Occult lesions are real and they need attention. Ignore them at your peril.
By Shahid Punwar, Margaret Hall-Craggs and Fares Haddad
The terms ‘bone bruises’ and ‘bone contusions’ are commonly used to refer to areas of localised bony tenderness after a traumatic injury, where there is no X-ray evidence of fracture. These painful injuries are classified as ‘sub-periosteal haematomas’; while they can persist for several weeks, they usually resolve spontaneously with no further consequences.
However, ‘bone bruise’ is also used to describe a group of traumatic bone lesions which are quite distinct from sub-periosteal haematomas. This second group is also known as ‘occult bone lesions’ as the injuries are usually not visible on X-ray or by observation (eg at arthroscopy), but they do show up on magnetic resonance imaging (MRI). They are a feature of certain common sports injuries, most notably knee ligament tears.
There has been much discussion about the importance of these lesions in predicting the severity and prognosis of injuries, in particular when associated with ACL rupture and the subsequent development of osteoarthritis.
First described by Mink and Deutsch in 1989(1), MRI bone bruises have only recently been widely recognised. This is mainly because of the increasing use of MRI in the investigation of acute knee injuries.
MRI shows bone bruises as localised abnormalities in the subchondral bone (just beneath the joint-lining cartilage) and marrow, because of the increased water content of the injured area. (The best MRI images of bone bruises are produced from Short T1 Inversion Recovery [STIR] sequences.)
These areas of injury are thought to represent micro-trabecular fractures of cancellous (spongy) bone, haemorrhage and swelling of the marrow, without significant disruption to the adjacent bone or overlying articular cartilage. In histological studies Rangger et al (1998)(2) found blood and fluid in the region of the MRI abnormalities, but we still await conclusive evidence of the exact nature of bone bruises.
In Mink’s original classification, a bone bruise was just one of several different types of occult lesion, however the term has now evolved to include all of the various sub-types.
Vellet et al (1991)(3) defined five types of bruise on the basis of their architectural appearance and relationship to cortical (compact outer layer) bone:
Geographic lesions are found close to the edge of the compact bone, whereas reticular lesions are entirely contained within the marrow. Impaction fractures show depression of the articular surface and osteochondral fractures communicate with the joint space.
How common are bone bruises?
Most of what we know about the incidence and prevalence of bone bruises comes from studies involving the knee, in particular after ACL tear.
Vellet’s study of 120 patients with acute traumatic haemarthroses (bleeding into the joint space) of the knee identified bone bruising in 72% of patients. Bone bruises are seen in 80% of patients with an acute ACL injury(1,4). The phenomenon is also seen in conjunction with meniscal, posterior cruciate and collateral ligament injuries of the knee.
Other anatomical sites where bone bruises have been identified include the hip, ankle, elbow, foot and spine. As they can only be detected on expensive MRI scanning, it is likely that most of these bruises go unrecognised.
Though occult bone lesions may occur wherever there is direct trauma to bone, most knee and ankle sprains are the result of forceful twisting injuries. It is thought that the bruises are caused as adjacent bones are forcefully impacted (‘kissing contusions’). In ACL tears, bone bruises are mainly seen in the lateral tibial plateau (outer top edge/surface of tibia) and femoral condyle (lower end of the femur), which is consistent with the valgus (outward force) mechanism of injury.
Why some people with the same injury get bone bruises and others don’t can be put down to variations in the exact mechanism and force of each individual injury.
How long do they take to heal?
How quickly a bone bruise heals will clearly depend on the severity of the initial injury, as well as the level of activity in the healing phase.
A recent systematic review of the literature (5) identified 266 articles but only included 13 as quality studies. The percentage of complete healing of bone bruises on MR scanning ranged from 88 % after 11-16 months of follow-up in one study to 100 % after 5-12 months of follow-up in another.
Comparison between studies was hindered by different outcome measures and varying lengths of both clinical and MRI follow-up.
As bone bruising in the knee is so often associated with ligamentous and meniscal damage, it is difficult to attribute clinical symptoms to bruising alone. For this reason MRI is the best way of demonstrating healing.
Will they lead to osteoarthritis?
Osteoarthritis is a recognised complication of conservatively treated ACL tears. It is generally thought that the global instability of the ACL-deficient knee predisposes it to ‘wear and tear’ of the joint surface. However, this does not explain why the incidence of osteoarthritis (OA) in reconstructed knees has been found to be equal to, or even greater than that in conservatively treated knees(6).
The fact that bone bruises are commonly found in ACL injuries has raised the possibility that they may either be a prognostic indicator or a direct cause of degenerative changes.
In the review by Boks(5), all cases of reticular bruising (confined to the bone marrow) proved to have healed fully at follow-up MR scanning. By contrast, cartilage loss was detected in all cases of initial cortical impaction and osteochondral fracture, although, as these lesions by definition disrupt the articular surface, this is not surprising.
Of most interest is the variable prognosis for geographic lesions. While Vellet found that 78% of patients (14 of 18) with a geographic lesion had evidence of degenerative changes next to the site of the initial lesion, Costa-Paz et al (2001)(7) found only one of 11 geographic lesions present after 24-64 months. This difference could be accounted for by the much longer follow-up period used compared with Vellet’s study, during which time further healing may have occurred.
There are two main theories explaining how occult osseous lesions that do not disrupt the joint surface (ie geographic bruises) may lead to OA.
Animal studies have shown that impactive cartilage loading, which causes cartilage changes consistent with osteoarthritis, is also associated with evidence of subcortical microtrabecular fractures(8). This suggests that the presence of geographic bruises may indicate a significant unseen injury to the articular cartilage caused by the initial trauma. In this way geographic bone bruises could be seen as a predictor of cartilage degeneration rather than a direct causative factor in OA.
An alternative hypothesis is that once geographic bruises heal, they leave a stiffer construction than the previously normal spongy bone. The decreased compliance might then generate greater mechanical loads on the overlying articular cartilage, leading to progressive degeneration(9).
Because MRI scanning is such a common part of the assessment of knee injuries, most of the study of bone bruises to date tends to have been of knees. As MRI gains popularity in the evaluation of other injuries, it is likely that more associations will appear. These lesions may provide an explanation for ongoing pain in clinically intact limbs and in at least some cases may tip the balance in favour of MRI scanning rather than diagnostic arthroscopy. Bone bruises may also provide a potential target for therapeutic agents aimed at preventing the development of OA after ACL and other injuries.
Johnson (2000)(10) has shown that severe geographic bone bruising after injury is linked to increased disability, including prolonged knee effusion, pain and increased time to achieve normal range of motion of the joint. When combined with ACL or other ligamentous and meniscal pathology, this may affect how long it takes for the patient to be ready for an operation and the length of post-operative rehab.
After ACL reconstruction the presence of bone bruises may require rehab therapists to modify their regimens. Some studies, for instance, have shown that closed kinetic chain rehabilitation programmes produce high axial forces, which would suggest that open kinetic chain work would be preferable. Indeed, there may be a general argument in all leg injury rehab where bone bruising is present to refrain from weight-bearing activities until the lesion has healed, to reduce the risk of later bone breakdown.
If not adequately protected during trabecular healing, bone bruises may represent regions of bone theoretically at risk from the subsequent development of further problems such as insufficiency fractures. There is clearly a need for further studies involving both long-term clinical and MR follow-up to clarify the natural history of bone bruises.
When Mink first described bone bruises in 1989, they were thought to be a purely benign abnormality. Early follow-up studies showed both clinical and MR resolution without any evidence of a progressive abnormality. However, it now seems probable that certain types of bone bruise are associated with changes in subchondral bone, which may contribute to the early degeneration of articular cartilage, and the subsequent development of osteoarthritis.
The term ‘bone bruise’ itself is nonspecific and should be used with caution. In order to avoid confusion, clinicians should make it clear whether they are referring to periosteal injuries or MRI findings.
1.Mink JH, Deutsch AL (1989) ‘Occult cartilage and bone injuries of the knee: Detection, classification and assessment with MR imaging’ Radiology 170:823-9
2. Rangger C, Kathrein A, Freund MC, Klestil T, Kreczy A (1998) ‘Bone bruise of the knee, histology and cryosections in 5 cases’ Acta Orthop Scand 69(3): 291-4
3.Vellet AD, Marks PH, Fowler PJ, Munro TG (1991) ‘Occult posttraumatic osteochondral lesions of the knee: Prevalence, classification and short-term sequelae evaluated with MR imaging’ Radiology178: 271-6
4. Spindler KP, Schils JP, Bergfeld JA et al (1993) ‘Prospective study of osseous, articular and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy’ Am J Sports Med 21(4): 551-6
5.Boks SS, Vroegindeweij D, Koes BW, Hunink MG, Bierma-Zienstra SM (2006) ‘Follow-up of occult bone lesions detected at MR imaging: Systematic Review’ Radiology238: 853-62
6.Friederich NF, O’Brien WR (1993) ‘Gonarthrosis after injury of the anterior cruciate ligament: A multicenter, long-term study’ Z Unfallchir Versicherungsmed1993; 86: 81-9
7.Costa-Paz M, Musculo DL, Ayerza M et al (2001) ‘Magnetic resonance imaging follow-up study of bone bruises associated with anterior cruciate ligament ruptures’ Arthroscopy17: 445-449
8.Radin EL, Parker HG, Pugh JW, Steinberg RS, Paul IL, Rose RM (1973) ‘Response of joints to impact loading, III. Relationship between trabecular microfractures and cartilage degeneration’ J Biomech6: 51-57
9. Rosen MA, Jackson DW, Berger PE (1991) ‘Occult osseous lesions documented by magnetic resonance imaging associated with anterior cruciate ligament ruptures’ Arthroscopy 7: 45-51
10.Johnson DL, Bealle DP, Brand JC Jr, Nyland S, Caborn DN (2000) ‘The effect of a geographic lateral bone bruise on knee inflammation after acute anterior cruciate ligament rupture’ Am J Sports Med28(2): 152-155