Ki Hyuk Sung

Pitfalls and Important Issues in Testing Reliability Using Intraclass Correlation Coefficients in Orthopaedic Research

Background Intra-class correlation coefficients (ICCs) provide a statistical means of testing the reliability. However, their interpretation is not well documented in the orthopedic field. The purpose of this study was to investigate the use of ICCs in the orthopedic literature and to demonstrate pitfalls regarding their use. Methods First, orthopedic articles that used ICCs were retrieved from the Pubmed database, and journal demography, ICC models and concurrent statistics used were evaluated. Second, reliability test was performed on three common physical examinations in cerebral palsy, namely, the Thomas test, the Staheli test, and popliteal angle measurement. Thirty patients were assessed by three orthopedic surgeons to explore the statistical methods testing reliability. Third, the factors affecting the ICC values were examined by simulating the data sets based on the physical examination data where the ranges, slopes, and interobserver variability were modified. Results Of the 92 orthopedic articles identified, 58 articles (63%) did not clarify the ICC model used, and only 5 articles (5%) described all models, types, and measures. In reliability testing, although the popliteal angle showed a larger mean absolute difference than the Thomas test and the Staheli test, the ICC of popliteal angle was higher, which was believed to be contrary to the context of measurement. In addition, the ICC values were affected by the model, type, and measures used. In simulated data sets, the ICC showed higher values when the range of data sets were larger, the slopes of the data sets were parallel, and the interobserver variability was smaller. Conclusions Care should be taken when interpreting the absolute ICC values, i.e., a higher ICC does not necessarily mean less variability because the ICC values can also be affected by various factors. The authors recommend that researchers clarify ICC models used and ICC values are interpreted in the context of measurement.

Long term outcome of single event multilevel surgery in spastic diplegia with flexed knee gait.

Distal hamstring lengthening (DHL) is a commonly performed procedure in flexed knee gait. However, the necessity of this procedure has been challenged due to the concerns on adverse effects in long-term follow-up. This retrospective study was undertaken to investigate the long-term outcome of single event multilevel surgery (SEMLS), including bilateral DHL, in ambulatory patients with cerebral palsy using 3D gait analysis. Twenty-nine ambulatory patients with spastic diplegic cerebral palsy who had undergone SEMLS including bilateral DHL were included. 3D gait analysis was performed preoperatively, 1 year postoperatively and over 10 years postoperatively. Preoperative temporal parameters, kinematics and GDI were compared with values obtained 1 and 10 year follow-up visits. The mean age of patients at time of first surgery was 8.3 years (range, 5.4-16.3 years), and mean time from first surgery to last 3D gait analysis was 11.8 years (range, 10.0-13.3 years). Mean pelvic tilt was not changed significantly after SEMLS including DHL. Mean knee flexion at initial contact decreased from 31.1° preoperatively to 26.0° at 1 year postoperatively (p=0.065), and then decreased significantly to 23.6° at 10 years postoperatively (p=0.038) versus the preoperative value. Mean GDI score significantly improved from 69.4 preoperatively to 77.9 at 1 year postoperatively (p=0.003) and continuously improved to 82.2 at 10 years postoperatively (p=0.017). Single event multilevel surgery including DHL provides a favorable outcome 10 years postoperatively in patients with spastic diplegic cerebral palsy.

Arthroplasty versus arthrodesis for end-stage ankle arthritis: decision analysis using Markov model

BackgroundTotal ankle arthroplasty and arthrodesis are the two mainstreams of treatment for end-stage ankle arthritis. This study was performed to determine which is a better choice for ankle arthritis, using a decision analysis and Markov model to reflect the repetitive nature of revision arthroplasty.MethodsBased on current published evidence, a decision tree was constructed to compare the clinical outcomes of total ankle arthroplasty and arthrodesis, which contained the possible clinical events and the probabilities. Total ankle arthroplasty was subject to revision arthroplasty, and a Markov model was adopted for this branch to reflect this repetitive trait of the procedure. Arthrodesis could cause adjacent arthritis, and a conventional decision analysis model was adopted for this branch. Quality well-being index score was used for clinical outcome assessment, which was the utility in the decision tree. Sensitivity analysis was performed to test the stability of the decision tree and the threshold values.ResultsThe model favoured total ankle arthroplasty over arthrodesis in terms of quality well-being index score. Sensitivity analysis showed that the model was considerably stable, unaffected by the changes in probabilities of failure after total ankle arthroplasty and adjacent arthritis after arthrodesis.ConclusionsBased on current evidence, total ankle arthroplasty was found to be a better treatment than arthrodesis for ankle arthritis. Future development in the implant materials, improved understanding of ankle biomechanics, and surgical techniques will further enhance the clinical outcome of total ankle arthroplasty.

Consensus and Different Perspectives on Treatment of Supracondylar Fractures of the Humerus in Children

Background Although closed reduction and percutaneous pinning is accepted as the treatment of choice for displaced supracondylar fracture of the humerus, there are some debates on the pinning techniques, period of immobilization, elbow range of motion (ROM) exercise, and perceptions on the restoration of elbow ROM. This study was to investigate the consensus and different perspectives on the treatment of supracondylar fractures of the humerus in children. Methods A questionnaire was designed for this study, which included the choice of pinning technique, methods of elbow motion, and perception on the restoration of elbow ROM. Seventy-six orthopedic surgeons agreed to participate in the study and survey was performed by a direct interview manner in the annual meetings of Korean Pediatric Orthopedic Association and Korean Society for Surgery of the Hand. There were 17 pediatric orthopedic surgeons, 48 hand surgeons, and 11 general orthopedic surgeons. Results Ninety-six percent of the orthopedic surgeons agreed that closed reduction and percutaneous pinning was the treatment of choice for the displaced supracondylar fracture of the humerus in children. They showed significant difference in the choice of pin entry (lateral vs. crossed pinning, p = 0.017) between the three groups of orthopedic surgeons, but no significant difference was found in the number of pins, all favoring 2 pins over 3 pins. Most of the orthopedic surgeons used a removable splint during the ROM exercise period. Hand surgeons and general orthopedic surgeons tended to be more concerned about elbow stiffness after supracondylar fracture than pediatric orthopedic surgeons, and favored gentle passive ROM exercise as elbow motion. Pediatric orthopedic surgeons most frequently adopted active ROM exercise as the elbow motion method. Pediatric orthopedic surgeons and general orthopedic surgeons acknowledged that the patients age was the most contributing factor to the restoration of elbow motion, whereas hand surgeons acknowledged the amount of injury to be the most contributing factor. Conclusions More investigation and communication will be needed to reach a consensus in treating pediatric supracondylar fractures of the humerus between the different subspecialties of orthopedic surgeons, which can minimize malpractice and avoid medicolegal issues.

Hamstring and psoas length of crouch gait in cerebral palsy: a comparison with induced crouch gait in age- and sex-matched controls

BackgroundPrevious studies have shown that hamstring lengths are often not short in patients with cerebral palsy, which raises concerns over the benefits of distal hamstring lengthening in patients with crouch gait. In this study, the authors measured lengths of hamstrings and psoas muscles in normal subjects mimicking crouch gait and compared these with lengths in cerebral palsy patients with crouch gait.MethodsThirty-six patients with cerebral palsy and crouch gait were included in this study, and in addition, 36 age- and sex-matched normal controls were recruited. Hamstring and psoas muscle lengths in patients were evaluated using gait analysis and interactive musculoskeletal modeling software. Muscle lengths were also measured in the normal control group during normal gait and while mimicking crouch gait, and these were compared with those of cerebral palsy patient with crouch gait.ResultsNo significant differences were observed between maximum hamstring (p=0.810) and maximum psoas (p=0.456) lengths of patients and controls mimicking crouch gait. However, patients showed significantly shorter excursions of hamstring (p=0.022) and psoas (p=0.036) muscles than controls, whereas no significant excursion differences were observed between controls during normal gait and mimicking crouch gait.ConclusionsNormal controls mimicking crouch gait and cerebral palsy patients with crouch gait demonstrate similar muscle length patterns. However, mimicked crouch gait did not reproduce the excursion pattern shown by patients with crouch gait, which suggests that reduced hamstring and psoas excursion is an innate characteristic of pathologic crouch gait.

Incidence of Deep Vein Thrombosis after Major Lower Limb Orthopedic Surgery: Analysis of a Nationwide Claim Registry

Purpose We aimed to evaluate the nationwide incidence and risk factors for symptomatic deep vein thrombosis (DVT) after major lower limb orthopedic surgeries. Materials and Methods The Korean Health Insurance Review and Assessment Service database was used to retrospectively identify International Classification of Disease-10 codes for DVT and operation codes representing hip arthroplasty, knee arthroplasty, and hip fracture surgeries. The age- and gender-adjusted annual incidence of DVT, rates of major lower limb orthopedic surgeries, and the postoperative incidence of DVT according to the surgical procedure were assessed. Results The age- and gender-adjusted annual incidence of DVT was 70.67 per 100000 persons/year. Compared to patients aged <49 years, the relative risk of DVT was five times higher in patients aged 50-69 and 10 times higher in patients aged >70 years (p<0.001). Females showed a greater relative risk for DVT than males (1.08; p<0.001). The incidence of postoperative DVT, according to the type of surgery, was significantly greater for knee replacement arthroplasty than for other forms of surgery (p<0.002). The relative risk of postoperative DVT was higher in females in knee replacement arthroplasty (1.47) and hip fracture surgery (2.25) groups, although relatively lower in those who underwent hip replacement arthroplasty (0.97). Conclusion Among major lower limb surgeries, advanced age, female gender, and undergoing a knee replacement arthroplasty were found to be risk factors for developing postoperative DVT. These findings further emphasize the need for orthopedic surgeons to consider the development of DVT after surgery in high-risk patients.

Prophylactic femoral varization osteotomy for contralateral stable hips in non‐ambulant individuals with cerebral palsy undergoing hip surgery: decision analysis

Aim  This study was undertaken to determine the need for concurrent prophylactic femoral varization osteotomy (FVO) of contralateral stable hips at the time of hip reconstructive surgery on unstable hips in non‐ambulant individuals with cerebral palsy (Gross Motor Function Classification System levels IV and V).

Risk factors associated with amputation-free survival in patient with diabetic foot ulcers.

Purpose To determine the 1-year survival rate, 1-year amputation-free survival rate and the risk factors of amputation for patients with diabetic foot ulcers. Materials and Methods One hundred seventy-three patients with diabetic foot ulcers were included in our study. Mean patient age was 67.5 (range, 29 to 87, SD ±11.4) years. 74% of the patients were male. Time from study entry to amputation and time to death were evaluated separately as censored event times by Kaplan-Meier curves and log-rank tests. A multivariate Cox proportional hazards regression analysis was carried out for determining the risk factors of amputation. Results The survival rate and amputation-free survival rate were 96.5% (n=167), 65.9% (n=114), respectively, over one year study period. Severity of ulcer was the strongest significant risk factor of amputation [hazard ratio (HR): 7.99; confidence interval (CI): 3.12 to 20.47]. Peripheral artery disease was also independent risk factor of amputation (HR: 2.64; CI: 1.52 to 4.59). Conclusion In assessing the prognosis of diabetic foot ulcers, clinicians should consider the severity of ulcer and presence of peripheral artery disease. Our study provides important insights into clinical practice and supplementary information for both physicians and patients.

Transcultural adaptation and testing psychometric properties of the Korean version of the Foot and Ankle Outcome Score (FAOS)

This study was performed to translate and transculturally adapt the English version of the Foot and Ankle Outcome Score (FAOS) into a Korean version, and to test psychometric properties of the Korean FAOS in terms of internal consistency, test–retest reliability, convergent validity, and dimensionality. Translation and transcultural adaptation of FAOS into a Korean version was performed according to internationally recommended guidelines. Internal consistency (N = 294) and test–retest reliability (N = 21) were evaluated. Convergent validity was analyzed using correlation with pain visual analogue scale (VAS) score. All subscales, except for the quality of life (Q) subscale (Cronbach’s alpha, 0.615), showed satisfactory internal consistency (Cronbach’s alpha > 0.7). Cronbach’s alpha of function in daily living (ADL) was highest (0.962), which might represent the redundancy of the items. All five subscales showed satisfactory reliability with ADL subscale showing the highest ICC (intraclass correlation coefficient; 0.851) and Q subscale the lowest ICC (0.718). Pain VAS score showed the highest correlation with pain (P) subscale of FAOS (r = 0.675, p < 0.001) and the lowest correlation with Q subscale (r = 0.495, p < 0.001). In the dimensionality test, a factor analysis was performed using the total items to rank their relative significance, which showed seven components solution. Considerable portion of the items showed a similar dimension according to their original subscales, except for ADL items. Translation and transcultural adaptation of FAOS into the Korean language was performed successfully. The items were understandable, and the subscales showed satisfactory test–retest reliability. Some minor revision might be needed to enhance the internal consistency of Q subscale and reduce the redundancy of ADL subscale.

Medial and lateral crossed pinning versus lateral pinning for supracondylar fractures of the humerus in children: decision analysis.

Background: The choice of pinning techniques in supracondylar fractures of the humerus in children has been a debate regarding its fixation stability and risk of iatrogenic ulnar nerve palsy. This study was performed to determine as to which fixating method (medial and lateral crossed pinning vs. lateral pinning) is better for the displaced supracondylar fractures using a decision analysis tool in terms of function. Methods: A decision analysis model was designed containing the probability of iatrogenic ulnar nerve palsy and malunion caused by unstable fixation for each of lateral pinning and medial and lateral crossed pinning techniques. The final outcome was function adjusted life year and used as a utility in the decision tree, where function was evaluated using the McBride disability evaluation. The probabilities of all cases were obtained by literature review and assumptions. A roll back tool was used to determine the better pinning technique, and sensitivity analysis was performed to compensate for the uncertainty of the model. Results: Overall, our decision model favored the lateral pinning technique over the medial and lateral crossed pinning with the utilities of 99.6 and 99.3 in terms of function adjusted life year. One-way sensitivity analysis showed that the threshold rate of iatrogenic ulnar nerve injury as a complication after medial and lateral crossed pinning was 0.7%, below which the model favored medial and lateral crossed pinning over lateral pinning. The decision model was found to be sensitive to the percentage of permanent ulnar nerve palsy after medial and lateral crossed pinning. Two-way sensitivity analysis showed that the lateral pinning technique was more beneficial than the medial and lateral crossed pinning technique. Conclusions: In our decision analysis model, the lateral pinning technique was found to be more beneficial than the medial and lateral crossed pinning technique for supracondylar fractures of the humerus in children, on the basis of current evidences. However, the results were sensitive to the data of ulnar nerve injury. Avoiding the worst clinical scenario (permanent ulnar nerve palsy) might be more important and affordable than obtaining favorable clinical results (stable fixation) at the potential cost of disastrous complications. Level of Evidence: Level III.