Incidence and gender differences of slipped capital femoral epiphysis in the Netherlands from 1998–2010 combined with a review of the literature on the epidemiology of SCFE

Abstract

Abstract

Purpose

The incidence of slipped capital femoral epiphysis (SCFE) among children living in the Netherlands has never been published.

Methods

The national hospitalization registration system of the Netherlands was searched for the incidence of surgical procedures for SCFE in the Netherlands among different pediatric age groups between 1998 and 2010. International Classification of Diseases, 9th Revision codes were used.

Results and conclusion

The incidence of surgical procedures for SCFE during the last decade was 11.6 per 100,000 children aged 5 to 19 years. No statistical difference in the incidence of SCFE was found between boys and girls, although the incidence of SCFE did significantly increase in girls during the study period. Based on our analysis, the Netherlands appears to be the first country in which no difference in the incidence of SCFE among boys and girls has been reported. However, during the study period there has been a concomitant increase in the number of girls with SCFE.

Introduction

Slipped capital femoral epiphysis (SCFE) is a disorder of the proximal femur that occurs mainly in peripubertal children. It is defined as the displacement of the femoral neck and shaft relative to the femoral head in the growth plate in which the proximal femoral neck and shaft move anteriorly and rotate externally relative to the femoral head, while the femoral head remains in the acetabulum. SCFE has been described as the most common hip disorder of adolescent children in the USA [1]. Boys are more susceptible than girls in developing this disorder (Table 1). It is most often diagnosed in obese children around puberty and in children with endocrinopathies or chronic systemic diseases. In the Netherlands, the incidence of SCFE has never been published. To gain an understanding of SCFE and to compare its incidence in other countries we performed a literature search aimed at determining its incidence in different countries among children of different ages. We then performed a Dutch population-based search to determine the incidence of SCFE in the Netherlands and whether there was a difference between boys and girls. National data on body length and body weight of the children were used to interpret the findings. We compared our results with those of other studies that provide a local incidence of SCFE (Table 1).

Table 1

Literature search on the incidence slipped capital femoral epiphysis

First author	Year of publication	Location of study	Incidence SCFE:100.000	Age of children enrolled in study (years)	Years of incidence noted	Male:female ratio
Henrikson [5]	1969	Gothenburg, Sweden	2.0–13.0	7–16	1947–1966	1.9
Kelsey [7]	1970	Connecticut	3.4	<25	1960–1967	2.7
			10.1	8–17
		New Mexico	0.7	<25	1960–1967	1.7
			2.1	8–17
Hagglund [4]	1984	Southern Sweden	61.0 (M), 30.0 (F)	5–23	1910–1982	2.3
Jerre [6]	1996	Gothenburg, Sweden	79.0	7–17	1946–1992	1.8
Loder [15]	1996	International multicenter study				1.4
Noguchi [10]	2002	Japan	2.2 (M), 0.8 (F)	10–14	1997–1999	3.1
Lehmann [1]	2006	USA	10.8	9–16	1997 and 2000	1.7
Benson [2]	2008	New Mexico	6.0	8–17	1995–2006	1.9
Lim [9]	2008	Singapore	1.2	5–14	1994–2006	4.1
Murray [14]	2008	Scotland	9.7	6–18	1981–2000	1.7
Song [11]	2009	Korea	0.3	10–14	1989–2003	3.1
Larson [8]	2010	Midwestern American (Olmsted) county	8.8	9–16	1965–2005	1.8
Nguyen [3]	2011	South Australia	2.8–8.2	10–19	1988–2007	1.7
Witbreuk (current study)	2012	Netherlands	11.6	5–19	1998–2010	1.1

SCFE slipped capital femoral epiphysis, M male, F female

Materials and methods

Patients

We performed a search of the national hospital registration system of the Netherlands, i.e. the “Landelijke Medische Registratie (LMR), PRISMANT kubus Ziekenhuis statistiek”, which registers all patients admitted to Dutch hospitals. Only age groups and sex are listed in this registration system. Our study period covered the period 1998–2010, and the search items were the diagnosis of non-traumatic SCFE [NT SCFE; International Classification of Diseases, 9th Revision (ICD 9) code 732.2] and traumatic SCFE (T SCFE; ICD 9 code 820.01) as well as a combination of these two groups. The ICD codes only contain figures for the age categories 1–4 years, 5–9 years, 10–14 years and 15–19 years. We restricted ourselves to the age group 5–19 years. NT SCFE can be confused with T SCFE; therefore we counted the entities separately and as a combined group to these different codes. For the population figures we used Statline (Centraal Bureau voor Statistiek, the Netherlands).

To determine the size of the population at risk over the entire study period 1998–2010, we considered all children aged between 5 and 19 years in that period. First, we considered different age-cohorts (one cohort for each age 5, 6,…19 years) that were at risk for developing SCFE in 1998. The follow-up period for each age-cohort was defined from 1998 onwards either up to the year the patient turned 19 years old or until 2010, whichever came first. We also considered different entry-cohorts (one cohort for each year from 1999, 2000,…2010) that consisted of children who turned 5 years old in that year. For each entry-cohort, the follow-up period was defined from the year of entry until 2010. We estimated the number of boys and girls in each cohort as the average number of boys and girls in that cohort over its follow-up period. The total number of children at risk over the period 1998–2010 was estimated by summing each estimated cohort-size (Table 2).

Table 2

Population demographic statistics for the Dutch population aged 5–19 years between 1998 and 2010

Statistical analysis

Incidence was defined as the population-proportion of surgical procedures for SCFE in children. The difference in the incidence rates between boys and girls was assessed by relative risks, with corresponding 95 % confidence intervals (95 % CIs). The Cochran–Armitage test for trend was used to test for a linear trend of the yearly incidence rates over time. A 95 % CI not containing the value 1 or a p value of <0.05 was considered to be significant, while a p value of <0.1 was considered to be a trend. All statistical analyses were performed using SPSS ver. 15.0 (SPSS, Chicago, IL).

Results

Incidence

The total number of surgical procedures for SCFE was 609 in the period 1998–2010 for children between 5 and 19 years in the Netherlands. This translated into an incidence of 11.6 surgical procedures per 100,000 children over that period in the combined NT and T SCFE group. When the T SCFE group was not included in the analysis, the incidence of surgical procedures for NT SCFE was 8.8 per 100,000 children. The total incidence over the combined group for children aged 10–19 years was 13.7 surgical procedures per 100,000 children.

Gender ratio

In the combined group of NT and T SCFE the relative risk of exposure of boys compared to girls aged 5–19 years was 1.11 (95 % CI 0.95–1.30) over the period 1998 to 2010. The 95 % CIs over all years imply that there was no significant difference in the risk of being exposed to SCFE between boys and girls (Table 3).

Table 3

Relative risk of being exposed to traumatic and non-traumatic SCFE for boys compared to girls aged 5–19 years, per year and over the entire study period 1998–2010

Study year	RR [95 % CI]: M vs. F
1998	1.49 [0.82–2.69]
1999	1.05 [0.58–1.89]
2000	1.47 [0.76–2.81]
2001	1.73 [0.94–3.19]
2002	1.22 [0.72–2.06]
2003	1.57 [0.86–2.87]
2004	0.70 [0.39–1.26]
2005	1.05 [0.57–1.93]
2006	1.20 [0.62–2.31]
2007	0.99 [0.58–1.70]
2008	0.88 [0.49–1.56]
2009	0.87 [0.54–1.41]
2010	0.91 [0.51–1.64]
1998–2010	1.11 [0.95–1.30]

RR relative risk, CI confidence interval

The annual incidence of surgical procedures for SCFE (Fig. 1) increased significantly for girls aged between 5 and 19 years (p = 0.034). The incidence of the total group (boys and girls combined) did not increase significantly (p = 0.384) nor did the incidence for boys (p = 0.438).

Fig. 1

Annual incidence of traumatic slipped capital femoral epiphysis (T SCFE) and non-traumatic SCFE (NT SCFE) for boys and girls aged 5–19 years and for the two groups combined for the period 1998–2010

The same results were seen in only the NT SCFE group, where the relative risk over the 13 years was 0.99 (95 % CI 0.82–1.18) (Table 4). The annual incidence of NT SCFE for girls also increased significantly (p < 0.001), but not for boys or the combined group (p = 0.309 and p = 0.293, respectively) (Fig. 2).

Table 4

Relative risk of being exposed to non-traumatic SCFE for boys compared to girls aged 5–19 years, per year and over the entire study period 1998–2010

Study year	RR [95 % CI]: M vs. F
1998	1.23 [0.61–2.47]
1999	0.96 [0.48–1.91]
2000	1.56 [0.74–3.31]
2001	1.67 [0.82–3.40]
2002	1.30 [0.72–2.36]
2003	1.31 [0.69–2.50]
2004	0.76 [0.40–1.47]
2005	1.09 [0.53–2.24]
2006	0.89 [0.43–1.85]
2007	0.68 [0.36–1.26]
2008	0.64 [0.32–1.25]
2009	0.92 [0.53–1.59]
2010	0.59 [0.30–1.18]
1998–2010	0.99 [0.82–1.18]

Fig. 2

Annual incidence of NT SCFE for boys and girls aged 5–19 years and for the two groups combined for the period 1998–2010

Age

In the combined NT and T SCFE group, 499 of the 609 patients (82 %) with SCFE were between 10 and 14 years of age (Fig. 3). The T SCFE group contained more boys in the age group 10–14 years, whereas the NT SCFE group contained more girls (Fig. 4). In the age group 10–14 years, the incidence rates for boys and girls did not differ significantly (relative risk 0.92, 95 % CI 0.77–1.10). In the age group 15–19 years, the incidence rates did differ: 4.3:100,000 for boys and 0.6:100,000 for girls. One possible explanation for this effect could be the earlier closure of the growth plate in girls (around 14 years of age) than in boys (around 16 years of age). In the age group 5–9 years, we observed the opposite effect: the incidence for boys was 0.6:100,000 and that for girls was 1.2:100,000.

Fig. 3

Percentage of patients (boys and girls combined) with NT and T SCFE divided into different age groups for the entire study period 1998–2010

Fig. 4

Percentage of patients with NT SCFE and T SCFE per gender and for different age groups over the entire study period 1998–2010

Discussion

Our data show that the presumed gender dominance in the incidence of SCFE was not confirmed in the interval 1998–2010 in the Netherlands. The data also show an increase in the incidence of surgical procedures for SCFE in girls starting from 1998.

The strong points of this study are its nationwide coverage and the use of uniform diagnostic criteria.

One limitation to our study is our use of national Dutch hospital registration data, which did not allow us to take into account the etiology of SCFE (i.e. endocrinological, mechanical or other causes). NT SCFE is sometimes confused with T SCFE and often treated as such. Therefore, we combined the T/NT SCFE groups and also considered the NT SCFE group separately. The T SCFE group may have contained some high-energy epiphyseal fractures, leading to an overestimation of T SCFE. However, we expect this overestimation to be very small because of the low frequency of this type of fracture in general and the low number of high-energy lesions in the Netherlands. A second limitation to our study is that the statistics did not show whether patients had a unilateral slip or a bilateral slip nor could we find information on different cultures or seasonal variations.

Incidence

It is difficult to compare the incidence we found in the Netherlands with the incidences reported in the literature because data on different age groups are presented. In addition, different estimation methods have been used. A few studies use ICD codes, as we did [1–3], but most studies used the hospital information system because this provided a defined framework [4–11]. The incidence we found is comparable to data from Sweden and the USA.

In general, it would appear that Asian countries like Japan, Singapore and Korea [9–11] have a lower incidence of SCFE than Western countries. Speculations on the reasons for these differences have focused on ethnic or nutritional differences [9, 11] (see Table 1).

Gender ratio

All published studies report a male:female ratio with preponderance for boys, ranging from 1.4 to 4.1 (Singapore) [9].

Our study is unique in determining that the 95 % CIs over each of the past years in the Netherlands imply that there was no significant difference in the risk of being exposed to NT and T SCFE between boys and girls. Despite these figures, however, there has been an increase in the incidence of SCFE in girls over the past 13 years: the annual incidence increased significantly for girls aged 5 to 19 years. The incidence for boys did not increase significantly nor did that for the combined group.

Figure 1 shows that the difference is small between girls and boys on an annual basis but that over the years of the study period there was an increase in the incidence of SCFE in girls compared to boys. To date, our study is the first to report such a difference at the national level while concomitantly describing an increase in the incidence of SCFE in girls over the last decade.

A decrease of male predominance has also been found in Japan and Sweden. Despite the low incidence in Japan, a fivefold increase in boys and tenfold increase in girls was found between 1974 and 1999 [10]. In Sweden, the predominance of boys has decreased from 85–90 to 60–65 % during the first decade of this century [4, 12].

Age

We found the same distribution of children with SCFE in different age groups as reported in the other studies. All studies describe a peak in incidence around the age of onset of puberty. In our group, the 609 patients with SCFE were divided as follows: 5 % (29) in the 5- to 9-year-old group, 82 % (499) in the 10- to 14-year-old group and 13 % (81) in the 15- to 19-year-old group (see Fig. 3). A majority of studies describe a decrease in the mean age of onset during recent years. Hagglund et al. [4] suggest that the decreasing age at onset of SCFE is probably caused by the onset of puberty at an earlier age during the last century and by the increased awareness of parents, the social environment and physicians, leading to an earlier diagnosis of SCFE.

Body length

In the Netherlands nation-wide growth studies have been conducted since 1955. The fifth Dutch Growth study was carried out between 2008 and 2010. These results can be compared to earlier studies in the Netherlands. Schonbeck et al. [13] report that there has been no increase in the body length of boys and girls between 1997 and 2009 in the Netherlands.

Body weight

Figures obtained from the Factsheet Results Fifth National Growth Study TNO (June 10, 2010) show that there has been a trend for an increase in overweight and obesity from 1997 to 2009. A comparison of data for the period 1997–2009 shows that 9.4 versus 13.3 % of Dutch boys and 11.9 versus 14.9 % of Dutch girls aged 2–21 years were overweight and 0.9 versus 1.8 % of the boys and 1.6 versus 2.2 % of the girls were classified as obese [13]. Although the boys have increased in body weight more than the girls, the incidence of SCFE for boys—but not for girls—seems to have stabilized. The increase of SCFE in girls in the past decade may also be due to these increases in body weight, but we have no data to support such a hypothesis for boys. Increasing obesity in adolescents has been found in many different countries, as has an increase in SCFE [2, 10, 11, 14]. In major cities in the Netherlands overweight prevalence has stabilized, with no increase in the body mass index, which leads to the hope that the rising trend in overweight is starting to turn [13].

Conclusion

Based on the results of our study, in the Netherlands there was no difference in the incidence of surgical procedures for SCFE between boys and girls during the period 1998–2010. In the past 13 years there has been an increase in the incidence of surgical procedures for SCFE among girls in the combined NT/T SCFE group and in the NT SCFE group alone. The incidence in boys and in the combined group has not increased. From other national registries we know that there has been an increase in the number of Dutch children who suffer from overweight and obesity, while the body length has not increased the past 10 years.

The incidence determined in our study seems to be comparable with that of other Western countries such as the USA and Sweden.