A prospective study was conducted among the patients with hip osteoarthritis and femoral neck fractures who underwent primary cementless THA via posterolateral approach by one experienced surgeon between July 2017 and June 2019. Exclusion criteria were patients with previous ipsilateral tibial fractures, total knee arthroplasties and knee deformity with a tibio-femoral angle more than 5° varus or 15° valgus. The trial was approved by the institutional review board (Code 42/60) and registered in the Thai Clinical Trials Registry. All patients gave their written informed consent prior to inclusion.
The surgery was performed by one experienced surgeon via posterolateral approach. The patient was positioned in the lateral decubitus. The leg was put in a stockinette and an EKG electrode (3M Red Dot, USA) was attached to the medial 1/3 of the tibial tubercle. Two plastic pipe clips (Thai Pipe, Thailand) were attached to the anterior part of the shin with nylon cable ties. The base of one clip was positioned at the midpoint of the most medial and most lateral points of the malleoli and the other was locked onto the Red Dot electrode. An aluminium pipe (Yunteng self picture monopod YT-188, China) was gently pressed over both clips until the bilateral grooves of the pipe were snugly captured between the clip edges. This pipe would represent the mechanical axis of the tibia. A spirit level (Haccury YK-3, China) was glued to the base of another pipe clip, and then connected to the pipe by pressing the clip over the pipe (Fig. 1).
When the trial stem was inserted, the femur was internally rotated with knee flexion. Three methods for estimating the stem anteversion were performed sequentially. Method A by visual estimation, the surgeon assessed the angle between the leg axis and the axis of metal rod of the stem inserter handle by eye (Fig. 2A). Method B by digital protractor alone, the assistant placed the leg vertically until the surgeon approved its position by visualization, without any concern to the spirit level. A digital protractor (Etopoo DC18, China) was placed on the flat surface of the stem handle and recorded as an estimated stem anteversion (Fig. 2B). Method C by digital protractor and spirit level, the assistant internally rotated the leg until a bubble in the spirit level was centered. The stem anteversion was then measured by placing a digital protractor on the handle (Fig. 2C).
Demographic data included patient age, gender, body mass index (BMI), diagnosis and stem type. The intraoperative estimation of stem anteversion angles by method A, B and C were recorded. All patients received CT scans postoperatively at 4–5 days, in supine position. The scans were obtained from the acetabulum to the proximal tibia with a 1.5-mm thickness using Philips Ingenuity Core 128 (Cleveland, USA). True stem version was defined as the angle between a line through the center of the neck of the femoral prosthesis and the posterior condylar line . The knee alignment was measured as tibio-femoral angle in the scout view. Two intramedullary midpoints were marked at a 10-cm distance from the knee joint surfaces, one at the distal shaft of the femur and the other at the proximal shaft of the tibia. The angle between the lines drawn from the center of the bases of the tibial spines to both midpoints was defined as a tibio-femoral angle .
All radiographic assessments were independently performed by 2 orthopaedic residents, who were not involved with the surgery and repeated in a blind manner 4 weeks later. The average of 4 measurements was used for data analysis. Inter-observer and intra-observer measurement reliabilities were determined with intra-class correlation coefficients (ICC) using the absolute agreement and 2-way random-effects model. ICC values < 0.5 indicated poor reliability, 0.5–0.74 moderate, 0.75–0.9 good, and > 0.90 indicated excellent reliability .
The sample size was calculated to detect a significant difference in percentages of hips with intraoperative estimation error within 5°. According to the results in 25 hips of our pilot study using the traditionally visual estimation technique, 52% (13 hips) had an estimation error within 5°. We hypothesized that our method could achieve this goal in 90% of hips. With a two-sided type I error level of 0.05 and a 90% statistical power of detection in a two-dependent proportions formula, the sample size was 100 hips.
The primary outcome was the percentage of stem placements with an error within 5°. The secondary outcome was the deviation degree of the estimated stem anteversion from the true stem anteversion. The Shapiro-Wilk test for normal distribution was used prior to further statistical analysis. Continuous data were analyzed by using the t-test and Mann-Whitney U test. Categorized data were analyzed by using the exact probability test. Correlation between the estimated and true anteversion was analyzed by the Pearson correlation coefficient. Angle overestimation and underestimation were defined when the estimated anteversion was above and below the CT measurement by more than 1° respectively. We evaluated the factors that influenced the angle overestimation and underestimation in method C by using multivariate regression analysis. The statistical analyses were performed using STATA version 10.1 (Stata Corp LP, College Station, Texas, USA) and a p-value of < 0.05 was considered significant.