Table 2

MRI case–control studies

StudyCunningham et al(2007)Robinson et al(2004)Zoga et al(2008)Brennan et al(2005)Verrall et al(2001)Paajanen et al(2008)
Case-study group participants (incl. sports type, gender, age)100 athletes (soccer)
95 Male, 5 Female. Age: 17–38
52 athletes (35 soccer, 13 rugby, 4 athletics), of whom 27 previous surgery
51 Male, 1 Female. Age 17–36
141 patients (127 athletes: 41 American football, 23 soccer, 15 running, 14 hockey, 34 others)
134 Male, 7 Female. Age 17–71
18 athletes (15 soccer, 3 rugby)
Male. Age 19–32
89 athletes (Australian football)
Male. Age 17–33
16 athletes (13 soccer, 1 cross-country skier, 1 runner, 1 ice-hockey)
14 Male, 2 Female. Age 22–38
Controls100 athletes (50 rowers, 50 soccer)
Sex unknown. Age :18–28
6 athletes (5 soccer, 1 rugby)
Male. Age 17–37
25 (physical activity unknown)
Male. Age 18–39
70 athletes (rowers)
Male, Age 17–34
Group 1: 17 distance runners
Group 2: 10 sedentary men
20 athletes (10 soccer, 10 ice-hockey)Male. Age 19–27
Inclusion criteriaCases: Debilitating groin pain and symptoms and signs at pubic symphysis
Controls: Asymptomatic, hip pain and pain from sacroiliac joints
Cases: Groin pain>3 months
Controls: no groin symptoms or injury, no groin surgery
Cases: Patients referred with diagnoses ‘athletic pubalgia” and ‘sports hernia”
Controls: Asymptomatic

Cases: Patients referred with suspicion of groin injury
Controls: Asymptomatic
Cases: recruited at end of pre-season after 6 weeks intensive training
Control 1: age-matched, min. 6 weeks intensive training
Control 2: no prior history of groin pain, no physical exercise within 6 weeks, age-matched
Cases: Osteitis pubis (established by exclusion of other groin disorders, typical clinical history and signs, pelvic radiographs isotope bone scan and MRI)
Controls: Asymptomatic
Exclusion criteriaCases: Sports hernia at clinical examination before referral
Controls: groin pain
Cases: Acute groin injury, insufficient clinical and surgical details availableCases: Inadequate MRI of pubic region. Images older than 120 days from time of physical examinationNot describedNot describedCases: Inguinal hernias, iliopsoas and abdominal muscle-related pain. chronic prostatitis.tendinitis of the groin, bursitis or hip disorders
Clinical findings (1) type of examination
(2) findings
(1) Not described
(2) Groin pain for average of 3 months
(1 and 2) Not described(1) Not reproducible but details given
(2) 93 positive for rectus abdominis tendon lesion, 15 for adductor compartment lesion,71 for both, 16 for osteitis pubis (not defined)
None(1) Not reproducible but details given
(2) 52 athletes positive for current groin symptoms and signs (tenderness on palpation of symphysis and superior pubic rami)
(1) Not described
(2) All cases have tenderness of pubic symphysis >3 months. 3 cases with adductor-type pain on palpation
Diagnostic entity used by authorsOsteitis pubisAdductor dysfunctionAthletic pubalgiaOsteitis pubisOsteitis pubisOsteitis pubis
Radiology protocol (1) Field strength
(2) Sequences
(1) 1.5T
(2 )Coronal T1& STIR, axial T2
(1) 1.5T
(2) Coronal: T1& STIR, axial T2. Oblique axial: T1, T2 FatSat, T1Fatsat postgadolinium
(1) Cases: various (117 at 1.5T, 2 at 3.0T, 16 at 0.2–0.3T, 6 at 0.6–0.7T)
Controls: 1.5T
(2)Protocols not defined
(1) 1.5T
(2) Cases: coronal T1 and STIR, axial T2
Controls: coronal STIR only
(1) 1.5 and 1.0 T
(2) Coronal and axial:T1 and T2 FatSat
(1) 1,0 T
(2) Coronal and axial: T1 and STIR
Additional radiologic examinationsNoneNoneNonePelvic radiographs
Fluoroscopy-guided injection of contrast in symphyseal joint
NonePelvic radiographs and isotope bone scan
Evaluation of MRI scan2 Radiologists blinded to side of symptoms. Diagnosis by consensus.
Presence of secondary cleft sign
Degenerative changes at symphyseal joint

2 Radiologists blinded to clinical details. Diagnosis by consensus
BMO (graded 0–2)
Degenerative changes at symphyseal joint
Musculotendinous structures of the groin & abdominal wall
Graded at 1 and 2 reading (0–2)
3 Radiologists blinded to clinical details. Diagnosis by consensus
BMO (not graded)
Osteitis pubis (BMO with degenerative changes at symphyseal joint)
Abnormal rectus abdominis and adductor tendons
Presence of secondary cleft sign
2 Radiologist: 1 blinded and 1performing fluoroscopy-guided injections.Diagnosis by consensus
Presence of secondary cleft sign
BMO (not graded)
Degenerative changes at symphyseal joint
2 Radiologists blinded to clinical details. Diagnosis by consensus
BMO (graded 0–3 and extent>2 cm)
Degenerative changes at symphyseal joint
Abnormal conjoint tendon
Inguinal hernia
2 Radiologists blinded to clinical details. Diagnosis by consensus
BMO (graded 0–3)
Degenerative changes at symphyseal joint
Abnormal conjoint tendon
Abnormal adductor enthesis
Reproducibility of radiologyNoneInterobserver agreementNoneInterobserver agreementNoneNone
‘Gold standard’ usedInjection of contrast in pubic symphyseal cleft to demonstrate secondary cleft sign (CSG only)NonePhysical examination in 141 cases
Surgery in 102 cases
Fluoroscopy-guided injection of contrast in symphyseal jointNoneSurgery in 8 patients not responding to conservative treatment of groin pain
Conclusions/resultsSecondary cleft sign in 48.8%, osteitis pubis in 9% of cases. Both in 43%.
Secondary cleft sign and BMO only present in cases but not controls.
Abnormal anterior pubis and enthesis enhancement correlates significantly with clinical side of symptoms
All other abnormal imaging findings show no correlation with clinical side of symptoms
Positive MRI findings in 138 cases: Secondary cleft sign in 66 cases, BMO in 63 cases.
MRI sensitivity 68% and specificity 100% for rect. abdom., 86% and 89% for adductor lesions compared to surgery
Secondary cleft sign identified with 100% sensitivity and specificity on MRI and fluoroscopy. Present in 12 cases. Secondary cleft sign corresponds to side of symptoms72% of athletes (62 cases and 8 runners) have pubic BMO.
62% of those (47 cases and 5 runners) have current groin pain and symptoms at clinical exam
The difference in BMO in the 3 groups (surgical, conservative and control) is not statistically significant, although grade 3 oedema is not found in asymptomatic controls
  • BMO, bone marrow oedema.