Outcomes of Temporary Hemiepiphyseal Stapling for Correcting Genu Valgum in Children with Multiple Osteochondromas: A Single Institution Study

Background: Multiple osteochondromas is a rare skeletal disorder characterized by the presence of osteocartilaginous protrusions causing bony deformities, especially around the knee. Guided growth by temporary hemiepiphyseal stapling is the treatment of choice to correct the deformity by modulating the residual physeal growth of the lower limbs. Although this procedure is increasingly practiced, inconclusive evidence exists regarding its effectiveness in children with multiple osteochondromas. The study aims to compare the outcomes of temporary hemiepiphyseal stapling for correcting genu valgum in children with multiple osteochondromas vs. idiopathic cases. Methods: In this retrospective cohort study, we included patients admitted at a single institution from 2008 to 2018. A total of 97 children (77 idiopathic, 20 multiple osteochondromas) were enclosed, accounting for 184 limbs treated by temporary hemiepiphyseal stapling. We investigated if children with multiple osteochondromas had a similar successful rate of correction, rate of complications, and correction velocity compared to children with idiopathic genu valgum. Results: Overall, 151 limbs (82%) achieved complete correction or overcorrection, with idiopathic cases having a significantly higher rate of success compared to pathologic cases (88% vs. 55%; p < 0.001). In addition, multiple osteochondromas children sustained a higher rate of major complications (p = 0.021) and showed significantly lower correction velocity (p = 0.029). Conclusion: Temporary hemiepiphyseal stapling is effective in both idiopathic and multiple osteochondromas children, although the latter often achieved incomplete correction, had a higher risk of complications, and required a longer time of stapling. We suggest to anticipate the timing of intervention; otherwise, children with multiple osteochondromas and severe valgus deformity, approaching skeletal maturity, could undergo combined femoral and tibial stapling.

Line from the top center of the femoral head to a line parallel to the distal medial and lateral condyles Tibial length (TL) Line from the center of the intercondylar eminence to the center of the tibial plafond (midpoint between the medial and lateral malleoli at distal tibia) Fibular length (FbL) Line from the apex to the lowest end of the fibula Mechanical axis of the lower extremity Line from the center of the femoral head to the center of the tibial plafond Absolute Mechanical axis deviation (MAD mm) The perpendicular distance from the mechanical axis of the lower extremity line to the center of the knee joint, expressed in mm 8 ± 7 mm medial. Genu valgus deformity was identified based on increased lateral deviation of the MAD Relative Mechanical axis deviation (MAD %) The perpendicular distance from the mechanical axis of the lower extremity line to the center of the knee joint, expressed as a percentage to one half of the width of the tibial plateau (to eliminate the bias created by the different tibial width in each patient)

Femoral mechanical axis
Line between the center of the femoral head and the midpoint between the medial and lateral condyles Femoral anatomic axis The mid-diaphyseal line of the femur Tibial mechanical axis Line from the center of the knee joint to the center of the ankle plafond The lateral angle formed between the lines of the mechanical femoral axis and the articular surface of the distal femur 88° ± 2° Anatomical lateral distal femoral angle (aLDFA) The medial angle formed between the line of the mechanical tibial axis and a line tangent to the joint surface of the proximal tibial plateau 81° ± 2° Mechanical medial proximal tibial angle (mMPTA) The medial angle formed between the line of the mechanical tibial axis and a line tangent to the joint surface of the proximal tibial plateau 87° ± 2° Anatomical medial proximal tibial angle (aMPTA) The medial angle formed between the line of the anatomical tibial axis and a line tangent to the joint surface of the proximal tibial plateau 87° ± 2° It normally corresponds to the mMPTA.  The child needed unplanned additional drugs, physiotherapy, or exams in case of pain, neurapraxia or thrombosis III A complication that is treatable but requires surgical, endoscopic, or interventional radiology procedure(s), or an unplanned hospital readmission A hospital re-admission was necessary, when surgical reinterventions were needed like reoccurrence of osteochondromas previously excised, hypo-or over-correction, plate malposition or breakage of hardware, infection IVa A complication that is life or limb-threatening, and/or requires ICU * admission, a complication with potential for permanent disability but treatable, a complication that may require organ/joint resection/replacement. No long-term disability

IVb
A complication that is life or limb-threatening, and/or requires ICU admission, a complication that is not treatable, a complication that requires organ/joint resection/replacement or salvage surgery. With long-term disability V Death Not observed in any case * ICU: Intensive Care Unit  Table S5. Adjustment comparisons between MO versus idiopathic cohort on post-operative MAD and correction velocity. Linear regression models.

Comparisons between MO versus idiopathic cohort using a Propensity Score approach
To adjust the comparisons between MO versus idiopathic cohort on HeS outcomes, we performed different analyses using a Propensity Score (PS) approach. A PS for the likelihood to have been diagnosed with MO was calculated using a logistic regression model from the following covariates: gender, Age at stapling, height zscore, weight z-score, affected side, site, bilaterality, and pre-operative MAD. Linear regression models were estimated including as a predictor the diagnostic group variable along with the PS. Besides, the PS was applied to create more balanced samples with a 1:1 ratio between MO and idiopathic children, using the nearest neighbor matching without replacement, with and without a caliper distance of 0.40 standard deviation of the PS distribution.
The analyses results, along with the estimates deriving from multivariable analyses, were reported in Table  S4.