In this study, multivariable logistic regression analysis was performed to investigate the factors affecting bone union of acute fractures in adolescent patients receiving conservative treatment for lumbar spondylolysis. We found that the factors affecting bone union were the main and contralateral side stages. Sex, age, level of lesion, and SBO had no effect on bone union. Notably, the main side progressive and contralateral side terminal stages were the negative factors for achieving bone union. In contrast, main side pre-lysis and early stages and absence of the contralateral side lesion were positive factors for achieving bone union.

Previous multivariable studies on the conservative treatment of lumbar spondylolysis have reported the affected level of lesion and stage, other factors such as gap size of the lesion, lumbar lordosis, contralateral side stage, and reduced flexibility as unfavorable factors for bone union [12, 13].

However, the bone union rate of these studies ranged from 17 to 67%, which is lower than those reported by recent studies [3, 9, 12, 13, 15]. Furthermore, the previous study included main side terminal stage lesions, in which bone union is considered impossible to achieve. Therefore, the treatment results in the previous study are not in line with the current understanding of various factors affecting the conservative treatment of lumbar spondylolysis [12, 13].

Our study is novel compared to previous reports in that: 1. the “pre-lysis stage” lesions, which is a new stage concept reported recently and defined as a lesion showing no fracture line on CT but with a signal change on MRI [3, 14, 16, 18], were included, accounting for 31% of all lesions in this study and 2. patients with the main side terminal stage lesion, in which bone union is difficult or unable to be achieved based on previous studies [1, 3], are excluded.

To the best of our knowledge, this is probably the largest conservative treatment report evaluating bone union with CT, both in terms of the number of patients and lesions [1,2,3, 9,10,11,12,13,14,15,16]. Unlike reports that examine single factors, such as SBO, level of lesion, and stage, the present study has the advantage of uniformly evaluating all cases and collecting a sufficient quality of data and number of patients for multivariable analysis.

From the results of this study, the factors affecting bone union in patients with lumbar spondylolysis receiving conservative statement were the stage of the main and contralateral sides, which is consistent with the results of previous studies that reported the association of the fracture of bone ring structure in the vertebral arch in the axial plane with local instability and bone union rate in the conservative treatment of lumbar spondylolysis [9, 11].

In other words, the more advanced the stage of the lesion, the larger the fracture gap, the greater the instability, and the less favorable it is for bone union. Bilateral lesions with two fractures are less likely to heal than unilateral lesions with a single fracture of the bone ring. The rate of bone union is also greatly reduced when the gap between the two fractures is large due to the increased instability [9, 11].

The previously reported factors, such as the level of lesion, SBO, and age, were not significantly associated with bone union in the multivariable analysis.

L5 lesions are known to negatively affect bone union, and the association between bone union and level of the lesion cannot be completely ruled out based on the results of this study alone. It is considered analogous that the stage carries a lot of weight, and the level of lesions and SBO are only confounding factors [3, 9,10,11,12,13].

Based on the results of the present study, the evaluation of factors affecting bone union at the time of the initial examination may assist in determining the treatment option (Fig. 3). As for imaging studies, only a single CT scan is sufficient to evaluate the stage of the lesion.

Treatment options according to factors affecting bone union. Evaluation of the main and contralateral staging at the initial visit will aid in selecting the treatment option. If there are more positive factors for bone union, conservative treatment should be selected. If there are more negative factors, pain control and surgical intervention should be considered. If intermediate, the use of low-intensity pulsed ultrasound (LIPUS) should be considered

If there are more positive factors for bone union, the rate of bone union is expected to be higher; thus, conservative treatment aimed at achieving bone union is the first choice. If there are negative factors for bone union, low-intensity pulsed ultrasound, which has been reported to improve the rate of bone union, pain, and functional disability and shorten the treatment period, may be an option in addition to conventional conservative treatment [19,20,21].

In lesions that are expected to have a particularly low rate of bone union as they have many negative factors for bone union, if conservative treatment failed to achieve bone union, pain control with oral painkillers or interventional pain injections, early surgery due to pain, or repair surgery for lesions that became pseudoarthrosis may be effective depending on age and sports level [6,7,8, 22,23,24,25]. To further develop the treatment of lumbar spondylolysis, prospective studies on the ability to predict the rate of bone union at the time of the initial visit are a future prospect and challenge.

The present study had several limitations. All patients in the study received uniform imaging evaluations and the conservative treatment protocol. The number of dropouts was therefore relatively high (26%). The remaining patients were those with good treatment compliance, which could be a selection bias affecting the outcome of conservative treatment. In addition, all patients were treated in an outpatient setting, and not all patients were evaluated for height, weight, body mass index, or pain course using a uniform scale. The study assessed local factors such as the level and stage of the lesion, as well as demographic and clinical variables such as patient sex and age, and the presence of SBO. However, the sagittal alignment of the lumbar spine was not included to avoid too many variables in our multivariable analysis. Given the limited number of patients, this would have reduced the statistical power of our findings. Thus, alignment was omitted; but it should be noted that this was a major limitation of the study. Moreover, it was difficult to perform the imaging evaluation and conservative treatment in this study at all facilities due to cost, insurance, and other factors. Careful judgment should be exercised when using the results of this study in actual treatment. We have presented treatment options based on the present results; however, further research is needed to determine whether our approach can consistently improve bone union rates. A multicenter study is warranted to further include patients with varying background characteristics.