The lisfranc push-up stress-test

G. Smith, C. Loizou

Introduction: The diagnosis of Lisfranc ligament disruption is notoriously difficult. Radiographs and MRI scans are often ambiguous therefore a stress-test examination under anaesthesia is commonly required. Two midfoot stress-tests are in current practice, namely the varus first ray stress-test and the pronation abduction test. The optimal type of stress-test is not however evaluated in the literature. We hypothesised that after the loss of the main plantar stabiliser (the Lisfranc ligament) the patient would demonstrate dorsal instability, not the classic 1^st/2^nd metatarsal diastasis commonly described. We therefore devised a push-up test (placement of a force under the 2^nd metatarsal in an attempt to elevate the base away from the middle cuneiform on the lateral radiograph). We aimed to initially test our hypothesis on a cadaveric model.

Methods: Twelve fresh frozen cadaveric specimens without previous foot injury were used. The 2^nd tarsometatarsal joint was exposed and the Lisfranc ligament and dorsal capsule were incised. An image intensifier was positioned and standard anteroposterior (AP) and lateral views were obtained. Two previously reported AP stress-tests (varus first ray stress test, pronation abduction test) and the novel test under investigation ('Lisfranc Push-Up' test) were duly performed. Images were obtained once the investigator felt the appropriate views were achieved.

Results: All twelve of the Lisfranc Push-Up tests showed dorsal subluxation of the 2^nd metatarsal on the middle cuneiform of greater than 2mm on the lateral radiograph. No diastasis of the 1^st/2^nd metatarsals was seen in any of the specimens on the AP radiograph for either of the other two stress-tests.

Conclusion: The authors have described a novel way of demonstrating the dorsal instability associated with the ligamentous Lisfranc injury. Our results support the Lisfranc Push-Up test as a reproducible and sensitive method for assessing ligamentous Lisfranc injuries. In our cadaveric model the previously described stress-tests do not work.