Is There a Connection Between Attention Deficit Hyperactivity Disorder and Osteochondritis Dissecans?

Abstract

Background:

To determine if children with Osteochondritis Dessicans (OCD) lesions of the distal femur are more likely to have a co-morbid diagnosis of Attention Deficit/ Hyperactivity Disorder (ADHD) than age matched controls and to assess the impact of ADHD on OCD outcomes.

Methods:

A retrospective chart review of patients treated at a single tertiary care hospital between 2000-2012 was performed. Charts were reviewed for a diagnosis of OCD of the distal femur in all skeletally immature patients (males < 16 years and females < 14 years). These were then screened for a comorbid diagnosis of ADHD. Age-matched controls with anterior knee pain without OCD were then reviewed to determine if ADHD was more common in the OCD population. Treatment and outcomes of the OCD lesions were then compared in children with and without ADHD.

Results:

The prevalence of ADHD was 23% in patients with OCD lesions and was significantly greater than the 11% found in the anterior knee pain age-matched controls (p<0.05). The average grade of lesions at presentation was similar in both groups (2.2 ADHD vs 2.1 no ADHD) however, at final follow-up, the average OCD grade was significantly worse for children with ADHD (1.4 vs 0.7, p<0.004).

Conclusion:

There is a significantly higher prevalence of ADHD in children with OCD lesions compared with age-matched controls. This study suggests children that with osteochondritis dessicans and ADHD may not have as favorable treatment course as children without the hyperactivity disorder.

Level of Evidence: III

Introduction

Osteochondritis Dissecans (OCD) is a term used to describe the separation of a segment of articular cartilage and subchondral bone from the remaining articular surface. Juvenile OCD (JOCD) describes a lesion found in skeletally immature children with the highest incidence occurring between the ages of 10 and 20. While OCD lesions can occur in other joints, the most common location is the knee joint on the lateral half of the medial femoral condyle. Knee OCD lesions are found more frequently in children involved in organized sports and is twice as common in males as in females.^1-4 The incidence of JOCD has been on the rise concordant with increasing participation in youth athletics.²

To the authors’ knowledge, no evaluation of pathologic hyperactivity has ever been linked as a potential cause or risk factor in developing an OCD lesion. However, similar orthopedic pathologies that involve altered blood supply, such as Legg-Calve-Perthes disease, have shown an association with such behaviors.⁵ Thus, our objective was to determine if the behavioral condition of Attention Deficit and Hyperactivity Disorder and/ or its treatment effects a patient’s likelihood for developing and healing osteochondritis dissecans lesions of the distal femur.

Methods

After Institutional Review Board approval for this study, a retrospective chart and radiographic review of consecutive patients treated for JOCD of the knee at a single teriary care institution (2000-2012) was performed. Inclusion criteria included diagnosis of OCD of distal femur, skeletal immaturity which is typically males less than or equal to 16 years of age and females less than or equal to 14 years of age. All included individuals had confirmed open tibial and femoral physes on radiographic review. Exclusion criteria included males older than 16 years of age and females older than 14 years of age. Patients with cartilage defects from acute trauma, OCD lesions outside the femur, and patients initially treated at an outside hospital were also excluded.

Data collected included patient age, gender, height, weight, BMI, behavioral comorbidities, medications, treatment (nonoperative vs. operative), state of the physis, radiographic size of lesion, follow-up length, and radiographic grade of lesion based on plain radiographs or MRI. A single reviewer (K.D.) reviewed all plain radiographs and/or MRI at presentation and final follow-up. The accuracy of measurements on plain films has previously been confirmed.⁶

Patient age was determined by their chronological age at the initial visit. Length of follow-up was determined from the time of their initial clinic visit to final clinic visit. At each visit patient height and weight measurements were taken and BMI was calculated. Individuals with ADHD were identified by a previous definitive diagnosis within the medical record or by reported use of stimulant medication typically used to treat ADHD. Treatment at the initial visit was recorded as operative or non-operative. Further chart review determined if the patient failed the initial non-operative treatment and crossed over to surgical treatment.

To determine the prevalence of ADHD in our general patient population, we used age-matched controls that presented to the clinic with a diagnosis of anterior knee pain. Radiographs confirmed these controls did not have JOCD of the knee. In a similar fashion, individuals with ADHD were identified by a previous definitive diagnosis within the medical record or by reported use of stimulant medication typically used to treat ADHD.

OCD Measurements: All OCD lesions were graded according to the Clanton Classification:⁷ Grade 1, depressed osteochondral fracture; Grade 2, osteochondral fragment attached by an osseous bridge; Grade 3, detached non-displaced fragment; and Grade 4, displaced fragment. Plain radiographs or MRI were utilized to determine whether the physis was closed or open. Length of the fragment was determined by measuring the OCD lesion with the ruler function on the PACS system (McKesson, San Francisco, CA) on the lateral x-ray or the widest measurement on sagittal MRI images. Width of the fragment was determined by measuring the OCD lesion with the ruler function on the PACS system on the AP x-ray or the widest measurement on coronal MRI images. The area of the lesions was calculated by multiplying the length and width determined of the lesion (Figure 1a and 1b), a method that was tested and validated by Wall et al.⁸

Figure 1a and 1b.

Measuring the length and width of the OCD lesion.

Statistical Analysis

The prevalence of ADHD among controls and patients with OCD lesions was compared using a chi-squared test. Age, height, weight, and body mass index were compared between groups using the standard t-test. Gender, race, and success of non-operative treatment were compared between groups using Fisher’s exact test. Lesion characteristics over time (length, width, area, and grade), follow-up time, and surgery were compared using Repeated Measures ANOVA.

Results

Eighty-two patients met the inclusion criteria from chart review. Nineteen patients (23%) with OCD also had a diagnosis of ADHD. The prevalence of ADHD among age- matched controls with a diagnosis of anterior knee pain but no OCD was 11% (9/80). The difference in the prevalence of ADHD between patients with anterior knee pain and OCDs was significant (p<0.05). No differences in age, height, weight, BMI, or race, was found between those children with and without ADHD and OCD lesions (Table 1).

Table 1.

Demograhic Characteristics of Children with OCD Lesions

	ADHD (n=19)	No ADHD (n=63)	p-value
Age (yr)	13.6 (2.0)	13.4 (2.0)	0.62
Height (in)	64.7 (6.6)	63.6 (5.8)	0.54
Weight (lb)	131.5 (47.5)	124.8 (36.7)	0.58
BMI	21.4 (4.8)	21.2 (3.7)	0.86
Female Gender (%)	0 (0.0)	16 (25.4)	0.02*
Caucasian Race (%)	13 (68.4)	56 (88.9)	0.07

Age, Height, Weight, and BMI using standard t-test. Gender and Race using Fisher’s exact test. * = Statistically significant (p<0.05). Reported as mean (SD), or n (%)

With regards to the OCD lesions, there was no difference in the length, width, area, or grade at presentation between cohorts (Table 2). However, at final follow-up the average OCD grade was significantly worse for children with ADHD (1.4 vs 0.7 p=0.004). Children without ADHD demonstrated a significant improvement in the grade of their lesions over time (2.1 to 0.7, p<0.001), whereas the children with ADHD only trended towards an improvement in grade over time (2.2 to 1.4 p=0.06). There was no significant difference in follow-up length for the patients with and without ADHD (1.56 to 1.50 years p=0.88).

Table 2.

Characteristic of OCD Lesions

	Time	ADHD	No ADHD	p-value
Length (mm)	Presentation	17.8 (7.9)	18.9 (5.7)	0.53
	Follow-up	10.2 (10.2)	7.9 (8.9)	0.36
Width (mm)	Presentation	12.9 (5.6)	13.8 (5.0)	0.44
	Follow-up	6.9 (7.0)	5.9 (6.8)	0.60
Area (mm2)	Presentation	257 (192)	276 (153)	0.61
	Follow-up	132 (185)	103 (151)	0.55
Grade	Presentation	2.2 (1.2)	2.1 (1.1)	0.63
	Follow-up	1.4 (1.5)	0.7 (0.9)	0.004*

Comparison based on ADHD status at each time point. Repeated Measures ANOVA p-values statistically significant (p<0.05)*. Reported as mean (SD), median (1st Q- 3rd Q), n (%)

The percentage of children requiring surgery was similar between groups with 55% of children with ADHD and 57% of children without ADHD. Of the patients initially offered non-operative treatments for their OCD lesion, 47% (8/17) of children with ADHD failed non-operative treatment compared with 33% (17/51) of children without ADHD, although this was not statistically significant (p=0.39).

Discussion

The incidence of OCD lesion⁹ and attention deficit hyperactivity disorder (ADHD)¹⁰ have both been on the rise. The results of this study suggest that ADHD may be more common in children with OCD lesions than other generalized knee conditions such as anterior knee pain. It also suggests that the comorbidity may lend to worse OCD outcomes. Although the exact etiology of OCD lesions still remains controversial,^9,11-16 the most accepted etiology of OCD lesions is acute or repetitive microtrauma resulting in a stress fracture of the subchondral bone⁹ as this has been shown to occur in athletes.¹⁷ While OCD lesions have not been linked to ADHD or its treatment, ADHD medications (stimulants), have been linked to mild growth supression.^7,8 As both non-operative^18,19 and operative^17,20-23 OCD treatment require children to remain non-weightbearing for extended periods of time (weeks to months); the diminished healing in those with ADHD may be directly due to the stimulants effect on blood supply and bone healing or indirectly through behavioral difficulty complying with treatment reccomendations.

We are not the first to make a connection between ADHD and orthopedic conditions. Previously ADHD was thought to be a risk factor in the development of Legg-Calve-Perthes (LCP) disease as 33% of children with LCP were found to have a diagnosis of ADHD, compared to the 5% incidence of ADHD in the general population at the time.⁵ A more recent study shows that perhaps the link is not with ADHD per se, but rather between a generalized behavioral disorder and LCP.⁶ Another study found that ADHD was present in 75% of children with a diagnosis of Osgood-Schlatter Disease (OSD).²⁴ Whereas still others have noticed increased fracture rates²⁵ and non-fatal injuries²⁶ in children with ADHD. Our current data was similar to that in these previous studies, as a three-fold increase in the diagnosis of ADHD in children with OCD lesions compared to the prevalence of ADHD among the general population (~8%)^10,27,28 and a two-fold difference between patients with OCD and anterior knee pain, (23% vs 11%; p < 0.05).

This study does have limitations. It was retrospective in nature and there was no formal protocol in determining treatment of the patients. Diagnosis of ADHD was made on chart review determined by the reported “problem” or “medication” list, so there was no formal confirmation of the psychiatric diagnosis. There were also no females in the ADHD group, but when the male subpopulation was analyzed (data not shown), similar results were obtained. A future prospective study with standardized treatment protocol for patients with OCD lesions and standardized criteria for the diagnosis of ADHD is warranted to confirm the results of this study.

In conclusion, we believe that there is a relationship between the diagnosis of ADHD and OCD lesions. From the current study we are unable to say whether this is due to the behavioral effects or treatments used to treat ADHD. When evaluating patients with OCD, we suggest obtaining a thorough history to determine if the patient has the diagnosis of ADHD as this may influence the outcome of treatment. Surgeons should be aware that children with OCD and ADHD might fare worse than their counterparts without ADHD regardless of treatment. Future prospective trials with controlled treatment protocols are warranted to better determine the relationship of ADHD in children with OCD lesions.