Rockwood Adults CH64

2850

SECTION FOUR • Lower Extremity

rule. 295,384 The proximal fibula fracture itself does not require fixation, indeed surgical exploration at this level carries a risk of damage to the common peroneal nerve. The level of the fibular fracture is a helpful but not infallible indicator of stability: Boden et al. 40 classically demonstrated that when cadaveric PER fractures, with associated deltoid rupture and syndesmotic injury, are created more than 3 to 4.5 cm prox- imal to the ankle, syndesmotic fixation was required to reestab- lish stability. However, this was based on the assumption that the syndesmosis (specifically the interosseous membrane) was torn up to the level of the fibular fracture. Nielson et al. 275 have subsequently demonstrated by MRI assessment of the syndes- mosis that this relationship is not constant. Indeed it has been found that even with a deltoid ligament tear fractures occurring up to 5 cm proximal to the ankle joint can be treated nonoper- atively with an equally satisfactory outcome both functionally and radiographically at 1 year. 183,291 Thus, although high fibular fractures are often unstable, the precise level of the fibular frac- ture cannot be used to judge the degree of syndesmotic injury. Measurements taken from plain radiographs (see Fig. 64-14) have been suggested to correlate with syndesmotic rupture: The most commonly used are a tibiofibular clear space (“syndesmo- sis A”) of greater than 5 mm and a tibiofibular overlap (“syn- desmosis B”) of less than 5 mm on the AP view or of less than 1 mm on the mortise view. 144 The tibiofibular clear space is the most reliable of these parameters. 144,304 Further assessment of these measurements within the normal population, however, has found a large variation in values with a mean tibiofibular clear space of 3.8 mm in females and 4.6 mm in males. 36,286 Dis- appointingly, cadaveric models have shown that no predictable increase in measurements on plain radiography can be found on sectioning of the syndesmotic ligaments, 37 and moreover clinical experience shows that some normal individuals have such a shallow incisura that they have no radiographic tibio- fibular overlap whatsoever. No consensus on an absolute value has been reached and as an alternative it has been suggested that an increase in the tibiofibular clear space in comparison to the contralateral ankle may be more accurate. Given the limita- tions of plain radiography, axial imaging has been evaluated and shown to demonstrate disruptions not evident on plain radio- graphs. 123 CT studies have shown wide variation in the shape of the syndesmosis with a deep concave incisura in some and a limited curve in others. 105 MRI has been shown to provide a more accurate assessment of syndesmotic injury 149,274 which correlates well with direct arthroscopic assessment. 279,369 MR arthrography may add further accuracy. 267 Such investigations, however, are expensive and are not commonly performed. Intraoperative stress testing with fluoroscopic assessment of the ankle joint can be performed in a variety of ways. Cotton 74 described applying a lateral force to the heel to displace the fib- ula laterally. A similar result can be obtained by directly pull- ing the fibula laterally with a hook (hook test) or a reduction clamp. Alternatively, an external rotation stress test can be per- formed. The various intraoperative stress tests correlate poorly with one another and have limited sensitivity when compared to a 7.5-Nm external rotation test performed with a standard- ized instrument. 290 As one would expect given the marked lim- itations of plain AP radiographs in assessing the syndesmosis

described above, and the uncertainty regarding the applicabil- ity of stress tests in distinguishing between SER 2 and SER 4 fractures, the rate of “positive” intraoperative tests also varies depending entirely upon the position of the foot and the crite- ria chosen. 148,169,290,291,353 Stoffel et al. 361 compared the hook and external rotation tests, reporting the most useful measurement of syndesmotic rupture to be the hook test. They found that the application of an external rotation force resulted in consistent increases only in the medial clear space but that this increase in medial clear space was also seen with deltoid ligament sectioning even when the syndesmosis was left undamaged. 361 Jenkinson et al. 169 reported that the standard external rotation test identified patients with syndesmotic injury not previously recognized on plain radiography. Lui et al. 224 found that arthroscopy diagnosed cases of syndesmotic rupture missed by stress radiography, but neither author sought to correlate their findings with functional instability. Finally Beumer et al. found in cadaveric experiments that all clinical tests resulted in radiographic “abnormalities” that could occur without any anatomic injury. 35,37 Which intraoper- ative test to use is therefore a matter of surgeon preference: It appears that the hook test is currently the most popular with 64% of orthopedic surgeons reporting it their first choice. 259 SER fractures are generally considered to be at low risk of syn- desmotic instability following stable fixation, because this type of injury is unlikely to have disrupted the interosseous ligament of the syndesmosis, or the interosseous membrane above. How- ever, instability has been reported by a number of authors: Stark et al. 353 reported that 93 of 238 (39%) patients with unstable SER 4 ankle fractures had instability on intraoperative stress testing, and postulated that this resulted from stretching or transection of the ligament. However, the limitations of these stress tests them- selves are gradually becoming clearer (see above) and the true level of syndesmotic instability after SER fractures remains uncer- tain. Certainly, there is clinical evidence that fixation of the syn- demosis in unstable SER 4 fractures does not improve outcome, implying that our ability to select those that need surgery is lim- ited. 192 Cadaveric attempts to investigate the importance of syn- desmosis injury should also be interpreted with caution as these have investigated the effect on pressure distribution following forceful subluxation of the talus. This experimental design may be flawed because clinically it is clear that a syndesmotic injury does not necessarily result in dynamic talar instability; as demon- strated by the stability seen in axial loading 186 and dynamic load- ing 251 cadaveric studies and clinical outcome studies. 183 Establishing a confident diagnosis of syndesmotic instability is of great practical importance because the decision to per- form surgical stabilization exposes the patient to additional risks. Fanter et al. 107 showed a risk of vascular injury in a cadaveric study and Kennedy et al. 183 compared patients with PER fractures treated with and without syndesmosis screws in a randomized trial and demonstrated a worse radiographic outcome among those treated with a screw. Malreduction (see Fig. 64-16) is widely considered to have a significant effect on outcome: In their influential paper, Sagi et al. reported that patients with syndesmotic malreduction had significantly worse functional outcomes than patients with an anatomical reduction after a minimum of 2 years. 326 However, this paper has been criticized in that the degree of malreduction was not