The results of the current study indicate that driving performance, as measured by the patient’s response to brake with the driving simulator, shows no significant difference between patients who have painful hip indicated for hip arthroscopy and the asymptomatic controls. Additionally, our results indicate that the patient brake response recovers to the preoperative state within 2 weeks after hip arthroscopy. The pain did not correlate with the time the patient took to return to the preoperative driving state.
The time required to return to driving following the orthopaedic intervention is of great concern to patients, but there are few studies dealing with this issue [1, 9]. Endangering patients by testing their driving performance in real-life situations is highly unethical; therefore, most of the available literature on testing postoperative driving fitness utilises driving simulators. While various parameters have been used for this assessment, the ability to brake is recognised as the most important ability for safe driving[1, 9]. Currently, there is no established threshold for guaranteed safe driving, but BRTs between 750 and 1500 ms have been suggested by various institutions [10, 11]. Using simulator data, studies have reported that patients return to average driving performance approximately 2–6 weeks after total knee arthroplasty [6, 12–14] and 4 to 8 weeks after total hip arthroplasty [15–17].
In contrast to arthroplasty surgeries, arthroscopy involves less injury to the periarticular muscle structure, resulting in less postoperative change to strength and function. It can therefore be hypothesised that driving performance could be recovering earlier with arthroscopy surgery [18]. The studies have shown that the brake reaction function returns to baseline at approximately 1 to 6 weeks following knee arthroscopy [19–22]. To our knowledge, there are 3 studies that looked into the time required to recover driving fitness after hip arthroscopy. Two studies specifically assessed patients undergoing arthroscopic FAI surgery using a simple driving simulator. Interestingly, these two studies reported conflicting results. Vera et al examined 19 patients, who underwent FAI surgery, and compared the response to braking events in an age- and sex-matched cohort [3]. The study reported no differences in patient BRT at 2 weeks after surgery compared with preoperative value or BRT of control. This study was largely limited by the relatively small sample size as it included only 11 patients with right hip arthroscopy. In contrast, the study by Balazs et al tested 59 patients undergoing FAI surgery [4]. This study reported patients undergoing arthroscopic FAI surgery have prolonged preoperative BRT compared with healthy controls, and significantly prolonged BRT postoperatively, which normalised at 4 weeks. This study, however, did not disclose the number of patients undergoing arthroscopy on the right side.
A study by Momoya et al analysed patient brake performance using a realistic driving simulator similar to the one used in our study. Their study tested 14 patients who underwent right hip arthroscopy surgery and compared various parameters related to braking performance with 17 healthy volunteers [2]. The authors noticed significant improvement in brake performance in the first 2 weeks and concluded that driving after 2 weeks following right hip arthroscopy is recommended. The study was limited by the different degrees of soft tissue and bony intervention, which may have influenced braking performance, and by its modest sample size.
In our study, we classified the studied subjects according to the degree of soft tissue and bony procedure performed. Our result showed that braking parameter in patients undergoing SA normalised by 1 week compared with 2 weeks in the FAI group, suggesting that the degree of procedure on the hip joint may influence brake response. Another important finding of the current study is that while FAI or hip pain may lead to daily activity necessitating intervention, it does not lead to poor driving performance. This result is consistent with the report by Momaya et al and by Veral et al, but conflicts with the one by Balaz et al. This conflict may be derived from the patients’ preoperative conditions. Our patients had a mean preoperative VAS of 5.9 and a mean iHOT score of 41.3, which is slightly higher than in the study by Balazs et al. Moreover, although we refrained from using strong analgesics such as opioids on the day of the testing, the patients were allowed to take mild pain medications such as acetaminophen or anti-inflammatory pills that do not have sedative effects. Additionally, the mean difference in BRT in the study by Balazs et al was 53 ms, which appears to be a very small difference. The mean difference in our study was 21 ms, and we believe this is a negligible difference.
To our knowledge, our study is the only one which tested the study group at weekly intervals. We believe that this short duration between test sessions may provide more precise timing on the normalisation of the braking response, which was 2 weeks in the FAI group and 1 week in the SA group. Again, this result is consistent with the report by Momaya et al and by Veral et al but conflicts with one by Balaz et al, who reported a significant difference at 2 weeks. We were unable to find why such a conflict occurred among the studies.
On the other hand, while the VAS improved consistently following the surgery, we found that the patients’ pain did not correlate with their performance to brake. It could be hypothesised that the pain may be evoked as the patients step on the brake pedal, and that this may potentially decrease the stepping force, but such a trend was not found. A potential reason for this would be that the pain has improved significantly following the surgery and the amount of pain the patients experienced at the test sessions was not significant enough to influence the result.
The simulation seems to be the only viable option to test the patients’ fitness to drive, but this also provides a number of limitations. First, we tried to make the simulation environment as close to a real driving setting as possible but driving in actual automobile will different as there would be vehicle movement during actual driving. Additionally, patients will likely be more cautious when they are driving on the real road and this may have affected the result. Normalisation of the reaction to braking may be the essential prerequisite for return to driving; however, driving is a complicated procedure and there are numerous other factors that may affect safe driving. Therefore, it should be noted that the return of the reaction to braking to the preoperative state presented in our study does not guarantee the safety of driving after 2 weeks. The second limitation of this study is the potential learning effect, namely, the participants getting acquainted to the simulation driving. To validate whether the learning effect influenced the result of the testing, the control groups underwent the driving test 3 times at 1-week intervals, and our result showed that there was no learning effect during the 3-week trial. This suggests that the subjects reacted to the braking event as soon as possible; therefore, the braking response did not improve even with acclimation. On the other hand, the patients underwent 8 sets of tests and the learning effect may have occurred during the span of the test. Similarly, some patients may be more familiar with simulation driving. This so-called learning phenomenon has been tested previously by Momoya et al, who reported that no such phenomenon occurred over the 8 weeks when test sessions were carried out at 2-week intervals. Third limitation is with the relatively small sample size. Our study used 47 patients undergoing hip arthroscopy, making it one of the largest series that underwent a driving simulation test. Nonetheless, the larger sample size may provide a more precise threshold for when the braking time may normalise.