Lisfranc injuries: Patient- and physician-based functional outcomes

Transcription

1 International Orthopaedics (SICOT) (2003) 27: DOI /s ORIGINAL ARTICLE P.A. O Connor S. Yeap J. Noël G. Khayyat J.G. Kennedy S. Arivindan A.J. McGuinness Lisfranc injuries: Patient- and physician-based functional outcomes Accepted: 24 August 2002 / Published online: 11 December 2002 Springer-Verlag 2002 Abstract The purpose of this study was to assess functional outcome of patients with a Lisfranc fracture dislocation of the foot by applying validated patient- and physician-based scoring systems and to compare these outcome tools. Of 25 injuries sustained by 24 patients treated in our institution between January 1995 and June 2001, 16 were available for review with a mean followup period of 36 (10 74) months. Injuries were classified according to Myerson. Outcome instruments used were: (a) Medical Outcomes Study 36-Item Short Form Health Survey (SF-36), (b) Baltimore Painful Foot score (PFS) and (c) American Orthopedic Foot and Ankle Society (AOFAS) mid-foot scoring scale. Four patients had an excellent outcome on the PFS scale, seven were classified as good, three fair and two poor. There was a statistically significant correlation between the PFS and Role Physical (RP) element of the SF-36. Résumé Le but de cette étude était d étudier le résultat fonctionnel du traitement d une fracture luxation du Lisfranc en utilisant des scores validés Patient et Médecin et en les comparant. A partir de 24 malades traités entre janvier 1995 et juin 2001 pour 25 lésions, 16 malades étaient disponibles pour la révision avec une moyenne de suivi de 36 (10 74) mois. Les blessures ont été classées d après Myerson. Les échelles d appréciation utilisées étaient: (a) l échelle de qualité SF-36, (b) le score Pied douloureux de Baltimore (PFS) et, (c), le score de l American Orthopedic Foot and Ankle Society (AOFAS). Quatre malades avaient un excellent résultat P.A. O Connor S. Yeap J. Noël G. Khayyat J.G. Kennedy S. Arivindan A. J. McGuinness Department of Orthopaedic Surgery, Cork University Hospital, Wilton, Cork, Ireland P.A. O Connor ( ), Orthopaedic Registrar, 50 Fosterbrook, Stillorgan, County Dublin, Ireland paoconn@indigo.ie Tel.: , Fax: sur l échelle de Baltimore, sept ont été classés comme bon, trois comme moyen et deux mauvais. Il y avait une corrélation statistiquement significative entre le score de Baltimore et le score SF-36. Introduction Jacques Lisfranc ( ) was a field surgeon in Napoleon s army. He actually described amputation through the tarsometatarsal joint of the forefoot in cavalrymen, but it is customary to refer to injuries in this area of the foot as Lisfranc s fracture dislocations [17]. This injury is uncommon, with an incidence variously reported between one per 5,500 fractures per year [1], one per 60,000 population per year [8] or accounting for 0.2% of all fractures [5]. Diagnosis is aided by high-quality radiographs (AP, 30 oblique and lateral views), observing Stein s lines [18], ruling out any diastasis between the first and second ray and searching for a fleck sign [14]. Diagnosis is often overlooked, especially in the multiply injured patient with other distracting injuries [7, 20], which can significantly reduce outcome [19] with resultant long-term disability [13, 16]. A spontaneous partial or complete relocation is not unusual [4, 14] and may complicate accurate assessment. A high index of suspicion must be maintained for these injuries, and additional imaging, stress radiographs [2], weight-bearing radiographs [6], CT [11] or MRI [15] performed as indicated. Although some controversy exists about the optimum method of treatment [3] most authors now advocate anatomic reduction with internal fixation as the treatment method of choice [2, 14, 16]. This study reviews the outcome of these fractures treated by a number of different methods. In addition, it attempts to determine whether there is a correlation between traditional physician-based outcome and patientbased outcomes in order to establish a benchmark for future assessments.

2 99 Materials and methods Patients treated for a Lisfranc injury at our institution between January 1995 and June 2001 were identified from a search of the computerised hospital database in a level 1 trauma centre. A historical prospective study was performed and all case notes and radiographs were examined. Data was collected from patient charts under the following headings: age, gender, date of injury, side of injury, delay in diagnosis, occurrence at work, mechanism of injury, classification of injury, associated injuries, treatment method, method of wound closure and any resultant complications. Injuries were classified on reviewing the initial radiographs according to the scale proposed by Myerson [14]. Outcome instruments used were the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36), the Baltimore Painful Foot score (PFS) [14] and the American Orthopedic Foot and Ankle Society (AOFAS) mid-foot scoring scale [9]. Patients were then either recalled to the clinic or contacted by telephone. In the case of patients contacted by telephone, all had a recent clinic attendance with adequate documentation of gait pattern and foot alignment to merit inclusion in the study. Data was collected on standardised pro-forma. Analysis of the SF-36 data was performed using normative control values in the manual and interpretation guide as per Ware [21]. Statistical analysis of the scores obtained was performed using the Student s t-test and Minitab software (V. 13) (Pennsylvania, USA). Results There were 25 Lisfranc injuries in 24 patients during the study period; 16 patients were available for follow-up, and this group comprises the basis for this report. Reasons for unavailability were: overseas home address in two patients, untraceable from address at presentation (four patients), one patient had since died of an unrelated illness and one patient remained psychotic since the time of injury and hence was unable to complete the questionnaire. There are on average 2,800 admissions to our service per annum, giving an annual incidence of Lisfranc injuries of 0.014%. Injuries were treated by several different surgeons and methods of treatment varied among consultant orthopaedic surgeons and with injury type. As a general rule, closed reduction was attempted first. Age distribution of the 16 patients available for follow-up was from 16 to 80 years with a mean of 37 years. There were 12 male and four female patients. The left foot was involved in nine patients and the right in seven. One patient had a diagnosis that was missed initially but was treated within 2 weeks of the original injury. Five patients were injured in the workplace. Mechanism of injury was a crush injury in five patients, a fall from a height in seven, a motor vehicle accident in two, a motorbike accident in one and one occurred during a rugby match. Associated injuries were neurological in one patient, long bone fractures in three, other metatarsal fractures in eight and other tarsal injuries in eight. Isolated foot injuries occurred in 12 patients. Most patients had associated tarsal or metatarsal fractures with only five having neither. Twelve patients had primary wound closure performed and three were treated by secondary closure, of which two had split skin grafting performed. Fig. 1 AP radiograph shows type A Lisfranc fracture dislocation Fig. 2 AP radiograph shows type B2 Lisfranc fracture dislocation with disruption of the medial two tarso-metatarsal joints Only one patient had an open wound. Average length of follow-up was 36 (range 10 74) months. Injuries were classified according to Myerson, and we found 25% (n=4) had a type A injury, 13% (n=2) a type B1, 56%

3 100 Fig. 4 AP radiograph shows type C2 Lisfranc fracture dislocation Fig. 3 AP radiograph shows type B2 Lisfranc fracture dislocation with disruption of the lateral four tarso-metatarsal joints (n=9) a type B2 and 6% (n =1) a type C2. No patient was found to have a type C1 injury. Treatment methods were closed reduction and plaster cast application in 25% (n=4), closed reduction and K-wiring in 19% (n=3), open reduction and K-wiring in 37% (n=6) and open reduction and screw fixation in 19% (n=3) (Figs. 1, 2, 3, 4, 5, and 6). Immobilisation was maintained for 10 weeks 6 weeks non-weight bearing followed by 4 weeks partial weight bearing. K-wires were removed on average 6 weeks following surgery. One patient developed a compartment syndrome within 12 h of the initial injury, necessitating a standard triple incision fasciotomy. Following delayed wound closure with skin grafting, the final outcome was good. One patient treated by closed manipulation and K-wiring developed an infection around the K-wires requiring wire removal at 3 weeks. Immobilisation was maintained with a below-knee cast for 12 weeks with no loss of reduction. Eleven patients attended a review clinic and completed the outcome scores, while 5 were contacted by telephone and had a documented recent clinic attendance to allow completion of data collection. Two outcome tools are physician administered, whereas the patient indepen- Fig. 5 AP radiograph shows reduction of Lisfranc fracture dislocation and maintenance of reduction by multiple K-wires

4 101 The SF-36 data was compiled for all eight items [Physical Functioning (PF), Role Physical (RP), Bodily Pain (BP), Social Functioning (SF), General Health (GH), Vitality (VT), Role Emotional (RE), Mental Health (MH)] along with the Reported Health Transition (HT). Using a regression analysis, a statistically significant relationship (P <0.05) was found only with RP and the PFS and AOFAS scores, which were linear (Fig. 7). No statistical difference was found between SF-36 data and comparative normative data for an American population (Ware [21]). We found no statistically significant correlation between method of treatment and functional outcome scores, nor between injury type and functional outcome. Discussion Fig. 6 AP radiograph shows reduction of Lisfranc fracture dislocation and maintenance of reduction by cross screws Fig. 7 dently completes the SF-36. PFS scores had a mean of 76 (range 34 94); AOFAS scores had a mean of 81 (range ). Four patients had an excellent outcome on the PFS scale (90 100), seven were classified as good (75 90), three fair (60 75) and two poor (<60). Lisfranc injuries are rare injuries that are traditionally associated with a poor prognosis. It has been proposed that anatomic reduction and internal fixation can reduce the onset of painful arthrosis. There has been a transition in reporting outcomes from early subjective reports to the application of numerous outcome tools. This study evaluated the use of three different outcome tools in a series of complex foot trauma. We analysed the relationship between physician- and patient-administered questionnaires using a generic outcome evaluation tool the SF-36 which is the most widely applied such tool in North America [12]. In addition, two similar injuryspecific outcome tools with comparable published data were applied [10, 13, 14]. They werethe mid-foot scoring scale of the AOFAS and the PFS. Although we found no statistically significant relationship between method of treatment and outcome nor between injury type and outcome. This is not to say that there is no relationship between these factors but rather that with the small cohort the differences could not be interpreted statistically as being significant. This was also the finding of Richter [16]. Our PFS scores were similar to those of Mulier [13] and indicate the difficulty in obtaining uniformly good results from these complex foot injuries. Classically, Lisfranc injuries have been associated with high-energy trauma. However, we did not have the same number of severely injured patients as has been reported by other authors [2, 14], and three injuries were caused by seemingly minor falls. Even trivial trauma such as stepping from a curbstone can cause these injuries by an indirect mechanism [1, 2, 20], particular in the elderly [14]. We did find a statistically significant relationship between elements of the SF-36 RP, the PFS and AOFAS (mid-foot) scores. The authors feel that this validates the use of the SF-36 for questionnaire followup in complex foot and ankle trauma. In treating these injuries, accurate reduction has been emphasised by several authors [1, 2, 7, 8, 10, 13]. Open reduction and temporary internal fixation with screws has gained increasing acceptance as the treatment method of choice [2, 10, 14], with the role of the second

5 102 metatarsal frequently being the keystone in allowing reduction. Quality and maintenance of reduction of the initial reduction would appear to be the most important determinants of long-term acceptable results rather than the specific type of treatment [13, 14]. The injury pattern does not appear to influence final clinical results [14]. Despite anatomic reduction, a portion may develop posttraumatic degenerative symptoms, and this appears to be related to the degree of articular involvement [13]. Although a significant number of these injuries develop radiographic evidence of arthritis, no relationship between degenerative changes and functional end results has been established [13, 14]. Certainly, direct crush injuries fare badly in functional outcome studies, regardless of the pattern of injury produced [13, 14]. In addition, pure ligamentous injuries or minor subluxations can be associated with a poor prognosis [6]. In summary, this study examines a small cohort of patients with complex foot and ankle trauma and validates the use of the SF-36 as an outcome tool when compared with specific foot and ankle outcome tools. Although commonly overlooked, Lisfranc injuries can have a significant detrimental effect on long-term functional outcome that can be measured by validated generic and specific outcome instruments. References 1. Aitken AP, Poulson D (1963) Dislocations of the tarsometatarsal joint. J Bone Joint Surg [Am] 45: Arntz CT, Veith RG, Hansen ST (1988) Fractures and fracture dislocations of the tarsometatarsal joint. J Bone Joint Surg [Am] 70: Brunet JA and Wiley JJ (1987) The late results of tarso-metatarsal joint injuries. J Bone Joint Surg [Br] 69: Early JS, Bucholz RW (1996) Lisfranc injuries and their management Current Orthopaedics 10: English TA (1964) Dislocations of the metatarsal bone and adjacent toe. J Bone Joint Surg [Br] 46: Faciszewski T, Burks RT, Manaster BJ (1990) Subtle injuries of the Lisfranc joint. J Bone Joint Surg [Am] 72: Goosens M, DeStoop N (1982) Lisfranc s fracture-dislocations: Etiology, radiology, and results of treatment. Clin Orthop 176: Harcastle PH, Reschauer R, Kutscha-Lissberg E et al (1982) Injuries to the tarsometatarsal joint: Incidence, classification and treatment. J Bone Joint Surg [Br] 64: Kitaoka HB, Alexander IJ, Adelaar RS et al (1994) Clinical rating system for the ankle-hindfoot, midfoot, hallux and lesser toes. Foot Ankle Int 15: Kuo RS, Tejwani NC, Digiovanni CW et al (2000) Outcome after open reduction and internal fixation of Lisfranc joint injuries. J Bone Joint Surg [Am] 82: Leenen LPH, van der Werken C. (1992) Fracture-dislocations of the tarsometatarsal joint, a combined anatomical and computed tomographic study. Injury 23: Martin DP, Engelberg R, Agel J et al (1997) Comparison of the Musculoskeletal Functional Assessment questionnaire with the Short Form-36, the Western Ontario and McMaster Universities Osteoarthritis Index, and the Sickness Impact Profile Health- Status Measures. J Bone Joint Surg [Am] 79: Mulier T, Reynders P, Sioen W et al (1997) The treatment of Lisfranc injuries. Acta Orthop Belg 63: Myerson MS, Fisher RT, Burgess AR et al (1986) Fracture dislocations of the tarsometatarsal joints: End results correlated with pathology and treatment. Foot Ankle Int 6: Potter HG, Deland JT, Gusmer PB et al (1998) Magnetic resonance imaging of the Lisfranc ligament of the foot. Foot Ankle Int 19: Richter M, Wippermann B, Krettek C et al (2001) Fractures and fracture dislocations of the midfoot: Occurrence, causes and long-term results. Foot Ankle Int 22: Rosenberg GA, Patterson BM (1995) Tarsometatarsal (Lisfranc s) fracture-dislocations. Am J Orthop [Suppl] l: Stein RE (1983) Radiological aspects of the tarsometarsal joints. Foot Ankle 3: Turchin DC, Schemitsch EH, McKee DM et al (1999) Do foot injuries significantly affect the functional outcome of multiply injured patients? J Orthop Trauma 13: Vuori JP and Aro HT (1993) Lisfranc joint injuries: trauma mechanisms and associated injuries. J Trauma 35: Ware JE, Snow KK, Kosinki M. (2000) SF-36 Health survey: Manual and interpretation guide. Quality Metric Incorporated, Lincoln