İntroduction The upper limb can be involved in children with cerebral palsy (CP), while the lower limb is more commonly affected. Little is known regarding the alterations in the upper limb kinematics after the orthopedic surgery of the lower extremity during gait. This study aimed to evaluate the differences in the upper limb kinematics in children with CP between preoperative and postoperative parameters.

Metods The study group included 30 children with CP (17 diplegics, 13 hemiplegics; mean age = 12,9 years, range =5-16 years; 18 boys, 12 girls) and the control group consisted of 29 healthy children (15 boys, 14 girls; mean age = 10,2 years, range=5-15 years). Video-based gait analyses of all children were retrospectively evaluated. The follow-up period for the 30 cerebral palsy children included in the study, based on the date of the latest walking video after surgery, was an average of 3.2 ± 1.5 years. The included patients were those who had undergone surgery on the lower extremity (Achilles tendon lengthening, varus derotation osteotomy, periacetabular osteotomy, gastrocnemius release, hamstring release, midfoot osteotomies) without any surgical intervention on the upper extremity and without contracture in the upper extremity. On April 28, 2023, the ethical committee decision 2023/801 was obtained from the Istanbul University Istanbul Medical Faculty Department of Orthopedics and Traumatology.

The inclusion criteria were as follows: diplegic or hemiplegic patients with surgical treatment of the lower extremity, no previous upper extremity surgery, and no contracture in the upper extremity. The exclusion criteria were: incomplete preoperative or postoperative data in the gait analysis records, walking with support or assistance during the gait analysis, and poor video quality that would not allow for measurement. The patients' preoperative and postoperative videos from the gait analysis laboratory and their medical records were retrospectively reviewed. Among the 300 patients who met the inclusion criteria from the 12,567 gait analysis data points, 76 had missing preoperative videos, 87 had missing postoperative videos, 55 were walking with parental assistance, and 52 had video quality that did not allow for measurement.

Coronal and sagittal view videos from the gait analysis were used for evaluation.The preoperative and postoperative videos of the patients were opened on a computer using the AVI or MP4 format and imported into the Kinovea program. The program allowed for slow motion, fast-forward, pause, and angular analysis of the movements in the videos. Using the program, the videos were paused at the initial contact (1), loading response (2), midstance (3), terminal stance (4), pre-swing (5), early swing (6), mid-swing (7), and terminal swing (8) phases in the sagittal plane, and the following angles were measured: the angle between the forearm axis and the third metacarpal axis (wrist sagittal plane angle), the angle between the humerus axis and the forearm axis (elbow sagittal plane angle), the angle between the humerus axis and the vertical line drawn on the ground (shoulder sagittal plane angle), the angle between the line drawn on the sagittal plane of the trunk and the vertical line on the ground (trunk sagittal plane angle), and the angle between the line drawn on the sagittal plane of the head and the vertical line on the ground (head sagittal plane angle). In the coronal plane, the videos were paused at initial contact and midstance, and the following angles were measured: the angle between the humerus axis and the vertical line drawn on the ground (shoulder coronal plane angle), the angle between the line drawn on the axis of the trunk and the vertical line on the ground (trunk coronal plane angle), and the angle between the line drawn on the coronal plane of the head and the vertical line on the ground (head coronal plane angle). In hemiplegic patients, the affected side was considered the patient side, and the unaffected side was considered the healthy side. In the diplegic group, the more affected side was considered the patient side, and the other side was considered the healthy side. If both sides were equally affected, the assignment was randomized between patient and healthy. In the control group, the right upper extremity was assigned as patient side, and the left upper extremity was assigned as healthy.To evaluate the efficacy of the surgery, the Edinburgh score was calculated in patients pre-and postoperatively.

Inclusion Criteria:

• children with hemiplegic and diplegic cerebral palsy
• Presence of preoperative and postoperative walking records from the lower extremity in our archive

Exclusion Criteria:

• upper extremity surgery
• Video recording does not allow measurement

children with hemiplegic and diplegic cerebral palsy and 5-15 age health children