Abstract
Background:
Osteochondritis dissecans (OCD) of the elbow is a disorder of subchondral bone and articular cartilage, of which the incidence among children is not clearly known.
Purpose:
To assess the demographics and epidemiology of OCD of the elbow among children.
Study Design:
Cross-sectional study; Level of evidence, 3.
Methods:
A retrospective chart review of an integrated health system for the years 2007 through 2011 was performed for patients with elbow OCD aged 2 to 19 years. Lesion location, laterality, and all patient demographics were recorded. OCD incidence was determined for the group as a whole as well as by sex and age group (2-5, 6-11, 12-19 years). Patient differences based on age, sex, and ethnicity were analyzed, and multivariable logistic regression models were used to assess the risk of elbow OCD by group.
Results:
A total of 37 patients with 40 OCD lesions fit the inclusion criteria. No lesion was found among 2- to 5-year-olds. A majority of lesions (n = 39, 97.5%) were in the capitellum, and 1 (2.5%) was in the trochlea. Twenty-five patients (67.6%) had right-sided lesions; 9 (24.3%), left-sided; and 3 (8.1%), bilateral. The incidence of elbow OCD for patients aged 6 to 19 years was 2.2 per 100,000 overall and 3.8 and 0.6 per 100,000 for males and females, respectively. The majority of OCD cases were seen in those aged 12 to 19 years, with an incidence of 3.4 per 100,000 versus 0.38 among 6- to 11-year-olds. Multivariable logistic regression analysis revealed a 21.7-times increased odds ratio of elbow OCD among patients aged 12 to 19 years versus 6 to 11 years, and males had a 6.8-times greater odds ratio of elbow OCD than females (P < .0001 for both). Based on race and ethnicity, non-Hispanic whites had the highest incidence of elbow OCD as compared with all other ethnic groups.
Conclusion:
In this population-based cohort study of pediatric elbow OCD, males had almost 7 times the risk of elbow OCD as compared with females, and 12- to 19-year-olds had nearly 22 times the risk of elbow OCD versus 6- to 11-year-olds. In keeping with many prior studies, the majority of patients had right-sided lesions.
Franz Konig 18 first described osteochondritis dissecans (OCD) in 1888. Incidence studies on OCD were performed by Linden 23 for OCD of the knee and by Kessler et al 14,15 on OCD of the knee and ankle. Anecdotally, elbow OCD is thought to be associated with overuse among young athletes, such as those in overhead throwing sports and gymnastics, and treatment often includes reducing stresses placed on the joint by restricting participation to allow healing. However, there are no true incidence studies on OCD of the elbow or any epidemiologic studies of this disease in a closed, self-contained population. Thus, the purpose of the present study was to evaluate the demographics and epidemiology of elbow OCD in an extremely large population-based cohort of children and adolescents in Southern California.
Methods
All patients aged 2 to 19 years from the entire database of patients enrolled as members of Kaiser Permanente Southern California from January 2007 until August 2011 (N = 1,068,215) were assessed per the methods described by Kessler et al 14,15 for OCD of the knee and ankle. From this population, we retrospectively reviewed the electronic health records of inpatient, outpatient, and emergency department encounters for the first occurrence of an International Classification of Diseases, Ninth Revision (ICD-9) code for elbow OCD. The ICD-9 codes used to identify patients with OCD of the elbow included 732.3, 732.7, 732.9, and 733.9. Inclusion criteria included isolated OCD lesions of the elbow among patients aged 2 to 19 years at the time of diagnosis. Exclusion criteria included osteochondral fractures, Panner disease, and all other intra-articular cartilaginous, ligamentous, or bony injuries. Radiographs were reviewed by a fellowship-trained pediatric orthopaedic surgeon (J.M.W.) to ensure that a true OCD lesion existed. All cases of OCD identified on radiographs were confirmed with magnetic resonance imaging (MRI). It was also noted whether patients had radiographs of the contralateral elbow within 1 year before or after the diagnosis of OCD. Diagnosis of OCD was confirmed per findings of focal subchondral lucency and edema versus more diffuse necrotic lesions among younger patients, characteristic of Panner disease. Ultimately, 37 patients were found to fit the inclusion and exclusion criteria.
Patient age at diagnosis, sex, race and ethnicity, joint involvement, side involved, and specific location within the elbow were included as variables. Specific locations included the capitellum and trochlea. Age at diagnosis was obtained from enrollment records and was categorized as 2 to 5, 6 to 11, or 12 to 19 years to reflect preschool, elementary school, and middle/high school, and race/ethnicity was documented in the same manner as prior OCD epidemiology studies 14,15 (Table 1). Incidence data were calculated only for the year 2010 because (1) the transition of all Kaiser Southern California hospitals and clinics to a fully integrated electronic coding system (Current Procedural Terminology and ICD-9) was incomplete until the end of 2009 and (2) only a portion of the year 2011 was used to identify patients. All information on patient participation in competitive sports was recorded, including whether there had been an antecedent injury or trauma noted by the patient or parents prior to the onset of OCD. The specific type of trauma was recorded, as was the specific type of competitive sport.
Demographics of Patients Included in the Elbow OCD Modeling a
a Values are presented as n (%) unless noted otherwise. P value based on t test for continuous variables and Mantel-Haenszel chi-square for categorical variables. OCD, osteochondritis dissecans; Q1, Q3, interquartile range.
The distribution and frequency of elbow OCD and joint location were calculated, and chi-square tests were performed to determine differences by age and sex. The associations between elbow OCD and race/ethnicity, age, and sex were assessed with multivariable logistic regression models to estimate odds ratios (ORs) and 95% CIs. An alpha level of 0.05 was used to determine statistical significance, and SAS Enterprise Guide (v 4.2; SAS Institute) was used for all analyses.
Results
No patients aged ≤5 years were found to have elbow OCD, and only 1 patient <10 years of age was diagnosed with elbow OCD. A total of 37 patients were found to have elbow OCD, and patient demographics are summarized in Tables 1 and 2. There were 32 male and 5 female patients, for a male:female ratio of 6.4:1. Overall, 25 (67.6%) patients had right-sided lesions; 9 (24.3%), left-sided (Table 2); and 3, bilateral (8.1%). Radiographs of the contralateral elbow within 1 year of diagnosis were available in 70% of the patients, and they demonstrated no evidence of contralateral OCD. There were 12.3 times as many OCD lesions in the 12- to 19-year-old group as in the 6- to 11-year-old group, with 37 and 3 lesions, respectively. The male:female ratio differed significantly between age groups, with 86.5% and 33.3% elbow OCD among males in the older and younger age groups, respectively (P < .01). The mean ± SD and median ages of disease onset for the entire group were 13.8 ± 2.0 and 14.0 years, respectively. Most lesions involved the capitellum (97.5%), with 1 involving the trochlea (2.5%) (Table 3). The location within the elbow did not differ significantly by sex, but logistic regression analysis demonstrated a significantly higher proportion of capitellar lesions in the older group versus the younger (P = .03).
Demographics for Patients With Elbow Osteochondritis Dissecans a
a Q1, Q3, interquartile range.
Location of Elbow Osteochondritis Dissecans by Sex and Age Group
In the analysis of sports participation and association of trauma, 31 of 37 patients were documented to participate in competitive sports (Table 4). The most common competitive sports were baseball and football, with 25% participating in baseball only, 25% in football only, and 6.25% in both. Only 10 patients (27.5% of all elbows) with OCD initially presented after having had an injury to the elbow in the prior weeks or months, after which the parents claimed the pain in the elbow began. Of note, among all 35 patients with unilateral elbow OCD, radiographs were performed because of ongoing pain. For 2 of the 3 patients with bilateral OCD, however, the findings on 1 side were purely incidental in an asymptomatic elbow. One of these 2 patients with incidentally found OCD was a competitive gymnast, whereas the second did not document participation in competitive sports.
Patient-Reported Competitive Sports Participation a
a Two patients played both football and baseball; 1 patient played both football and basketball.
In the incidence analysis, the incidence of elbow OCD for patients aged 6 to 19 years was 2.2 per 100,000 overall (Table 5). The incidence for males was >6 times that of females, at 3.8 and 0.6 per 100,000, respectively. The older group represented the majority of elbow OCD, with an incidence of 3.4 per 100,000 among 12- to 19-year-olds versus 0.38 among 6- to 11-year-olds. Older males had the highest incidence of disease, at 6.2 per 100,000.
Incidence Rates for Osteochondritis Dissecans of the Elbow by Age for 2010
Logistic regression analysis further confirmed the univariate analysis. The OR of having elbow OCD was 21.7 times greater for patients aged 12 to 19 years compared with patients aged 6 to 11 years (P < .0001; 95% CI, 5.2-90.3) (Table 6). In addition, the odds of a male having elbow OCD was 6.8 times greater than that of females (P < .0001; 95% CI, 2.6-17.4). Consistent with this, older males had the highest risk of OCD, with ORs of 39.0 and 37.1 as compared with 6- to 11-year-old males and females, respectively (P = .0003 and .0004; 95% CI, 5.3-285.8 and 5.1-272.1). The OR was also higher for older males versus older females (OR, 8.2; P < .0001; 95% CI, 2.9-23.3).
Multiple Logistic Regression Models for Elbow Osteochondritis Dissecans
a Age groups are indicated by year range: 6-11 and 12-19.
bP value indicates the significance of the effect in predicting elbow osteochondritis dissecans.
In the OR analysis done by race and ethnicity, the risk of elbow OCD was greatest for non-Hispanic whites, who had 4.4 and 9.4 times the risk of elbow OCD as compared with Hispanics (P = .0002; 95% CI, 2.0-9.6) and those with unknown/other ethnicity (P = .003; 95% CI, 2.2-39.8), respectively. The risk for non-Hispanic whites trended toward being greater than that for Asians and blacks but did not reach statistical significance (P = .09 for both) (Table 6). Incidence analysis by ethnicity confirmed the highest incidence for non-Hispanic whites, at 7.4 per 100,000 (Table 7).
Incidence Rates for Elbow Osteochondritis Dissecans by Ethnicity for 2010
Discussion
Takahara et al 40 provided the largest study on OCD of the elbow, reporting on 106 patients. Their study was a retrospective review to assess treatment outcomes, which found that patients with open capitellar growth plates healed completely with rest. The Takahara et al 40 study, however, included a mixture of children and adults. In addition, this and all other studies of elbow OCD failed to assess the incidence of disease or demographics in a self-contained population.
The present study found a lower incidence of elbow OCD as compared with the incidences of knee OCD in the studies of Linden 23 and Kessler et al 14 and ankle OCD in the study by Kessler et al. 15 In terms of sex, prior literature on elbow OCD among children and adults has demonstrated a predilection toward males. ¶ The present epidemiology study clearly confirms this finding, with a 6.4:1 male:female ratio and a 6.8-times greater OR of disease among males. In terms of age of presentation, the majority of studies have shown that elbow OCD among children usually occurs at ≥10 years. # The present study unequivocally confirms that older children have the higher frequency of the disease, with a nearly 22-times increased risk of OCD among 12- to 19-year-olds versus 6- to 11-year-olds.
Prior studies on laterality have indicated that OCD tends to affect the right elbow more frequently.** Our study also demonstrates this right-side dominance, with just over two-thirds of patients having right elbow OCD, with a relatively low incidence of bilateral disease. We assume that the right-sided dominance is due to the fact that the majority of the population is right-handed. In terms of location of elbow OCD, the literature has consistently shown that lesions are usually in the capitellum. †† Our epidemiologic study supports this, with 97.5% of the sample having capitellar lesions and only 2.5% (1 patient) having a trochlear location. In addition, our study supports the findings of many prior studies on elbow OCD suggesting that these injuries tend to occur among patients who participate in competitive sports, ‡‡ although rarely is the OCD related to an actual acute injury. We suspect that development of elbow OCD may be related to the repetitive microtrauma produced during activities such as gymnastics or overhead throwing (baseball and football). The adolescent-male, dominant-sided predilection for elbow OCD could then be explained by increased participation in competitive sports during an important stage in joint development.
Little is known about the association between race/ethnicity and OCD of the elbow. The present study sheds light on the risk of elbow OCD by ethnicity and, to our knowledge, is the first to do so. The logistic regression analysis demonstrated that, as with ankle OCD, 15 non-Hispanic whites have a higher risk of disease as compared with all other races/ethnicities, although it is important to note that this risk did not reach statistical significance when compared with Asians, blacks, and Hispanics individually. The increased risk of elbow OCD among whites was confirmed by the incidence analysis.
Our study has a number of weaknesses. Similar to our prior studies on ankle 15 and knee OCD, 14 the incidence data were obtained only from 2010, while the demographic/epidemiological data were obtained from January 2007 to August 2011 (for reasons cited previously). True incidence may be underestimated, as some lesions may remain asymptomatic and go undiagnosed. Other weaknesses include the retrospective nature of the study and the fact that in the multivariable logistic regression modeling by ethnicity/race, socioeconomic status, and sports participation measures were not assessed as potential confounders. In addition, our population demographics may not be representative of other populations, and the small numbers of certain ethnic groups in our study may limit the extrapolations that can be made.
Conclusion
The present study represents the first and only true incidence and epidemiologic study of OCD of the elbow among patients aged ≤19 years. Similar to our prior study of knee OCD 14 and in contrast to our study of ankle OCD, 15 males were found to have an increased incidence of disease and a significantly increased risk of elbow OCD as compared with females. However, similar to both these studies, teenagers had a markedly increased risk of disease versus younger patients. In addition, this study demonstrates that, as with ankle OCD, non-Hispanic whites have the highest incidence of elbow OCD.
Footnotes
One or more of the authors has declared the following potential conflict of interest or source of funding: J.I.K., J.M.W., and K.G.S. are members of the ROCK (Research in OsteoChondritis of the Knee) research group, which receives unrestricted educational grants from Allosource of Vericel. K.G.S. has received grants from DePuy Orthopaedics and Sanofi-Aventis and hospitality payments from DePuy Orthopaedics. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
Ethical approval for this study was obtained from a subcommittee of the Kaiser Permanente Southern California Institutional Review Board.
Notes
Acknowledgment
The authors thank Heidi Fischer, PhD, for her help with biostatistical analysis.
