2315124код для вставки
Clinical Anatomy E76-79 (1994) Variation in Tibia1 Torsion D.G. ECKHOFF, R.C. KRAMER, J.J. WATKINS, B.J. BURKE, C.A. ALONGI, E.R. STAMM, AND D.P. VAN GERVEN Departments of Orthopaedics (D.G.E., R.C.K., J.J. W., B.J.R., Radiology (E.R.S., C.A.A.) and Univenity of Colorado, Denver, Colorado and Department of Anthropology, University of Colorado, Roulder, Colorado (0.P. VanG.) A skeletal collection from the African country of Sudan was examined to determine if torsion of the tibia in an African population varies from established norms for European, Asian, and American populations. A large variation in tibial torsion was observed. There was a distinct limb asymmetry with more torsion on the right than left. No associationof tibial torsion with gender was identified. A small but appreciable increase in average torsion (7") was identified when this population was compared with previously reported populations. This study documents a variation in tibial torsion not appreciated in earlier studies. o 1994 wiley-Liss, Inc. Key words: African, black, torsion, rotation, leg, limb INTRODUCTION Rotation of the tibia (torsion) begins in utero and progresses throughout childhood and adolescence to skeletal maturity (Staheli and Engel, 1972). Average tibial torsion has been documented in adult Caucasian and Oriental populations between 20" and 30" (Le Damany, 1909; Elftman, 1945; Hutter and Scott, 1949; Elgeti et al., 1980; Jakob et al., 1980; Turner and Smillie, 1981; Takai et al., 1985; Yagi and Sasaki, 1986; Clementz, 1989; Yoshioka et al., 1989). No black skeletal or patient population has been compared with these studies to determine whether a variation in tibial torsion exists, despite the fact that certain morphologic characteristics of the human limb demonstrate racial variation. For instance, it is well documented that the average tibia is longer and narrower in black as compared with Caucasian populations (Dupertuis and Haddon, 1951; Trotter and Gleser, 1958; Farrally and Moore, 1965). Several studies (Le Damany, 1909; Hutter and Scott, 1949; Clementz, 1989) have addressed the difference in tibial torsion between right and left limbs, while only one study has identified variation in tibial torsion based on gender (Yoshioka et al., 1989). These characteristics have not been addressed in a black population. 0 1994 Wiley-Liss, Inc. This study was undertaken to determine if a black population from Africa demonstrates tibial torsion comparable to previously reported American, European, and Japanese populations. T h e study was also designed to address issues of variation based on gender and limb asymmetry. MATERIALS AND METHODS One hundred and twelve well-preserved skeletons (224 matched limbs) were examined by goniometry and computed tomography (CT) scan to determine the torsion present in both right and left tibiae. These skeletons are part of an anthropologic collection from Africa (Sudan) maintained in the Department of Anthropology at the University of Colorado. T h e anthropologic characteristics of this population have been well documented and reported (Van Gerven et al., 1981). T h e age and gender of each specimen were determined by an experienced anthropologist (D.P.V.) according to recognized anthropologic techniques and Received for publication March 8, 1993; revised September 3, 1993. Address reprint requests to D.G. Eckhoff, Department of Orthopaedics, University Pavilion, 4701 E. Ninth Avenue-Box E203, Denver, CO 80262. Variation in Tibial Torsion 77 established practice (Ubelaker, 1978). Gender was es- two most posterior points of the plateau, and it was tablished by measurements of pelvic dimensions with determined in this study by the surface upon which 51 males and 61 females. Bone age was established by both condyles of the plateau rested during measureanalysis of the symphysis pubis. All specimens were ment. T h e trans-tibia1 axis of the distal tibia (Fig. 1, skeletally mature, with an average age of 39 (range, line D) was defined by the line connecting the distal tip 21-51) for females and 36 (range, 20-52) for males. of the medial malleolus and the midpoint of the lateral Seven right and nine left specimens demonstrating border (fibular sulcus). It was determined by direct evidence of fracture or traumatic deformity were omitted. visual inspection of the bone or by inspection of approTorsion was defined as the twist of the bone about its priate cross-sectional images on the C T scan. Direct goniometric measurements were performed long axis. Torsion was measured in degrees as the angle between the posterior axis of the proximal plateau (Fig. by resting one arm of the goniometer on the supporting 1, line A) and the trans-tibia1 axis of the ankle (Fig. 1, surface (parallel to the posterior condyles) while the line D). T h e measured angle of tibial torsion is positive other arm was aligned visually to the distal trans-tibia1 when the distal axis is externally rotated with respect to axis. T h e angle observed between the arms of the the proximal axis. T h e posterior axis (Fig. 1, line A) of goniometer equalled the recorded angle of tibial torthe proximal plateau was defined as the line joining the sion. This same angle was obtained by CT using the console computer of the C T scanner. T h e posterior condylar axis and the distal-tibia1 axis were visually identified on the scan and traced with the computer cursor, following which the console provided the angle between these two axes. This choice of methods and definition of landmarks are discussed below. Goniometric and CT measurements were each performed by two independent observers and reported at two different times. Analysis of the data was performed using a standard paired t-test. RESULTS ;..i / Analysis of the goniometric data revealed the following. T h e average amount of torsion in right tibiae was 38 ? 11". T h e average amount of torsion in left tibiae was 33 2 9". T h e combined average tibial torsion by goniometer was 35 k 9" (Table 1). Analysis of the CT scan data revealed the following. T h e average amount of torsion in right tibiae was 40 5 9". T h e average amount of torsion in left tibiae was 32 ? 10". T h e combined average tibial torsion by C T scan was 36 k 9" (Table 1). Statistical analysis was performed using a paired Student's t-test. This test demonstrated that there was a significant difference between right and left limbs ( P < TABLE 1. Data by observer and method Observer Fig. 1. Tibial torsion is measured as the angle between axes at the top and bottom of the tibia as illustrated by lines A and D at center right. Proximal and distal axes of the tibia referenced in this study and the literature are the following: A, posterior condylar axis of the tibia connecting the two most posterior points of the condyles; B, transcondylar axis through the midpoint of each condyle; C , anterior condylar axis joining the anterior margins of the medial and lateral articular surfaces of the plateau; D, bimalleolar axis joining the distal tip of the tibial malleolus with the midpoint of the lateral border (fibular sulcus). Method of Measure Goniometric Right leg Left leg Combined CT scan Right leg Left leg Combined 1 2 Average 37 30 34 39 34 36 38 33 35 40 32 36 40 32 36 40 32 36 78 Eclchoff et al. 0.05) by both goniometric and C T scan methods. condylar axis is variably defined by a line joining the There was no significant difference between measure- mid-points of the medial and lateral plateau (line B, ments made by the two techniques. There was no Fig. 1) or a line passing through the most anterior significant difference between observations of the points on the articular surface of the medial and lateral same or different individuals using the C T method. A plateaus (line C, Fig. 1). This axis has also been designificant difference in goniometric measure of the fined (Hutter and Scott, 1949) as the perpendicular left tibiae was found between the two observers ( P < line to the medial border of the foot. Comparison of the present study to previous studies 0.001); however, the average difference between the two observers was only 2". No significant difference can be made based on the proximal axis selected for existed on the right between observers with the gonio- measurement. There is one reported cadaver study metric technique. Regression analysis demonstrated (Elftman, 1945) that referenced the posterior tibial that there was no influence of age or sex on the amount condyles to define the proximal tibial axis. T h e posterior femoral condyles were measured in this study asof torsion present. suming they parallel the posterior condyles of the tibia. This is a valid assumption, as the posterior condylar DISCUSSION axis and the posterior tibial axis have been shown to be Torsion of the tibia has been reported on European, parallel in the non-arthritic knee (Eckhoff et al., 1993). Asian, and American populations, with averages rang- Comparison of Elftman's average 28" tibial torsion with ing between 19" and 30" (Table 2). T h e variation of the average 35" presented here reveals 7" more tibial torsion between these studies can be attributed to the torsion on average in this African population when meamethod of measure and choice of landmarks. In studies sured by goniometer (Table 3). There are three patient surveys that have defined reporting rotation of the tibia in the absence of the the proximal axis of the tibia as the posterior tibial fibula, e.g., cadaver studies, the distal axis is detercondyles on C T scan (Elgeti et al., 1980; Jakob et al., mined only by the morphology of tibia. By contrast, the 1980; Takai et al., 1985). These studies demonstrate studies based on patient exam, C T scans, or fluoroscopic images include the fibula. T h e distal axis referenced to the fibula is externally rotated 5" to 6" TABLE 3. Literature torsion vs. this study when compared with the axis determined by the tibia Study Average alone. Material (method) (torsion in degrees) difference Variation in the measure of tibial torsion also occurs Cadaver (7) Elftman (28) as a consequence of the proximal axis of the tibia se(goniometer) Observer avg this study (35) lected. T h e principle choices for the proximal refer- Pa tien t Elgeti (29) (7) ence axis are the posterior condylar axis and the trans(CT scan) Jakob (30) Takai (28) condylar axis. T h e posterior condylar axis is defined by Average (29) the line joining the two most posterior points on the Observer avg this study (36) medial and lateral plateau (line A, Fig. 1). T h e transTABLE 2. Comparison of tibial torsion studies' External tibial torsion (degrees) 0 Study (population) 10 Clement2 (Caucasian) Elftman (Caucasian) Elgeti (Caucasian) Hutter (Caucasian) Jakob (Caucasian) La Damany (Caucasian) Takai (Oriental Turner (Caucasian) Yagi (Oriental) Yoshioka (Caucasian) 40 + X * * + X + * This study (African) Goniometric method C T scan method "Combined (R&L); 30 20 * * * * + + X, right; + left. * X * x 50 Variation in Tibial Torsion 29, 30, and 28" of tibial torsion, respectively, for a combined average of 29" of tibial torsion. By comparison, this study of an African population reports 36" average tibial torsion or 7" more tibial torsion using CT scan as the method of measurement (Table 3). T h e average of left and right data (combined data) was used for comparison since the three studies being compared do not report left and right limbs separately. There are three previous cadaver studies that identify the transcondylar axis as the proximal reference (La Damany, 1909; Hutter and Scott, 1949; Yoshioka et al., 1989). These three studies demonstrate a combined average of 22" of tibial torsion. There is one previous patient study that references the transcondylar axis on C T scan (Yagi and Sasaki, 1986) to identify 24" of tibial torsion. Comparison between these studies and the African population reported here is difficult due to the disparate nature of the axes-transcondylar versus posterior condylar. T h e transcondylar axis was not measured in the present study because it is hard to define, has limited clinical relevance, and was not reproducible in our hands. Torsional variation between left and right tibia has been previously reported, with right tibial torsion consistently greater than the left (Le Damany, 1909; Hutter and Scott, 1949; Clementz, 1989). 'I'he data reported in this study also demonstrate a consistent difference between the right and left tibial torsion, with the right significantly greater than the left. Torsional variation between males and females has been documented only once (Yoshioka et al., 1989), with males averaging 21" and females averaging 27". This variation in tibial torsion based on gender was not observed in the African population reported here. All of the studies previously reported document a large variation of tibial torsion within the populations studied. This study confirms the wide variability of tibial torsion between individuals as demonstrated in the large standard deviations. Clinically, this variation may be very significant. For instance, it will make the application of alignment guides, such as those keyed to rotational landmarks of the tibia in total knee arthroplasty, difficult to apply without individual modification. As additional examples, correction of traumatic malunion or congenital maltorsion of the tibia may not be individually accomplished by simply recreating the average torsion for the population reported in these studies. In this or any other clinical application of these 79 studies, individual variation with respect to tibial torsion should be addressed, given the variability reported. REFERENCES Clementz, B.G. 1989 Assessment of tibial torsion and rotational deformity with a new fluoroscopic technique. Clin. Orthop. 245:199-209. Dupertuis, C.W. and J.A. Haddon 1951 On the reconstruction of stature from long bones. Am. J. Phys. Anthropol. 9: 15-53. Eckhoff, D.G., E.R. Stamm, and R.F. Kilcoyne 1993 Femoral and tibial rotational morphometry in osteoarthritis. Orthop. Trans. Elftrnan, H. 1945 Torsion of the lower extremity. Am. J. Phys. Anthropol. 3:255-265. Elgeti, H., R. Grote,and G. Giebel 1980 Bestimmung der Tibiatorsion mit der axialen Computertomographie. 2. Unfallheilk. Traumatol. 83:14-19. Farrally, M.R. and W.J. Moore 1975 Anatomical differences in the femur and tibia between Negroids and Caucasoids and their effects upon locomotion. Am. J. Phys. Anthropol. 43: 63-70. Hutter, C.G. and W. Scott 1949 Tibial torsion. J. Bone Joint Surg. 32-A:511-518. Jakob, R.P., M. Haertel, and E. Stussi 1980 Tibial torsion calculated by computerized tomography and compared to other methods of measurement. J. Bone Joint Surg. 62-B: 238-242. Le Damany, P. 1909 La torsion due tibia, normale pathologique, experimentale. J. Anat. Physiol. 45598-615. Staheli, L.T. and G.M. Engel 1972 T h e natural history of torsion and other factors influencing gait in childhood. Clin Orthop. 86:183-186. Takai, S., K. Sakakida, F. Tamashita, F. Sum, and F. Izuta 1985 Rotational alignment of the lower limb in osteoarthritis of the knee. Int. Orthop. (SICOT) 9209-216. Trotter, M. and G.C. Gleser 1958 A re-evaluation of estimation of stature taken during life and of long bones after death. Am. J. Phys. Anthrop. 16:79-123. Turner, M.S. and I.S. Smillie 1981T h e effect of tibial torsion on the pathology of the knee. J Bone Joint Surg. 638:396-398. Ubelaker, D. H. 1978 Human Skeletal Remains-Excavation, Analysis, Interpretation. Aldine Publishing Co., pp. 42-43, 53-55. VanGerven, D.P., M.K. Sanford, and J.R. Hummert 1981 Mortality and culture change in Nubia Batn el Hajar. J. Hum. EvoI. 20:395-408. Yagi, T. and T. Sasaki 1986 Tibial torsion in patients with medial-type osteoarthritic knee. Clin Orthop. 223:177-182. Yoshioka, Y., D.W. Siu, R.A. Scudamore, and T.D.V. Cooke 1989 Tibial anatomy and functional axes. J. Orthop. Res. 7: 132- 137.