Treatment of lumbar spondylolysis causes athletes to leave competition for at least a few months. All patients in this study returned to their original sport after the initial healing of lumbar spondylolysis but returning to sports too early caused recurrence. From this survey, the period to recurrence was within six months after returning to sports in 16 of 20 patients.
Early return to sports activity may cause recurrence of spondylolysis because bone strength of the healed lesion is insufficient. In this study, the average period after healing until recurrence was 149 days; the recurrence of spondylolysis did not occur immediately after returning to exercise. Suspending sports activity for longer can reduce the recurrence rate but lengthening the time to return to sports is not desirable for athletes. Therefore, strength of the lesion was sufficient and the timing to resume sports was appropriate. The average treatment period for the initial spondylolysis was 101 days, and the treatment period after the recurrence was 100 days. Therefore, we conclude that the cause of the recurrence was not that the treatment period was too short. All 20 patients returned to sports after the 2nd conservative treatment but two patients returned with a 3rd spondylolysis (2nd recurrence). Because adolescents are not physically mature, it is difficult to prevent 100% recurrence of lumbar spondylolysis in this age group, but that the patient had only partially recovered bone strength and the spondylolysis unfortunately recurred.
Recurrence in the earliest case occurred 42 days after healing. Therefore, recurrence may occur after an early return to sports. It will be necessary to instruct the athletes and parents to refrain from sudden high-intensity exercise immediately after resuming sports.
Males with lumbar spondylolysis with bone marrow edema constituted 79.6% of the subject population in our previous report [2]. The recurrence rate for males was 18/125 (14.4%) and for females was 2/27 (7%). The recurrent proportion of males in this study was 90%. Although the number of patients in this study was small, recurrence may be more likely to affect males.
There were initially five cases of bilateral spondylolysis and two after recurrence. Once suffering from lumbar spondylolysis, the patient learns related symptoms such as the location of pain. As a result, because the patient tends to visit the hospital earlier and more cases can be diagnosed as unilateral before reaching bilaterality, the number of bilateral cases at the time of recurrence is lower. Therefore, post-treatment follow up is necessary given the possibility of recurrence and patients and their parents should be informed about the possibility of recurrence.
Both the age of patients initially treated for spondylolysis and its recurrence, and the predominance of L5 in most cases of spondylolysis were consistent with previous reports [3].
Even in cases in which union was obtained once, some patients had non-union after recurrence. It has been proposed that there is a predisposition for lumbar spondylolysis in some individuals [4], and it is possible that the predisposition also affects the union rate. Individual characteristics may influence both fusion and the tendency for lumbar spondylolysis in the first place. As all recurrent cases had a history of bony fusion once, they have some potential to regenerate bone. However, this study showed that patients may not always heal even if the same treatment led to bony union in the past. We are paying close attention to recurrence after returning to exercise and are considering ways to detect recurrence earlier. As MRI is useful for early detection of lumbar spondylolysis [5], it is important to regularly perform additional MRI analyses to confirm the presence or absence of recurrence. In addition, when low back pain appears after returning to sports following treatment for lumbar spondylolysis, MRIs should be taken early for diagnostic purposes.
The limitation of this study is that the number of cases is small and statistical examination is insufficient for univariate or multivariate analysis.