This study was approved by our institutional review board and was registered in a prospective database (IRB#2016-08-010, KCT 0002643). Patients who had undergone hip arthroscopy between August 2017 and July 2019 and who fulfilled our inclusion criteria were enrolled in the study. The inclusion criteria were, as follows: 1) age between 18 and 50 years and a valid driving license, 2) history of routine commute by driving before symptom onset, 3) no influence of underlying diseases or consumption of drugs that may potential affect the driving performance, 4) history of hip arthroscopy on the right hip, and 5) provision of consent to undergo a simulator test at a weekly interval for 6 weeks. Of the 55 patients who were initially enrolled, 8 patients were excluded owing to 1) motion sickness during driving simulation (n=2) and 2) inability to attend the entire 6-week test session (n=6). The 47 remaining patients were classified into the following two groups based on the type of the surgery that they underwent: 1) femoroacetabular impingement (FAI) surgery group (n=29) and 2) simple hip arthroscopy (SA) group (n=18). Patients who underwent osteochondroplasty for a femoral head cam lesion and repair of the labrum with or without rim trimming were allocated to the FAI surgery group. Patients who underwent surgeries that did not require bone resection or labral repair and for whom the entire operation duration was <60 min were allocated to the SA group.
We enrolled 20 healthy volunteers who were routine commuting drivers aged between 18 and 50 years and classified them as the controls (control group). The participants in the control group underwent thorough physical examination to exclude any hip problems, including range of motion, pain on movement, point tenderness, impingement test, flexion-abduction-external rotation test, and straight leg elevation test. Only the subjects who had no abnormalities during physical examination were selected.
Prior to the test, all subjects were questioned regarding their driving experience, including the number of hours spent driving each week and number of years since they acquired a driving license. For the patients who had undergone hip arthroscopy, an additional question was asked to determine if they had ceased driving or the number of hours they spent driving had decreased after the initiation of hip pain.
All surgeries were performed by a single surgeon. The indications for surgery were as follows: 1) minimum of 3 months of conservative treatment, 2) improvement in pain with intra-articular injection, 3) detection of a pathologic lesion in radiographic images, and 4) pain severe enough to interfere with daily life. The surgeries were performed only when the diagnosis was confirmed preoperatively, which was based on the physical examination and the radiographic analysis. The patients were operated on in the lateral decubitus position, and the surgery was initiated using standard lateral and anterolateral portals. The 29 patients who underwent arthroscopy for FAI (FAI surgery group) underwent interportal capsulotomy to enhance instrument manipulation and osteochondroplasty of the femoral head with labral repair using one to three anchors. Fourteen of these patients underwent rim trimming for pincer lesions. Of the 18 patients who underwent simple arthroscopy, simple synovectomy was performed for 6 patients; labral debridement and capsular shrinkage, 5 patients; synovial chondromatosis removal, 4 patients; calcification debridement, 2 patients; and ligamentum teres debridement, and shrinkage, 1 patient. Interportal capsulotomy was performed in 15 patients from the SA group. Capsule repair was performed only when the surgeon decided that capsulotomy was excessive, which was conducted in 21 patients in the FAI surgery group and 10 patients in the SA group.
All patients were allowed to ambulate with the assistance of crutches on the first postoperative day. Range-of-motion exercises were initiated as soon as the pain was tolerable. While the patients who had undergone simple arthroscopy were not restricted from any postoperative movement, the patients who had undergone labral repair were discouraged from squatting deeply until the 6th week after the procedure. For pain management, intravenous narcotics were introduced immediately following the surgery, and additional tramadol injections were prescribed according to the patients’ need. Oral pain control pills were prescribed for a minimum of 2 weeks and were re-prescribed when the patients felt it was necessary. The institution typically allows 1-week admission for patients undergoing hip arthroscopy.
Simulator set-up and test protocol
A modern immersive driving simulator (Carnetsoft BV, Groningen, Netherlands), developed for driver training and research, was used in the current study. This driving simulator has shown high validity, excellent test-retest reliability, and significant sensitivity for testing fitness to drive [11, 12]. The simulator was composed of three monitors, a steering wheel, and a pedal unit. As the simulator was configured with an automatic transmission model, a clutch or a shifting gear was not used. Three 24-inch monitors were used for display, so that a 180-degree field of view could be provided, and an additional user interface monitor was used to control the driving scenario. The simulator provided a dashboard that included a speedometer and a tachometer. A stereo audio system, which included engine sounds and natural road traffic noises, was also provided. In addition, an adjustable driving seat was manufactured for comfortable access to the pedal unit, so that the driving simulator could imitate real driving as much as possible (Figure 1). The test scenario, i.e., driving in a three-dimensional realistic suburban road, was developed specifically for the current study.
The test included an initial 5 min of practice driving in a suburban environment followed by a 5-min test driving session. A stopping event was described using the flashing red stop sign on the screen, which was triggered by the investigator using the separate user interface monitor. The stopping event was initiated only when the driving speed exceeded 60 km/h. Five stopping events were tested during the course of test driving.
Overall, all patients participated in eight sessions of simulated driving during the course of the study. The index driving test was performed prior to the surgery to set the baseline. Prior to the test, the patients underwent detailed instruction sessions on what to expect during the simulation. The first postoperative driving test was performed when the patients felt comfortable on sitting on the driving seat and when they felt that they are ready to attempt simulated driving. The second postoperative driving test was performed on the 7th day following the surgery, and the tests were repeated at weekly intervals for 6 weeks. On the day of the simulated driving test, all patients refrained from taking opioid medication, if they had been prescribed any. The 20 healthy volunteers (control group) also underwent the same protocol, where they practiced driving for 5 min followed by 5 min of test driving. This was repeated thrice at weekly intervals to determine if acclimatization to the driving simulator may have had any potential effects on the results of the simulation.
For the outcome, the brake reaction time (BRT), total brake time (TBT), and brake pedal depression (BPD) for each of the five stopping events were measured, and the means of the five results for each parameter were used for the analysis. The BRT was defined as the time period from the first flash of the red stop sign on the screen to the patients setting their foot on the brake pedal. The TBT was defined as the time period between the stopping event and the car stopping completely. The BPD was measured as a percentage of the brake pedal pushed by the participants with respect to the pedal being fully pushed .
For the patient-reported outcomes, the visual analog scale (VAS) and international hip outcome tool (iHOT-12) scores were measured for all subjects prior to the test. Additionally, the VAS score was measured before each driving test being performed. All patients were recommended to not drive for 6 weeks following the surgery.
Statistical analysis was performed using the SPSS 21 software (SPSS Inc., USA). Continuous variables were expressed as means and standard deviations (SDs). The independent t-test was performed to compare the preoperative driving performance of the patients with that of the controls, while analysis of variance (ANOVA) was used to compare the variables among the SA, FAI surgery, and control groups. The paired t-test was used to analyze the time required to regain the preoperative driving performance level after hip arthroscopy, and repeated-measures ANOVA was performed to assess whether a learning effect was noted in the control group. The sphericity of the repeated ANOVAs was tested using the Mauchly test, and Greenhouse-Geisser correction was used when the sphericity was violated. Linear regression analysis was performed to assess the correlation between the patient-reported outcomes (VAS score) and braking parameters. The significance level was set to an alpha value of 0.05 for all analyses. Sample size estimation in priori with 80% power recommended 20 subjects per group to detect a difference of 150 ms in the BRT.