, this decline continued steadily throughout the remaining study period (Figure 1). This decline was observed across all four time periods evaluated, with rates per 100,000 of 16.19 (95% CI 15.11 to 17.34) in 1985 to 1992, 13.72 (95% CI 12.77 to 14.71) in 1993 to 2000, 9.50 (95% CI 8.69 to 10.36) in 2001 to 2008, and 7.37 (95% CI 6.48 to 8.35) in 2009 to 2016. This decline was apparent, irrespective of sex or age at diagnosis (Figure 2, Table I), with a 69% reduction in Perthes’ disease cases observed in the 2009 to 2016 period compared to the 1985 to 1992 period. Peak age at diagnosis consistently occurred from three to seven years in each time period (Figure 2, Table I).
Fig. 1
Trends in annual incidence of Perthes’ disease, diagnosed at ages two to 12 years, per 100,000 live births relative to the year of birth. The shaded ribbons represent the 95% CIs. The dotted trendlines represent the linear trend across the period during which ICD-8 was used in Denmark, and the dashed trendlines represents the linear trend following the implementation of ICD-10-DK in 1994 (indicated by a dashed vertical line). ICD, International Classification of Diseases of the World Health Organization.
Fig. 2
Incidence per 100,000 live births relative to age at diagnosis for four birth periods (1985 to 1992, 1993 to 2000, 2001 to 2008, 2009 to 2016). The reduction of the incidence among boys and girls to zero in the older age groups for the birth period 2009 to 2016 represents the young age of these children and consequently very small sample sizes in these age groups. The shaded ribbons represent the 95% CIs.
Table I.
Crude and adjusted relative risk for Perthes’ disease by infant sex, birth period, and age at diagnosis.
| Coefficient | RR (95% CI) | aRR (95% CI) |
|---|---|---|
| Sex | ||
| Female | Reference | |
| Male | 4.21 (2.83 to 6.49) | 4.20 (3.66 to 4.85) |
| Birth period | ||
| 1985 to 1992 | Reference | |
| 1993 to 2000 | 0.85 (0.51 to 1.41) | 0.85 (0.74 to 0.97) |
| 2001 to 2008 | 0.59 (0.33 to 1.03) | 0.59 (0.51 to 0.68) |
| 2009 to 2016 | 0.31 (0.14 to 0.62) | 0.31 (0.26 to 0.38) |
| Age at diagnosis, yrs | ||
| 2 | Reference | |
| 3 | 2.07 (0.99 to 4.66) | 2.07 (1.64 to 2.64) |
| 4 | 2.17 (1.04 to 4.85) | 2.17 (1.71 to 2.75) |
| 5 | 2.10 (1.00 to 4.72) | 2.10 (1.66 to 2.67) |
| 6 | 1.86 (0.87 to 4.24) | 1.86 (1.47 to 2.38) |
| 7 | 1.34 (0.58 to 3.16) | 1.34 (1.03 to 1.74) |
| 8 | 0.98 (0.40 to 2.44) | 0.99 (0.75 to 1.30) |
| 9 | 0.57 (0.19 to 1.58) | 0.57 (0.41 to 0.79) |
| 10 | 0.42 (0.12 to 1.27) | 0.42 (0.29 to 0.60) |
| 11 | 0.26 (0.05 to 0.91) | 0.26 (0.16 to 0.39) |
| 12 | 0.28 (0.06 to 0.94) | 0.28 (0.18 to 0.41) |
-
aRR, adjusted relative risk; RR, relative risk.
Associations with perinatal conditions
Our analysis of perinatal conditions revealed that the absolute risks (AR) of Perthes’ disease were generally low, ranging from one to two per 1,000 live births across most investigated conditions (Table II). However, several groups of perinatal conditions (ICD-10-DK) were positively associated with an increased risk of Perthes’ disease development, although the CIs while excluding the null were wide, indicating imprecise effect sizes (Table II). Other birth injuries (DP15), a broad diagnostic category encompassing a number of organ and tissue injuries as well as unspecified injuries which occur in the foetal or perinatal period,19 exhibited the highest RR of 7.48 (95% CI 3.37 to 16.63), followed by intracranial non-traumatic haemorrhage of foetus and newborn (DP52) (RR 2.96 (95% CI 1.33 to 6.60)), other conditions of integument specific to foetus and newborn (DP83) (RR 2.49 (95% CI 1.29 to 4.79)), and foetus and newborn affected by complications of placenta, cord, and membranes (DP02) (RR 2.26 (95% 1.13 to 4.52)).
However, it is important to consider these risk estimates in the context of the low baseline risk of Perthes’ disease (1.28 per 1,000) in this group. Transitory disorders of carbohydrate metabolism specific to the foetus and the newborn (DP70) were also positively associated with the development of Perthes’ disease, with CIs that excluded the null (RR 1.41 95% 1.12 to 1.78). Exclusively among children born prior to 1994 (i.e. based on ICD-8 codes), we observed that other maternal conditions unrelated to pregnancy (761) exhibited the highest risk (RR 2.00 (95% CI 1.04 to 3.85)), followed by difficult labour with other and unspecified complications (ICD-8: 768) (RR 1.48 (95% CI 1.08 to 2.04)) and immaturity, unqualified (777) (RR 1.36 (95% CI 1.04 to 1.78)). While these RR estimates suggest potential associations, it is important to acknowledge that they do not directly translate to a high absolute risk of Perthes’ disease development due to the underlying low prevalence of the disease. Associations for the remainder of perinatal conditions evaluated using either ICD-10-DK or ICD-8 codes ranged from inverse to positive, but none had CIs that excluded the null (Table II).
Table II.
The number and absolute risk of Perthes’ disease (diagnosed at two to 12 years of age) per 1,000 infants diagnosed with conditions originating in the perinatal period, born between 1985 and 2016. International Statistical Classification of Diseases and Related Health Problems (ICD) codes where fewer than five Perthes’ disease cases were identified have been removed from the table in accordance with Danish data protection laws and regulations. The relative risk represents the risk of developing Perthes’ disease following exposure to a condition originating in the perinatal period. The ICD codes presented in this table represent higher-level categories within the ICD coding hierarchy, capturing all subcategories nested below them.
| ICD | ICD category title | n | AR/1,000 (95% CI) | RR (95% CI) |
|---|---|---|---|---|
| All | All conditions originating in the perinatal period | 564 | 1.28 (1.17 to 1.39) | 1.05 (0.95 to 1.15) |
| ICD-10-DK (1994 to 2016) | ||||
| DP01 | Foetus and newborn affected by maternal complications of pregnancy | 26 | 1.38 (0.90 to 2.02) | 1.35 (0.92 to 2.00) |
| DP02 | Foetus and newborn affected by complications of placenta, cord, and membranes | 8 | 2.31 (1.00 to 4.54) | 2.26 (1.13 to 4.52) |
| DP03 | Foetus and newborn affected by other complications of labour and delivery | 53 | 1.25 (0.93 to 1.63) | 1.22 (0.93 to 1.61) |
| DP05 | Slow foetal growth and foetal malnutrition | 49 | 1.34 (0.99 to 1.78) | 1.32 (0.99 to 1.76) |
| DP07 | Disorders related to short gestation and low birth weight, not elsewhere classified | 91 | 1.14 (0.92 to 1.40) | 1.12 (0.91 to 1.39) |
| DP08 | Disorders related to long gestation and high birth weight | 65 | 0.92 (0.71 to 1.17) | 0.89 (0.70 to 1.15) |
| DP12 | Birth injury to scalp | 5 | 1.07 (0.35 to 2.49) | 1.04 (0.43 to 2.51) |
| DP15 | Other birth injuries | 6 | 7.64 (2.81 to 16.56) | 7.48 (3.37 to 16.63) |
| DP20 | Intrauterine hypoxia | 8 | 1.64 (0.71 to 3.23) | 1.60 (0.80 to 3.21) |
| DP21 | Birth asphyxia | 67 | 1.23 (0.95 to 1.56) | 1.21 (0.95 to 1.55) |
| DP22 | Respiratory distress of the newborn | 64 | 0.99 (0.77 to 1.27) | 0.97 (0.75 to 1.25) |
| DP24 | Neonatal aspiration syndromes | 6 | 0.99 (0.36 to 2.15) | 0.96 (0.43 to 2.15) |
| DP28 | Other respiratory conditions originating in the perinatal period | 7 | 1.31 (0.53 to 2.69) | 1.28 (0.61 to 2.68) |
| DP36 | Bacterial sepsis of the newborn | 34 | 1.14 (0.79 to 1.60) | 1.12 (0.79 to 1.57) |
| DP39 | Other infections specific to the perinatal period | 11 | 1.43 (0.71 to 2.56) | 1.40 (0.77 to 2.53) |
| DP52 | Intracranial non-traumatic haemorrhage of the foetus and the newborn | 6 | 3.03 (1.11 to 6.58) | 2.96 (1.33 to 6.60) |
| DP59 | Neonatal jaundice from other and unspecified causes | 80 | 1.01 (0.80 to 1.26) | 0.99 (0.79 to 1.24) |
| DP70 | Transitory disorders of carbohydrate metabolism specific to the foetus and the newborn | 77 | 1.43 (1.13 to 1.78) | 1.41 (1.12 to 1.78) |
| DP83 | Other conditions of integument specific to the foetus and the newborn | 9 | 2.54 (1.16 to 4.82) | 2.49 (1.29 to 4.79) |
| DP92 | Feeding problems of the newborn | 42 | 0.86 (0.62 to 1.16) | 0.83 (0.61 to 1.13) |
| ICD-8-DK (1985 to 1993) | ||||
| 761 | Other maternal conditions unrelated to pregnancy | 9 | 3.54 (1.62 to 6.72) | 2.00 (1.04 to 3.85) |
| 768 | Difficult labour with other and unspecified complications | 39 | 2.60 (1.85 to 3.56) | 1.48 (1.08 to 2.04) |
| 769 | Other complications of pregnancy and childbirth | 10 | 2.04 (0.98 to 3.74) | 1.15 (0.61 to 2.13) |
| 772 | Birth injury without mention of cause | 8 | 2.03 (0.88 to 4.00) | 1.14 (0.57 to 2.29) |
| 776 | Anoxic and hypoxic conditions not elsewhere classified | 61 | 2.11 (1.61 to 2.70) | 1.20 (0.92 to 1.55) |
| 777 | Immaturity, unqualified | 56 | 2.38 (1.80 to 3.09) | 1.36 (1.04 to 1.78) |
| 778 | Other conditions of the foetus or the newborn | 81 | 2.02 (1.61 to 2.51) | 1.15 (0.92 to 1.44) |
-
AR, absolute risk; ICD, International Statistical Classification of Diseases and Related Health Problems; RR, relative risk.
Associations between congenital malformations, coagulation defects, and Perthes’ disease
In the subset of children born between 1994 and 2011 (n = 1,117,246, Perthes’ disease cases = 1,327), those with syndromic or non-syndromic major congenital malformations were associated with increased Perthes’ disease risk compared to those with no reported congenital malformation or syndrome (isolated Perthes’ disease) (Table III), although some estimates were imprecise. Among individuals with diagnosed syndromes, those with Down syndrome (RR 11.90 (95% CI 4.98 to 28.46)) or osteochondrodysplasia with defects of growth of tubular bones and spine (RR 16.37 (95% CI 6.86 to 39.07)) showed the highest associations. Among those with non-syndromic presentations of major malformations, the highest risks were observed for reduction defects of the lower limb (RR 7.05 (95% CI 3.79 to 13.10)) and other congenital malformations of the limb(s) (RR 3.26 (95% CI 1.75 to 6.07)). Other coagulation defects (DD68) showed a five-fold increased risk for Perthes’ disease (RR 4.77 (95% CI 1.79 to 12.69)); however, the number of children affected was less than five, and so this increased risk should be interpreted with caution.
Table III.
The number and absolute risk of Perthes’ disease (diagnosed at two to 12 years of age) per 1,000 infants diagnosed with congenital malformations, deformations, and chromosomal abnormalities, born between 1994 and 2011. International Statistical Classification of Diseases and Related Health Problems (ICD) codes where fewer than five Perthes’ disease cases were identified have been removed from the table in accordance with Danish data protection laws and regulations. Diagnoses were assessed hierarchically in the order shown in Supplementary Table i. The relative risk represents the risk of developing Perthes’ disease following exposure to a congenital malformation. The ICD codes presented in this table represent higher-level categories within the ICD coding hierarchy, capturing all subcategories nested below them.
| ICD-10-DK | ICD-10 category title | n | AR/1,000 (95% CI) | RR (95% CI) |
|---|---|---|---|---|
| All congenital malformations | 161 | 2.30 (1.96 to 2.69) | 2.07 (1.76 to 2.44) | |
| Syndromic congenital malformations | 36 | 3.38 (2.37 to 4.68) | 2.90 (2.08 to 4.04) | |
| Chromosomal abnormalities | 14 | 4.47 (2.45 to 7.49) | 3.80 (2.24 to 6.42) | |
| DQ90 | Down syndrome (trisomy 21) | 5 | 14.08 (4.59 to 32.56) | 11.90 (4.98 to 28.46) |
| DQ99 | Other chromosome abnormalities, not elsewhere classified | 6 | 6.17 (2.27 to 13.39) | 5.22 (2.34 to 11.60) |
| Genetic syndromes | 21 | 3.08 (1.91 to 4.71) | 2.62 (1.71 to 4.03) | |
| DQ77 | Osteochondrodysplasia with defects of growth of tubular bones and spine | 5 | 19.38 (6.32 to 44.64) | 16.37 (6.86 to 39.07) |
| DQ780–DQ789 | Other osteochondrodysplasias | 5 | 3.74 (1.22 to 8.71) | 3.16 (1.31 to 7.58) |
| DQ87 | Other specified congenital malformation syndromes affecting multiple systems | 5 | 2.08 (0.68 to 4.85) | 1.76 (0.73 to 4.22) |
| Teratogenic syndromes | < 5 | NE | 1.20 (0.17 to 8.49) | |
| Non-syndromic major congenital malformations | 125 | 2.11 (1.76 to 2.51) | 1.86 (1.55 to 2.23) | |
| Major congenital malformations of the heart | 30 | 1.86 (1.26 to 2.66) | 1.58 (1.10 to 2.27) | |
| DQ21 | Congenital malformations of the cardiac septa | 12 | 1.54 (0.79 to 2.68) | 1.30 (0.74 to 2.29) |
| DQ24 | Other congenital malformations of the heart | 7 | 2.43 (0.98 to 4.99) | 2.05 (0.98 to 4.30) |
| DQ25 | Congenital malformations of the great arteries | 5 | 1.97 (0.64 to 4.59) | 1.66 (0.69 to 3.99) |
| Major congenital malformations of the nervous system | 6 | 2.22 (0.81 to 4.82) | 1.87 (0.84 to 4.17) | |
| Major congenital malformations of the renal system | 10 | 1.90 (0.91 to 3.50) | 1.61 (0.86 to 2.99) | |
| Orofacial clefts | 5 | 2.47 (0.80 to 5.76) | 2.08 (0.87 to 5.01) | |
| Major congenital malformations of the limb(s) | 51 | 2.56 (1.91 to 3.37) | 2.20 (1.67 to 2.91) | |
| DQ65 | Congenital deformities of the hip | 22 | 2.16 (1.35 to 3.26) | 1.83 (1.20 to 2.79) |
| DQ72 | Reduction defects of the lower limb | 10 | 8.32 (4.00 to 15.25) | 7.05 (3.79 to 13.10) |
| DQ74 | Other congenital malformations of the limb(s) | 10 | 3.85 (1.85 to 7.08) | 3.26 (1.75 to 6.07) |
| Major congenital malformations of the genitalia | 9 | 1.88 (0.86 to 3.56) | 1.59 (0.82 to 3.05) | |
| DQ54 | Hypospadias | 5 | 1.49 (0.48 to 3.48) | 1.26 (0.52 to 3.02) |
| Major congenital malformations of digestive system | < 5 | NE | 1.05 (0.34 to 3.26) | |
| Major congenital malformations of the eye | < 5 | NE | 1.12 (0.36 to 3.47) | |
| Major congenital malformations of the ear, face, and neck | 5 | 3.12 (1.01 to 7.25) | 2.63 (1.09 to 6.32) | |
| Major congenital malformations of abdominal wall | < 5 | NE | 4.32 (0.61 to 30.54) | |
| Major congenital malformations of respiratory system | < 5 | NE | 1.79 (0.25 to 12.68) | |
| Other major congenital malformations | < 5 | NE | 0.57 (0.08 to 4.01) | |
| Isolated: Perthes’ disease (no congenital malformation recorded) | 1,166 | 1.11 (1.05 to 1.18) | 0.48 (0.41 to 0.57) | |
-
AR, absolute risk; ICD, International Statistical Classification of Diseases and Related Health Problems; NE, not estimated; RR, relative risk.
Associations with neurodevelopmental disorders
Among the ASD and ADHD diagnoses evaluated, pervasive developmental disorder, unspecified (DF849) (RR 1.82 (95% CI 1.02 to 2.95)), disturbance of activity and attention – attention deficit (DF900) (RR 1.58 (95% CI 1.16 to 2.09)), and hyperkinetic conduct disorder (DF901) (RR 2.62 (95% CI 1.04 to 5.32)) (Table IV) displayed increased risk of developing Perthes’ disease with CIs that excluded the null. After adjusting for child sex and birth period, these diagnoses were no longer significantly associated with Perthes’ disease.
Table IV.
Crude and adjusted relative risk for Perthes’ disease with autism spectrum or attention deficit hyperactivity disorder diagnoses by period of birth and infant sex. Diagnoses were assessed hierarchically in the order shown in Supplementary Table ii.
| Coefficient | RR (95% CI) | aRR (95% CI) |
|---|---|---|
| Sex | ||
| Female | Reference | Reference |
| Male | 4.17 (3.48 to 5.04) | 4.14 (3.40 to 5.09) |
| Birth period | ||
| 1994 to 1999 | Reference | Reference |
| 2000 to 2005 | 0.74 (0.61 to 0.91) | 0.74 (0.62 to 0.88) |
| 2006 to 2011 | 0.58 (0.46 to 0.72) | 0.58 (0.47 to 0.70) |
| Diagnosis | ||
| POPULATION | Reference | Reference |
| ICD10_DF841 | 1.41 (0.82 to 2.22) | 1.1 (0.48 to 2.12) |
| ICD10_DF845 | 0.82 (0.29 to 1.76) | 0.68 (0.13 to 2.00) |
| ICD10_DF848 | 0.89 (0.43 to 1.62) | 0.69 (0.22 to 1.61) |
| ICD10_DF849 | 1.82 (1.02 to 2.95) | 1.64 (0.67 to 3.29) |
| ICD10_DF900 | 1.58 (1.16 to 2.09) | 1.28 (0.81 to 1.93) |
| ICD10_DF901 | 2.62 (1.04 to 5.32) | 1.93 (0.44 to 5.25) |
| ICD10_DF908 | 3.29 (0.55 to 10.16) | 3.23 (0.15 to 15.11) |
| ICD10_DF909 | 0.9 (0.32 to 1.93) | 0.84 (0.16 to 2.47) |
-
aRR, adjusted relative risk; ICD-10, 10th revision of the International Statistical Classification of Diseases and Related Health Problems; RR, relative risk.
Co-occurrence of Perthes’ disease and DDH
A total of 4.5% (60/1,327) of Perthes’ disease cases also had a diagnosis of DDH. The proportion of Perthes’ disease with coexisting DDH varied by year of birth, ranging from 0.0% to 10.7%. However, there was no temporal trend and no sex-specific variation in co-occurrence. Restricting the analysis to Perthes’ disease cases diagnosed between two and five years of age (n = 771) resulted in a slightly lower co-occurrence rate of 3.6% (28/771) with DDH. However, the sample size in this subgroup was relatively small.
Discussion
We used nationwide Danish registry data to investigate the incidence and potential risk factors associated with Perthes’ disease in children aged two to 12 years at diagnosis. Our findings revealed an initial increase in Perthes’ disease incidence between 1985 and 1991, matching an increase noted in a Swedish study over the same time period.20 However, we noted a declining incidence of Perthes’ disease regardless of sex or age at diagnosis, with a 69% decrease between the first observation period (1985 to 1992: 16.19 per 100,000) and the last observation period (2009 to 2016: 7.37 per 100,000). This observed decline aligns with similar trends reported in other regions, including a 61% decrease over 15 years in Northern Ireland11 and a 4.2% year-to-year decline over 19 years in the UK,21 with this latter decline more pronounced in the northern areas of the UK, with an annual decline of 5.1%.21 These findings suggest that potential changes in environment, lifestyle, or healthcare factors may influence disease presentation, e.g. improvements in prenatal care, delivery practices, and neonatal management could potentially reduce birth-related trauma. Additionally, public health initiatives promoting healthy lifestyles and injury prevention might also influence Perthes’ disease risk. However, the magnitude of the decrease (69%) raises the question of whether environment, lifestyle, and healthcare factors alone can fully account for it. It is possible that some of the observed decline might also be attributable to changes in how Perthes’ disease is diagnosed or coded over time.
Our examination of perinatal conditions identified several diagnostic categories with statistically significant associations with Perthes’ disease risk. Notably, other birth injuries (DP15) emerged as the strongest risk factor, highlighting the potential role of birth-related trauma in the aetiology of Perthes’ disease. However, this ICD-10-DK diagnostic code comprises a group of heterogeneous but quite severe and rare birth-related traumas such as lesion of internal organs, face, or muscles. One may speculate that the children in this diagnostic group may represent infants with particularly traumatic birth, and as such our finding highlights the potential role of birth-related trauma in the aetiology of Perthes’ disease. Intracranial haemorrhage of the foetus and newborn may itself be a consequence of multiple aetiologies, including vascular malformations,22 which may indicate a shared aetiology.
Both syndromic and non-syndromic major congenital malformations were associated with an increased risk of developing Perthes’ disease. Among syndromic cases, Down syndrome (DQ90) and osteochondrodysplasia (DQ77) showed the highest risk, suggesting potential genetic or developmental disruptions contributing to both conditions. No published evidence exists linking Down syndrome to Perthes’ disease; however, a study of Turkish children with Down syndrome and Perthes’ disease reported that the occurrence of Perthes’ disease in these children is similar to that in children without Down syndrome.23 Although individuals with Down syndrome frequently present with musculoskeletal findings, such as hypotonia and ligament laxity,24 these findings suggest that the diagnosis of Perthes’ disease is not made by chance in these children. Interestingly, findings of specific non-syndromic limb malformations (DQ65, DQ72, and DQ74) and increased Perthes’ disease risk suggest possible shared developmental processes affecting both skeletal and hip development. Indeed, the presence of anisomelia in Perthes’ disease is associated with the extent of involvement and growth arrest.25
Perthes’ disease has been reported to be associated with the factor V Leiden mutation, high fibrinogen levels, and increased factor VIII levels.26 Supporting these associations, our analysis of coagulation defects revealed an association between other coagulation defects (DD68). We were unable to assess factor V Leiden mutation status or fibrinogen levels using our data.
The relationship between Perthes’ disease and ADHD was postulated based on overlapping behaviour characteristics;27 however, the literature is mixed. A small study of 16 children with Perthes’ disease and their non-Perthes’ disease siblings observed no significant difference in ADHD markers between the two groups.28 However, an earlier study which examined 4,057 Perthes’ disease cases compared to 40,570 controls (matched by sex and date of birth) determined a higher risk of ADHD in individuals with Perthes’ disease.29 In contrast, our study found that prior to adjusting for sex and period of birth, several ASD and ADHD diagnoses were associated with an increased risk of Perthes’ disease, but this relationship was confounded by a shared association with sex and period of birth. While there is no evidence to link Perthes’ disease with ASD, we assessed ASD and ADHD together in order to account for the substantial overlap between these two childhood neurodevelopmental disorders.30
Our findings suggest that a complex interaction among diverse factors may play a role in the development of Perthes’ disease, and future studies should be performed to elucidate the mechanisms underlying these associations. In particular, life course surveillance studies of individual patients may clarify the role of the associations described here.
The main strength of our study is that it is a nationwide evaluation of all Perthes’ disease cases identified from publicly available administrative register data. However, the use of ICD diagnostic codes at an aggregate level limits our analysis to the population level. To our knowledge, these data have not been validated for Perthes’ disease diagnoses. Consequently, there is the potential for misclassification of some Perthes’ disease diagnoses recorded, e.g. we cannot completely exclude DDH as confounding factor, but the overall co-occurrence of Perthes’ disease and DDH was low in our study. Existing evidence suggests high positive predictive values (83%) for orthopaedic surgery-related diagnoses within the Danish National Patient Register,31 providing some assurance of the quality of diagnoses identified. Additionally, the overall quality and completeness of these data are thought to have improved over time.31 Therefore, although data validation remains a limitation, it is unlikely to have a considerable impact on our study’s overall findings. It is possible, however, that the observed decline in Perthes’ disease incidence might have been partially influenced by temporal trends in diagnostic practices or data recording. Although we selected the codes thought to best capture Perthes’ disease diagnoses throughout the study period, subtle changes in coding practices over time cannot be entirely excluded. These limitations can be addressed in future studies by examining personal electronic health records for individual level assessment of these cases.
In conclusion, we report a declining incidence of Perthes’ disease in Denmark and identified various risk factors associated with the disease, specifically perinatal complications, congenital malformations, and coagulation defects. Further research into the pathological mechanisms at play in the development of Perthes’ disease is crucial for improving preventive strategies and early intervention.
References
1. Bansal T , Jaiswal R . Legg-Calve-Perthes disease: a must know entity for anaesthesiologists . Indian J Anaesth . 2016 ; 60 ( 1 ): 68 – 69 . Crossref PubMed Google Scholar
2. Loder RT , Skopelja EN . The epidemiology and demographics of legg-calvé-perthes’ disease . ISRN Orthop . 2011 ; 2011 : 504393 . Crossref PubMed Google Scholar
3. Perry DC , Machin DMG , Pope D , et al. Racial and geographic factors in the incidence of Legg-Calvé-Perthes’ disease: a systematic review . Am J Epidemiol . 2012 ; 175 ( 3 ): 159 – 166 . Crossref PubMed Google Scholar
4. Pavone V , Chisari E , Vescio A , Lizzio C , Sessa G , Testa G . Aetiology of Legg-Calvé-Perthes disease: a systematic review . World J Orthop . 2019 ; 10 ( 3 ): 145 – 165 . Crossref PubMed Google Scholar
5. Rodríguez-Olivas AO , Hernández-Zamora E , Reyes-Maldonado E . Legg-Calvé-Perthes disease overview . Orphanet J Rare Dis . 2022 ; 17 ( 1 ): 125 . Crossref PubMed Google Scholar
6. Mills S , Burroughs KE . Legg-Calve-Perthes Disease . Treasure Island, Florida : StatPearls Publishing , 2024 . Google Scholar
7. Dustmann HO . Etiology and pathogenesis of epiphyseal necrosis in childhood as exemplified with the hip . Z Orthop Ihre Grenzgeb . 1996 ; 134 ( 5 ): 407 – 412 . Crossref PubMed Google Scholar
8. Galloway AM , Pini S , Holton C , et al. “Waiting for the best day of your life”. A qualitative interview study of patients’ and clinicians’ experiences of Perthes’ disease . Bone Jt Open . 2023 ; 4 ( 10 ): 735 – 741 . Crossref PubMed Google Scholar
9. Perry DC , Arch B , Appelbe D , et al. The British Orthopaedic Surgery Surveillance study: Perthes’ disease: the epidemiology and two-year outcomes from a prospective cohort in Great Britain . Bone Joint J . 2022 ; 104-B ( 4 ): 510 – 518 . Crossref PubMed Google Scholar
10. Ali MS , Khattak M , Metcalfe D , Perry DC . Radiological hip shape and patient-reported outcome measures in healed Perthes’ disease . Bone Joint J . 2023 ; 105-B ( 6 ): 711 – 716 . Crossref PubMed Google Scholar
11. Mullan CJ , Thompson LJ , Cosgrove AP . The declining incidence of Legg-Calve-Perthes’ disease in Northern Ireland: an epidemiological study . J Pediatr Orthop . 2017 ; 37 ( 3 ): e178 – e182 . Crossref PubMed Google Scholar
12. Mazloumi SM , Ebrahimzadeh MH , Kachooei AR . Evolution in diagnosis and treatment of Legg-Calve-Perthes disease . Arch Bone Jt Surg . 2014 ; 2 ( 2 ): 86 – 92 . PubMed Google Scholar
13. Bahmanyar S , Montgomery SM , Weiss RJ , Ekbom A . Maternal smoking during pregnancy, other prenatal and perinatal factors, and the risk of Legg-Calvé-Perthes disease . Pediatrics . 2008 ; 122 ( 2 ): e459 – 64 . Crossref PubMed Google Scholar
14. Lindblad M , Josefsson A , Bladh M , Sydsjö G , Johansson T . Risk factors during pregnancy and delivery for the development of Perthes’ disease, a nationwide Swedish study of 2.1 million individuals . BMC Pregnancy Childbirth . 2020 ; 20 ( 1 ): 192 . Crossref PubMed Google Scholar
15. Balasa VV , Gruppo RA , Glueck CJ , et al. Legg-Calve-Perthes disease and thrombophilia . J Bone Joint Surg Am . 2004 ; 86-A ( 12 ): 2642 – 2647 . Crossref PubMed Google Scholar
16. No authors listed . Danish Biobank Register . Danmarks Nationale Biobank . 2022 . https://biobanks.dk/ ( date last accessed 19 December 2024 ). Google Scholar
17. Nguyen-Nielsen M , Svensson E , Vogel I , Ehrenstein V , Sunde L . Existing data sources for clinical epidemiology: Danish registries for studies of medical genetic diseases . Clin Epidemiol . 2013 ; 5 : 249 – 262 . Crossref PubMed Google Scholar
18. Nørgaard-Pedersen B , Hougaard DM . Storage policies and use of the Danish Newborn Screening Biobank . J Inherit Metab Dis . 2007 ; 30 ( 4 ): 530 – 536 . Crossref PubMed Google Scholar
19. No authors listed . ICD-10-CM Section P10-P15 . ICD.Codes . 2024 . https://icd.codes/icd10cm/chapter16/P10-P15 ( date last accessed 19 December 2024 ). Google Scholar
20. Johansson T , Lindblad M , Bladh M , Josefsson A , Sydsjö G . Incidence of Perthes’ disease in children born between 1973 and 1993 . Acta Orthop . 2017 ; 88 ( 1 ): 96 – 100 . Crossref PubMed Google Scholar
21. Perry DC , Bruce CE , Pope D , Dangerfield P , Platt MJ , Hall AJ . Legg-Calvé-Perthes disease in the UK: geographic and temporal trends in incidence reflecting differences in degree of deprivation in childhood . Arthritis Rheum . 2012 ; 64 ( 5 ): 1673 – 1679 . Crossref PubMed Google Scholar
22. Tan AP , Svrckova P , Cowan F , Chong WK , Mankad K . Intracranial hemorrhage in neonates: a review of etiologies, patterns and predicted clinical outcomes . Eur J Paediatr Neurol . 2018 ; 22 ( 4 ): 690 – 717 . Crossref PubMed Google Scholar
23. Talmac MA , Kadhim M , Rogers KJ , Holmes L , Miller F . Legg-Calvé-Perthes disease in children with Down syndrome . Acta Orthop Traumatol Turc . 2013 ; 47 ( 5 ): 334 – 338 . Crossref PubMed Google Scholar
24. Winders P , Wolter-Warmerdam K , Hickey F . A schedule of gross motor development for children with Down syndrome . J Intellect Disabil Res . 2019 ; 63 ( 4 ): 346 – 356 . Crossref PubMed Google Scholar
25. Grzegorzewski A , Synder M , Kozłowski P , Szymczak W , Bowen RJ . Leg length discrepancy in Legg-Calve-Perthes disease . J Pediatr Orthop . 2005 ; 25 ( 2 ): 206 – 209 . Crossref PubMed Google Scholar
26. Vosmaer A , Pereira RR , Koenderman JS , Rosendaal FR , Cannegieter SC . Coagulation abnormalities in Legg-Calvé-Perthes disease . J Bone Joint Surg Am . 2010 ; 92-A ( 1 ): 121 – 128 . Crossref PubMed Google Scholar
27. Loder RT , Schwartz EM , Hensinger RN . Behavioral characteristics of children with Legg-Calvé-Perthes disease . J Pediatr Orthop . 1993 ; 13 ( 5 ): 598 – 601 . PubMed Google Scholar
28. Berman J , Aran A , Berenstein-Weyel T , Lebel E . Exploring the association between Legg-Calvé-Perthes disease and attention deficit hyperactivity disorder in children . Isr Med Assoc J . 2016 ; 18 ( 11 ): 652 – 654 . PubMed Google Scholar
29. Hailer YD , Nilsson O . Legg-Calvé-Perthes disease and the risk of ADHD, depression, and mortality . Acta Orthop . 2014 ; 85 ( 5 ): 501 – 505 . Crossref PubMed Google Scholar
30. Mattheisen M , Grove J , Als TD , et al. Identification of shared and differentiating genetic architecture for autism spectrum disorder, attention-deficit hyperactivity disorder and case subgroups . Nat Genet . 2022 ; 54 ( 10 ): 1470 – 1478 . Crossref PubMed Google Scholar
31. Schmidt M , Schmidt SAJ , Sandegaard JL , Ehrenstein V , Pedersen L , Sørensen HT . The Danish National Patient Registry: a review of content, data quality, and research potential . Clin Epidemiol . 2015 ; 7 : 449 – 490 . Crossref PubMed Google Scholar
Author contributions
P. L. Hedley: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing
U. Lausten-Thomsen: Conceptualization, Methodology, Writing – review & editing
K. M. Conway: Methodology, Writing – review & editing
K. Hindsø: Methodology, Writing – review & editing
P. A. Romitti: Methodology, Writing – review & editing
M. Christiansen: Conceptualization, Writing – review & editing
Funding statement
The authors disclose receipt of the following financial or material support for the research, authorship, and/or publication of this article: this work was funded in part by grant U01 DD001307 awarded to the Iowa Center for Birth Defects Research and Prevention (paid to the University of Iowa) from the US Centers for Disease Control and Prevention.
ICMJE COI statement
P. A. Romitti reports a grant (paid to University of Iowa) from the US Centers for Disease Control and Prevention, related to this study.
Data sharing
The data for this study are publicly available at https://biobanks.dk/.
Acknowledgements
The authors are grateful to the team at the Department of Digital Infrastructure, Statens Serum Institut for their work in developing and maintaining the Danish Biobank Register online interface.
Open access funding
The authors report that the open access funding for their manuscript was self-funded.
Open access statement
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
Supplementary material
Tables listing the major congenital anomalies and syndromes, as well as neurodevelopmental disorders assessed regarding risk of developing Perthes’ disease.
Social media
Follow Statens Serum Institut on X @SSI_DK
Follow U. Lausten-Thomsen on X @UlrikLausten
Follow M. Christiansen on X @MIC_CPH
This article was primary edited by S. P. F. Hughes.
