A prospective case-series study was performed. In order to maintain validity throughout the process, we sought to adhere to the checklist of 22 items outlined in the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement . The procedures for study design, data collection and analysis were comparable to earlier studies performed in adolescent tennis players [4, 7].
Eighteen male participants between the ages of 16 and 18 (mean age 17.8, range 16.9–18.6 years) contracted to a single professional football academy in England were recruited. The recruitment was conveniently undertaken from a single squad within the club. The participants included 2 goalkeepers, 10 defenders, 5 midfielders and 1 attacker. None of these players had a cardiac pacemaker, cochlear implant, aneurysm clips (brain), or other bodily metalwork. Importantly, none of the players had any clinical signs or symptoms suggestive of lumbar spine pathology (e.g. lower back pain, radicular symptoms).
Contracted to a single professional club in England.
Any bodily metal work (e.g., cardiac pacemaker, aneurysm clips) which would preclude a safe MRI.
No clinical signs or symptoms of lumbar spine pathology (e.g., lower back pain, radicular symptoms).
A risk assessment was performed prior to proceeding with the study in order to identify possible associated risks, their likelihood, and their potential impact on stakeholders. Ethical approval was obtained from the Ethical approval was obtained from the University of Birmingham School of Sport and Exercise Science ethics committee (MCR040219-1). Participants were issued a Participant Information Sheet and provided informed written consent prior to data collection. Incidental and untoward findings were managed according to The Royal College of Radiologists’ guidelines on the Management of Incidental Findings During Research Imaging . For example, if unsuspected findings were deemed of clinical significance, arrangements were made to share this information with the player through discussion with the medical team, and where appropriate, further investigations were arranged.
Imaging was carried out over two separate days of data collection at (redacted for blinded manuscript), using a 3T Siemens Prisma scanner. Sagittal T1, T2 and FS (fat-saturated) T2 sequences were obtained as well as axial T2 sequences through the lower three levels and a 3D T1 spoiled gradient echo VIBE (Volumetric Interpolated Breathhold Examination) sequence. The 3D T1 VIBE sequence is isotropic and of thin slice section allowing multiplanar reformatting and relatively high-resolution evaluation of the pars interarticularis, similar to computed tomography (CT). It potentially negates the need to expose athletes to ionising radiation as is the case when using CT to exclude or further evaluate pars interarticularis fractures. . In two patients, the pars interarticulares were not completely imaged on the sagittal sequences as the MRI technician had not protocolled the sequence to image sufficiently laterally. Although the pars interarticulares were imaged completely on the remaining sequences (including the 3D T1 spoiled gradient echo VIBE sequence), these two participants were recalled for completion sagittal sequences.
Two consultant musculoskeletal radiologists (redacted), both with over a decade of experience, reviewed the images independently with findings agreed by standard consensus. Each MRI scan was assessed according to the following categories; facet joint arthropathy, presence of synovial cysts, pars abnormalities, disc abnormalities, spinal cord abnormalities and any other radiological findings as per previous studies [4, 7].
Abnormalities were reported using validated classification criteria. Facet joint arthropathy was classified using a system developed by Weishaupt et al., with grades classed as normal (grade 0), mild (grade 1), moderate (grade 2), or severe (grade 3) . Synovial cysts were described as rounded, fluid containing lesions related to the facet joint, found in the epidural, foraminal, or paravertebral positions. Pars injuries were assessed using a classification system by Alyas et al..7 which evolved from a system originally proposed by Hollenberg et al. . They were described as normal (grade 0); chronic stress reaction (grade 1); subtotal stress reaction (grade 2); acute stress fracture (grade 3); or chronic stress fracture (grade 4). Disc degeneration was reported according to classification criteria by Pfirmann et al. (grade 1 to grade 5; normal to severe) . Disc herniations were described as circumferential (> 50% of the disc circumference); broad-based herniation (25–50% of the disc circumference), central herniation (< 25% of the disc circumference), extrusion or sequestration . Modic end plate changes were recorded if present as type 1 to 3 . Indentation of the thecal sac was noted as present or absent. Nerve root involvement was noted as none, contact but no deviation, deviation and compression. The presence of intervertebral disc herniations, such as Schmorl’s nodes and limbus vertebrae, were also documented. These classifications enabled comparison of results with existing literature in different sports.