We suggest that arthroscopy should be considered early in athletes with clinical presentation of exertion ankle pain, effusion and palpation tenderness along the joint line after a previous ankle sprain. The definition of "early" could of course be argued, but 7 months is certainly a long delay for a professional or semi professional player. In fact, most full-time professional players in this study (14/17) had even longer than average 7 months duration from injury to arthroscopy (10–36 months). A number of these cases were high profiled professional players who said that they had struggled for long time but "put up with the pain" with aid of symptomatic treatment. In most of these cases, the team medics often referred to a "normal MRI" as an important delaying factor, but also lack of definite diagnosis assumed to be a "low grade sprain", pressure to play from managers and financial implications of a long absence from playing were discussed. Only 3 international footballers were operated early (within 3 weeks from injury) due to severe pain and hemarthrosis and video recorded injuries demonstrating a substantial direct impact to the ankle from opponent's shoes or studs in tackle situations. They all had undergone MRI showing effusion with no chondral injuries, but were found to have major full thickness cartilage tears on arthroscopy. They were all back to play within 8 weeks from surgery, which is shorter than the average of 16 weeks for the other more chronic injuries.
Arthroscopic treatment was successful in a majority (73%) of all the studied players, and 15/17 professional top players could resume full play after surgery within reasonable time. In two cases, these injuries obviously ended their career. The long duration of symptoms prior to appropriate diagnosis and treatment could well be suspected to have aggravated the outcome of treatment. Since all the patients in this study were athletes, performance from injury to treatment was also likely to have been affected, even though we lack detailed information. Secondly, the fact that 36% of the players still had some degree of discomfort despite treatment may suggest that some of these injuries may turn into chronic problems and, possibly in the long term, osteoarthritis. An ankle arthroscopy is a relatively straight-forward procedure, which can be done as an outpatient's procedure, [3, 7, 8] and its complications from arthroscopy are rare .
The majority of cartilage lesions in our study were located on the Talus dome, which corresponds to other studies [1, 3, 6]. This can, to some extent, be explained by the work of Athanasiou et al.  who pointed out differences in the mechanical properties of specific human cartilage regions. For example, we know that the Tibia cartilage is stiffer than that of the Talus. The anterior and posterior regions of the lateral and medial sites of the tibia were found to be 18–37% stiffer than the anatomically corresponding sites in the Talus. Our findings correspond with Hintermann et al  and Hirose et al  who found that 62% of cartilage lesions in his study were on the medial aspect of the Talus. Only 17% were found on the lateral part of Talus. Some patients in our study had minor lateral ligament damage as well but stated that instability was a minor problem compared with the pain they experienced. Isolated chondral lesions may not be as common as lesions associated with ligament or bony damage. However, it is possible that our cases may include sprains as well as sprain-direct impact injuries, as demonstrated from some video analyses from the injury made available for us. There are unfortunately no systematic data available on this, but it certainly warrants further investigations. It is logical to assume that direct stud or shoe impact injuries to the ankle in a tackle situation can cause chondral damages as well as ligament tears from the pure impact and we suggest this mechanism as an important explanation for chondral injuries of the ankle as well as biomechanical factors included in sprains as discussed by others . Van Dijk et al  found an association between the rupture of the lateral ligaments and medial joint pain. They found 19 cartilage lesions located on the medial malleoli or medial talus in 30 patients with lateral ligament ruptures. These results can be explained by Noguchi et al . They showed that an increase of stress distribution on the medial side of the ankle joint occurred when the lateral ligaments were released. Harrington , on the other hand, suggested unbalanced loading of the medial joint space as be the primary cause for the development of degenerative arthritis. As the Tibia cartilage on the medial and posterior sides is 18–37% stiffer than its corresponding Talus sites, even minimal impact can cause damage to the cartilage of the medial Talus dome .
Within the limitations of our sample size, the use of conventional MRI to delineate articular cartilage injuries must be argued, in particular since 50% of the talar dome cartilage lesions were full thickness tears. Whilst both the initial and second radiologists agreed completely on the presence of subchondral oedema, the presence of chondral injuries was more controversial (19% versus 41%). Nevertheless, both of these figures are inadequate for clinical practice. The only sequences that accurately depicted a chondral lesion were the sagittal T2 Turbo spin echo (TSE), T1 spin echo (SE), fat saturated fast spin echo (FSE), T1 FSE and coronal fat sat FSE. We noted that all patients had an axial scan. The only sequence that accurately and clearly delineated the cartilage of the talus was the T2 TSE. The fast spin echo was no better than the conventional spin echo. MRI of articular cartilage is challenging because most articular surfaces are curved and thin and images are also prone to artefacts. The closely applied joint surfaces with thin cartilage and complex osseous anatomy make conditions challenging for the radiologist [4, 6]. We speculate that the inaccuracies of the scans are due to the fact that the majority of sequences can not define the cartilage from the subchondral bone. The cartilage is either the same intensity as the bone or fluid. Secondly, the slices are too thick. Images were usually 3 mm thick, taken at 1 mm intervals. One scan was actually taken with 4 mm slices. As the cartilage in the ankle is so thin, these slices are far too thick to accurately delineate a lesion. A higher spatial resolution is therefore needed. In the ankle where there is minimal fat, suppression is not needed and results in fluid and cartilage producing the same signal intensity. It should be stressed that we do not know the initial detailed request to the MRI department by the referring Team doctor, which could of course influence the choice of sequences. Our findings correlate with the majority of authors who have studied the usefulness of MRI in the detection of similar lesions in the knee. Friemart stated that the sensitivity of detecting cartilage lesions in the knee by MRI can range from 15% to 96% . Rubin et al demonstrated that the presence of subchondral oedema in the knee may be indicative of a defect in the overlying articular surface . Even small defects of the articular surface were frequently associated with relatively large areas of subchondral marrow oedema. This seems to apply also for ankle injuries. The depth of oedema should also be taken into account as generally the deeper the oedema the more severe the cartilage lesion. More research is needed in this field. A technique that reliably detects cartilage lesions will greatly aid in the diagnosis and outcome of these injuries [19–22].
In conclusion, chondral injuries to the talo crural joint shows rather characteristic clinical symptoms and signs including subjective exertion pain, effusion and joint line tenderness on palpation and may occur with or without clinical signs of ligament laxity. In such cases, even with a "normal MRI" or an MRI showing subchondral oedema, an early arthroscopy should be considered. We suggest that the mechanism of some of these injuries in contact sports may involve direct impact to the ankle with or without sprain, which warrants further studies.