Estimation of return-to-sports-time for athletes with stress fracture – an approach combining risk level of fracture site with severity based on imaging

BACKGROUND The aim was to compare the return-to-sports-time (RTST) following stress fractures on the basis of site and severity of injury. This retrospective study was set up at a single institution. Diagnosis was confirmed by an interdisciplinary adjudication panel and images were rated in a blinded-read setting. METHODS 52 athletes (female, n = 30; male, n = 22; mean age, 22.8 years) with stress fracture (SFX) who had undergone at least one examination, either MRI or bone scintigraphy, were included. Magnetic resonance images (MRI) and/or bone scintigraphy (BS) of SFX were classified as either low- or high-grade SFX, according to existing grading systems. For MRI, high-grade SFX was defined as visibility of a fracture line or bone marrow edema in T1-, T2-weighted and short tau inversion recovery (STIR) sequences, with low-grade SFX showing no fracture line and bone marrow edema only in STIR and/or T2-weighted sequences. In BS images, a mild and poorly defined focal tracer uptake represented a low-grade lesion, whereas an intense and sharply marginated uptake marked a high-grade SFX. In addition, all injuries were categorized by location as high- or low-risk stress fractures. RTST was obtained from the clinical records. All patients were treated according to a non-weight-bearing treatment plan and comprehensive follow-up data was complete until full recovery. Two-sided Wilcoxon's rank sum test was used for group comparisons. RESULTS High-risk SFX had a mean RTST of 132 days (d) [IQR 64d - 132d] compared to 119d [IQR 50d - 110d] for low-risk sites (p = 0.19). RTST was significantly longer (p = 0.01) in high-grade lesions [mean, 143d; IQR 66d - 134d] than in low-grade [mean, 95d; IQR 42d - 94d]. Analysis of high-risk SFX showed no difference in RTST (p = 0.45) between high- and low-grade [mean, 131d; IQR 72d - 123d vs. mean, 135d; IQR 63d - 132d]. In contrast, the difference was significant for low-risk SFX (p = 0.005) [low-grade; mean, 61d; IQR 35d - 78d vs. high-grade; mean, 153d; IQR 64d - 164d]. CONCLUSION For SFX at low-risk sites, the significant difference in RTST between low- and high-grade lesions allows more accurate estimation of RTST by this approach. Both location of the injury and severity determined by imaging should therefore be considered for prediction of RTST.