In the current study, partially localized abnormal signals in the bone marrow on axial MR imaging was found even among patients with no abnormalities at both initial and follow-up radiographic exams. Abnormal signals extending throughout the entire bone marrow consistently showed radiographic abnormalities in later images. MR imaging with the initial radiographic abnormalities showed partial or entire bone marrow abnormalities. Thus, the area of abnormal signals in the tibial bone marrow in the axial MR imaging might predict later radiographic changes. In addition, abnormalities in the medial-posterior cortex in the axial MR imaging tended to be related to radiographic changes.
Group 1 showed negative finding on examination in both initial and follow-up radiographs, which recognized as shin splint. Group 2 showed positive findings in follow-up radiograph that recognized as stress fracture. Positive finding in initial radiograph in Group 3 was not clear which results of stress fracture or acute injuries on previous stress fracture site. This study showed the characteristic axial MR imaging of tibial stress injuries in each condition. There were no differences in MR images according to middle or distal part of tibia.
Aoki reported that coronal MRI scans can distinguish stress fractures and shin splints before radiographs show a detectable periosteal reaction of the tibia . Abnormally high signals in the bone marrow on short tau inversion recovery (STIR) or fat-suppressed MRI are wide and continue to the cortex in stress fractures; however abnormally high linear signals in the bone marrow are seen in shin splints. Ahovuo reported that detection of stress injuries in axial images of the tibial shaft shows higher rates of diagnostic accuracy than in coronal images . This study showed that an axial MR imaging detected relative involvement in the periosteum, bone marrow, and cortical bone.
In the current study, abnormal periosteal signals were found in the medial or medial-posterior surface of the tibia in all groups. This observation might be related to traction periostitis along the insertion of the soleus fascia, flexor digitorum longus, and tibialis posterior, implying a potential cause of shin splints [14–16].
Partial bone marrow abnormalities did not involve periosteal reactions at follow-up radiographs in this study. Similarly, abnormal high linear signals on STIR in coronal MRI scans are related to shin splints . This abnormally high signal intensity in bone marrow might be secondary to edema or hemorrhage related to microdamage and the associated repetitive response. Johnell reported that the metabolic activity in the biopsy bone was increased at the medial edge of the tibia with shin splints . Shweitzer reported that an increase in bone marrow signal intensity on MR imaging was caused by altered weight-bearing . This partial bone marrow abnormality might be related to changes in load and metabolism of tibial bone marrow.
Abnormalities in the entire bone marrow were related to abnormalities in the medial-posterior cortex of the tibia on axial MR imaging. Stress fractures show intramedullary abnormal signal continuing with the cortex of the tibia on coronal MRI scan . This cortical abnormality might predict a periosteal reaction such as a stress fracture, as well as the area of abnormal signals in bone marrow.
The relationship between stress fracture and shin splint is controversial. Aoki et al. reported that patients with shin splints did not have a stress fracture at the late phase. Therefore they think that shin splints may have some relation to stress reaction of bone but are likely different clinical entities from stress fractures. Other authors reported that shin splints are part of continuum of fatigue damage of stress fracture [2, 5].
Anderson et al. reported that stress fractures shows abnormal marrow signal intensity with linear signals abnormally extending through the anterior cortex with periosteal fluid . Fredericson et al. reported that periosteal edema reflects periostitis or shin splints . Abnormal linear signals involving the cortical surface are thought to represent a relatively early physiologic response of bone to stress . More severe edema of both the periosteum and bone marrow are found to represent increasing severity with periosteal edema . Therefore they concluded that periosteal edema may be seen as the initial injury on the spectrum, which, if allowed to progress, would evolve into a more serious bone injury .
In the current study, no patients with periosteal abnormalities and partial bone marrow abnormalities on axial MR imaging showed findings of periosteal reaction on follow-up radiograph. Patients with abnormalities in the entire bone marrow or medial-posterior cortex showed periosteal reaction on follow-up radiograph. This finding suggests that periosteal reaction on follow-up might represent the initial response of the tibia to repetitive stress. The bone marrow abnormality initially occurs in part of the bone marrow, and then extends to the entire tibia if the abnormally relates to micro damage of the bone marrow. The medial-posterior cortex might react to the stress and a periosteal reaction occurs if the damage to the bone marrow expands to the entire bone marrow.
Conventional radiographs are the primary tool for diagnosing suspected stress injuries; however, radiographic findings sometimes lag behind the MR imaging . The MR findings with initial radiographic abnormalities might represent a subsequent reaction following initial findings. This subsequent reaction might represent the healing course of edema or hemorrhage related to microdamage.
There are some limitations to this study. The first limitation is that the interval between the symptom onset and MR imaging was not definite. However the mean duration of symptoms before MR imaging was not different in Groups 1 and 2. In addition, the duration between radiographs and MRI did not show differences. Therefore we believe that findings in MR imaging are available for early discrimination of tibial stress injuries. The second limitation is that follow-up MR imaging was not evaluated in this study. The MR findings with initial radiographic reaction were thought to represent a reaction to tibial stress injuries. However a study including consecutive MR imaging is needed in the future. Despite these limitations, there have been no reported studies of tibial stress injuries using high resolution MR imaging.