AAP National Conference & Exhibition 2022
Section on Orthopaedics-POSNA Young Investigator Awards


Epidemiology of Preschool-Age Children Presenting to the Orthopaedic Clinic with Limping

Megan Hannon, MD; Blair Stewig, BS; Lois Lee, MPH, MD

Boston Children’s Hospital, Boston, MA

DOI: 10.55275/JPOSNA-2023-688

Volume 5, Number 2, May 2023

jposna2023688_fx1.jpg

Hip/Lower Extremity

Recipient: Blair Stewig, BS

Abstract

Background: Limping in preschool-age children is a common chief complaint in the orthopaedic clinic. Given the breadth of limping etiologies, there is variability in the evaluation of these patients. The objectives of our study are: 1) to describe the epidemiology of preschool-age children presenting with limp and 2) to determine predictors identifying children at risk for non-benign etiologies.

Methods: This is a retrospective study of children <6 years old presenting to a tertiary care pediatric orthopaedic clinic with a chief complaint of limping from 1/1/2019 to 12/31/2020. Subjects were identified from the electronic medical record using natural language processing. Exclusion criteria included: limp >3 weeks, lower extremity fracture or surgery <12 weeks, or chronic conditions affecting the lower extremity (i.e., malignancy, neuromuscular disorder, sickle cell disease). Patient demographic and clinical characteristics were collected. Outcome diagnoses were classified as “benign,” “receiving advanced treatment (i.e., antibiotics, steroids, surgery),” or “inconclusive.” We calculated frequencies of outcomes and compared them using the Chi-square test. Bivariate analyses were conducted using Student’s t-test, the Mann-Whitney U-test, or chi-square test. We conducted a multivariable logistic regression analysis to determine predictors associated with the outcome of receiving advanced treatment vs. benign/inconclusive diagnoses while controlling for patient characteristics.

Results: We identified 169 children <6 years old with limping: 105 benign causes (62%), 27 receiving advanced treatment (16%), and 37 inconclusive diagnoses (22%). The most common diagnoses among those receiving advanced treatment were juvenile idiopathic arthritis (8/27, 30%) and Lyme disease (7/27, 26%) (Table 1). On bivariate analysis, increased odds for requiring advanced treatment were identified for females (OR 2.8, 95% CI 1.20, 6.44) and presence of joint edema (OR 5.3, 95% CI 2.11, 13.47) (Table 2). Multivariate logistic regression found increased odds of receiving advanced treatment for females (OR 3.05, 95% CI 1.13, 8.24), joint edema (4.49, 95% CI 1.48, 13.61), and limp >9 days (OR 3.43, 95% CI 1.24, 9.51).

Table 1. Patient and Diagnosis Characteristics

All subjects (N=169) Advanced treatment (n=27) Benign/inconclusive (n=142)
Characteristic Freq. (%) Freq. (%) Freq. (%) P
Age at presentation (years; mean (SD)) 3.3 (1.44) 3.5 (1.63) 3.3 (1.40) 0.57
Sex (% male) 110 (65%) 12 (44%) 98 (69%) 0.03
Diagnosis
 Juvenile idiopathic arthritis 8 (5%) 8 (30%) 0 (0%)
 Lyme 7 (4%) 7 (26%) 0 (0%)
 Other 6 (4%) 3 (11%) 3 (2%)
 Legg-Calves-Perthes disease 3 (2%) 3 (11%) 0 (0%)
 Reactive arthritis 2 (1%) 2 (7%) 0 (0%)
 Developmental dysplasia of the hip 1 (1%) 1 (4%) 0 (0%)
 Septic arthritis 1 (1%) 1 (4%) 0 (0%)
 Stress Fracture 1 (1%) 1 (4%) 0 (0%)
 Kohler disease 1 (1%) 1 (4%) 0 (0%)
 Transient synovitis 64 (38%) 0 (0%) 64 (45%)
 Toddler fracture 19 (11%) 0 (0%) 19 (13%)
 Soft tissue injury or contusion 9 (5%) 0 (0%) 9 (6%)
 Toddler fracture (foot) 6 (4%) 0 (0%) 6 (4%)
 Reactive synovitis 3 (2%) 0 (0%) 3 (2%)
 No conclusive diagnosis 38 (23%) 0 (0%) 38 (27%)

Conclusion: The majority of preschool-age children presenting with limp had benign diagnoses. Understanding predictors for specific etiologies of limping “receiving advanced treatment” may assist in standardizing care for preschool children with limping.

Reprinted with Permission from AAP. The Young Investigator Awards (YIA) recognize the best abstract presentations by residents, fellows, and students at the annual scientific session during the National Conference & Exhibition.

Table 2. Presentation and Clinical Characteristics by Diagnosis Category

All subjects (N=169) Advance treatment (n=27) Benign/ inconclusive (n=142)
Characteristic Freq. (%) Freq. (%) Freq. (%) P
Laterality 0.39
 Left 71 (42%) 14 (52%) 57 (40%)
 Right 89 (53%) 11 (41%) 78 (55%)
 Unclear 9 (5%) 2 (7%) 7 (5%)
Presenting location 0.26
 Orthopaedic clinic 113 (67%) 15 (56%) 98 (69%)
 Urgent clinic 56 (33%) 12 (44%) 44 (31%)
Previous treatment 143 (85%) 23 (85%) 120 (85%) 0.50
 PCP 117 (81%) 19 (83%) 98 (82%) >0.99
 ED 31 (21%) 4 (17%) 27 (23%) 0.79
Duration of limp (days; median (IQR)) 5 (2-10) 6 (2-13) 5 (2-10) 0.39
Pain 90 (53%) 18 (67%) 72 (51%) 0.20
Morning stiffness 12 (7%) 4 (15%) 8 (6%) 0.25
Tick bite 3 (2%) 1 (4%) 2 (1%) 0.98
Previous infection 56 (33%) 11 (41%) 45 (32%) 0.49
Upper respiratory infection 34 (61%) 4 (36%) 30 (67%) 0.13
Strep 4 (7%) 0 (0%) 4 (9%) 0.71
Flu 4 (7%) 2 (18%) 2 (4%) 0.35
viral infection 7 (13%) 2 (18%) 5 (11%) 0.90
Ear infection 9 (16%) 2 (18%) 7 (16%) >0.99
Hand foot, and mouth disease 2 (4%) 1 (9%) 1 (2%) 0.85
Minor trauma 28 (17%) 4 (15%) 24 (17%) 0.99
Fall 17 (57%) 3 (75%) 14 (58%) 0.94
Blunt injury 1 (3%) 1 (25%) 0 (0%) 0.30
Sport/playground 9 (30%) 0 (0%) 9 (38%) 0.36
Hx of refusal to bear weight 111 (66%) 16 (59%) 95 (67%) 0.59
Weight-bearing 0.50
 No 12 (7%) 3 (11%) 9 (6%)
 Yes, fully 56 (33%) 6 (22%) 50 (35%)
 Yes, with limp 96 (57%) 18 (67%) 78 (55%)
 Yes, partially 3 (2%) 0 (0%) 3 (2%)
 Unknown 2 (1%) 0 (0%) 2 (1%)
Swelling/edema 27 (16%) 11 (41%) 16 (11%) <0.001
Skin sign 11 (7%) 2 (7%) 9 (6%) >0.99
Bruising 3 (27%) 1 (50%) 2 (22%) >0.99
Rash 3 (27%) 0 (0%) 3 (33%) 0.94
Abrasions 2 (18%) 0 (0%) 2 (22%) >0.99
Redness 1 (9%) 1 (50%) 0 (0%) 0.39
Warm to touch 1 (9%) 0 (0%) 1 (11%) >0.99
Other 1 (9%) 0 (0%) 1 (11%) >0.99
Localized tenderness 33 (20%) 6 (22%) 27 (19%) 0.74
Time from limp onset to presentation (days; median (IQR)) 5 (2-10) 6 (2-13) 5 (2-10) 0.42
Time from presentation to diagnosis (days; median (IQR)) 0 (0-4) 4 (0-26) 0 (0-0) <0.001
Time from presentation or follow-up (days; median (IQR); N=88)* 28 (12-197) 306 (31-655) 22 (8-58) 0.003

*The number in parentheses represents the number of patients with available data for the given characteristic.

IQR, interquartile range; SD, standard deviation.