Kinematics of the knee joint in deep flexion: a radiographic assessment

Medical Engineering & Physics 20 (1998) 302–307

Communication

Kinematics of the knee joint in deep flexion: a radiographic assessment Mohamed Samir Hefzy *, Brian P. Kelly, T. Derek V. Cooke Biomechanics Laboratory, Departments of Biological and Medical Research and Orthopaedic Surgery, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia Received 29 July 1997; accepted 5 March 1998

Abstract The purpose of this study is to describe the kinematics of normal knees in vivo, assessed in deep flexion, using bi-planar radiographs. Antero-posterior and lateral views were obtained from five healthy males during three sequential positions of kneeling. In the first position, the subject knelt with the knees fully flexed (deep flexion between 150 and 165°) and torso upright. In the second position, the subject bowed forward to an intermediate position (about 120° of knee flexion). In the third position, the subject bowed further until his head touched the floor, supporting the upper torso with hands and with the knees flexed at about 90°. The results show that past 135° of knee flexion, the patella cleared the femoral groove and was in contact only with the condyles. For these particular postures, and during deep flexion, motion of the femur on the tibia did not reveal the classical femoral ‘roll back’. Rather the lateral femoral condyle rolled further over the postero medial aspect of the lateral tibial plateau while contact of the medial femoral condyle occurred more anteriorly, but still in the posterior part of the medial plateau. This asymmetric rolling motion indicated an element of internal tibial rotation. Furthermore, the tibia was found to articulate with the femur at the most proximal points of the condyles in deep flexion. These data on the kinematics and contact characteristics of the tibio-femoral joint must be considered in any approach to design for a Deep Flexion Knee Implant.  1998 IPEM. Published by Elsevier Science Ltd. All rights reserved. Keywords: Deep knee flexion; Total knee replacement; Radiographs; Knee kinematics; Osteoarthritis

1. Introduction Osteoarthritis (OA) is one of the most commonly encountered forms of arthritis associated with the progressive loss of articular cartilage that may eventuate in exposure of subchondral bone. It is usually focally seen and associated with reactive changes in the joint, osteophytosis and bone damage. Total Knee Replacement (TKR) is a widely used strategy in surgical treatment of the knee when severely damaged by arthritis (OA or rheumatoid arthritis) [1,2]. However, the results of this procedure vary considerably among patients [3]. In North America and Europe, the TKR surgical intervention has proven successful in relieving pain and in allowing a return to daily living activities for most * Corresponding author. The Biomechanics Laboratory, Department of Mechanical, Industrial and Manufacturing Engineering, The University of Toledo, Toledo, OH 43606, USA. Tel.: + 1-419-530-8234; Fax: + 1-419-530-8206; e-mail: [email protected]

patients [4,5]. For instance, in 1991 alone, it has been reported that more than 177 000 TKRs have been performed in North America [6]. This number of surgeries is also increasing annually. While highly successful and widely used in North America, Europe and Australia, TKR is far less well accepted in the Middle and Far East. Very few knee arthoplasties are performed in Saudi Arabia and elsewhere in the Middle East compared with the Western World. A major factor that contributes greatly to the resistance to TKRs in this region of the world is the quality of the outcome of this surgical procedure which is affected by several factors that are implant-related, surgeon-related and patient-related [7]. Patient-related factors that affect the quality of a prosthesis such as physical condition, age and activity level differ greatly across populations. Almost all of the TKRs designed in the United States, Europe and Japan allow for a range of motion up to about 120° of knee flexion. This is satisfactory to the Western population as it accommodates

1350-4533/98/$19.00  1998 IPEM. Published by Elsevier Science Ltd. All rights reserved. PII: S 1 3 5 0 - 4 5 3 3 ( 9 8 ) 0 0 0 2 4 - 1

M.S. Hefzy et al. / Medical Engineering & Physics 20 (1998) 302–307

the range of motion needed for most of their daily activities. However, highly geographical variations exist in the normal range of joint motions. For instance, it has been reported that Saudi males have a difference of more than 15° in knee flexion compared with Scandinavian people [8]. Most of the population in the Middle East routinely deeply flex the knee up to 165°. During praying, most of the Moslem people kneel with the limbs fully flexed (between 150 and 165°) and torso upright with the heel reaching the posterior surface of the upper thigh as shown in Fig. 1(a). Since most of the presently available commercial TKRs are not designed to flex more than 120°, they will not meet the needs of patients in the Middle East. Also, hyperflexion with existing implants beyond their design configuration may provide subluxation and dislocation. A potential solution to this problem is the development of a TKR implant that would provide for full flexion of the knee. It has been reported that a total knee arthroplasty system was developed in Japan [9] that allows for flexion up to 165°. Yet, maintaining a good functional stability remains an issue when using such a system. This is mainly caused by the present lack of knowledge of normal knee mechanics past 140° of knee flexion. The literature review reveals that few studies have investigated the kinematics of the knee joint in Deep Flexion [10]. As a result, our understanding of the complex three-dimensional dynamic tibio-femoral and patello-femoral motions past 140° of knee flexion is limited. The purpose of this study was to determine the differential motions of femur, tibia, and patella, as well as resultant positional aspects of medial, lateral and patellofemoral compartments in deep flexion, during in vivo activities of prayer using radiography applied to healthy adults of Arabic descent.

303

Subjects were fully informed and consented to having radiographic images taken of the knee during three sequential positions of kneeling for prayer. In the first position, the subject knelt with the knees fully flexed and torso upright as shown in Fig. 1(a). In the second position, the subject bowed forward to an intermediate position (about 120° of knee flexion). In the third position, the subject bowed further until his head touched the floor, supporting the upper torso with the hands and with a knee flexion close to a right angle as shown in Fig. 1(b). Bi-planar radiographic images were taken using a custom designed wooden platform to support the subjects as shown in Fig. 2. A central slot in the platform allowed a vertical cassette to be placed between the legs to obtain lateral views of the knee. The area of the platform directly supporting the knee was made of plexiglass, and was designed to hold a cassette placed horizontally beneath it to obtain antero-posterior (A-P) views. The subject did not move between taking A-P and lateral views and was instructed not to change position of the feet in relation to the buttocks or rotate the torso away from the flexion plane while changing positions. With this set-up, A-P views taken at right angle to the platform were available only at the initial and intermediate positions (full flexion and about 120° of flexion, respectively). These A-P views were non-standard images. Tibial, femoral and patellar axes were identified on the radiographs as shown in Figs 3 and 4. Both tibial and femoral axes were parallel to the tibial and femoral shafts, respectively. The patellar axis was defined as the line joining the midpoint of the attachment of the patellar tendon distally, to the midpoint of the attachment of the

2. Methods Five healthy male subjects with an average age of 31.6 years and of Arabic descent participated in this study.

Fig. 1. Praying positions. (a) Kneeling in deep fluxion; (b) bowing position.

Fig. 2. Experimental set-up. This figure illustrates the custom-made cassette holder which allowed views to be obtained at 90° to the platform.

304

Fig. 3.

M.S. Hefzy et al. / Medical Engineering & Physics 20 (1998) 302–307

A representative set of lateral views in deep flexion, at the intermediate position and in the bowing position. This set is for subject 2.

Fig. 4.

Antero-posterior views in deep flexion for subjects 1–5.

quadriceps tendon proximally. The angle between tibial and femoral axes in the lateral view differs from the true knee flexion angle as demonstrated by the fact that these two lines do not overlap in the A-P view. Therefore, the knee flexion angles in positions 1 and 2 were calculated by writing a computer program that combines the measured angles between the tibial and femoral axes in both lateral and A/P views at a given position. In position 3, the flexion angle was measured directly from the lateral view because an A/P view could not be obtained at this position.

3. Results Fig. 3 displays a representative set of lateral views of the three positions; this set was obtained from subject 2. Radiographs were taken for the right knee in all subjects, except for subject 4 where X-rays were obtained for the left knee (subject 4 has had previous problems with his right knee). Because of technical difficulties, a lateral view at the intermediate position was not obtained for subject 1. Fig. 4 shows the A-P views in deep flexion for each of the subjects. The tibial and femoral axes in

M.S. Hefzy et al. / Medical Engineering & Physics 20 (1998) 302–307

the A-P view obtained from subject 1 overlapped because this subject held a symmetric position, placing the ankles directly underneath the buttocks. The other subjects placed their ankles laterally to their buttocks (which may produce some valgus at the knee joint) which caused the tibial and femoral axes to not overlap in the A-P views. Fig. 4 also shows that in deep flexion, and for all five subjects, the lateral joint space appears smaller than the medial. The lateral views show that on the femur, contact of the tibia appeared to occur on the most proximal points of the posterior condyles when the knee is in deep flexion. As subjects moved from the kneeling to the bowing positions, these contact points moved progressively distally on the posterior femoral condyles. For the tibia, the lateral views show that the medial contact appeared to occur in the posterior half of the tibial plateau, whereas the lateral contact was even more posterior, occurring near the posterior edge of the tibia. The lateral views also show that in deep flexion, the patella was very flexed with respect to the tibia. These figures also show that at this position, the patella was completely clear of the femoral groove and in contact with the femoral condyles only. Table 1 lists the values of the angles describing the position of the patella with respect to the femur and tibia at the three tested positions for all five subjects. In deep flexion, and at position 1, knee flexion angle averaged 157.3 ± 4.9°. At this position, the angle between patellar axis and tibial axis (angle ␣) averaged 62.7 ± 3.3°, while patellar–femoral flexion (angle ␤ between patellar axis and femoral axis) averaged 101.2 ± 3.2°. At position 3, the knee flexion angle averaged 102.6 ± 9.9°, and angles ␣ and ␤ averaged 30.5 ± 4° and 72.1 ± 6°, respectively. At the intermediate position, where knee flexion angle averaged 138.5 ± 6.6°, angles ␣ and ␤ averaged 52.4 ± 4.2° and 88.3 ± 6.3°, respectively.

305

4. Discussion and conclusions Bi-planar radiographs of the knee (A-P and lateral views) obtained at 90° to each other were taken for knees in subjects during a prayer movement at three flexion positions. The A-P view was non standard; yet clear and identifiable outlines of the femur, patella and tibia were seen, enabling some depiction of anatomic axial orientation of each bone at maximum flexion. This correlated with images seen on the lateral X-rays of these knees in which rotation between femur and tibia was observed. The results indicate an asymmetric rolling motion to occur in deep flexion in a pattern that is not typical of femoral ‘roll back’. In deep flexion, the contact points of the lateral compartments of the tibio-femoral joints were more posterior than those of the medial compartment; all contact points being located in the posterior aspect of the tibial plateaus. These data indicate a varied amount of internal tibial rotation, suggesting that in deep flexion there is a lesser amount of femoral ‘roll-back’ of the medial compartment on the tibia than that of the lateral compartment. This finding (the lateral femoral condyle rolled posteriorly more than the medial, to the extent that the contact in deep flexion came onto the edge of the bone) may reflect the particular postures chosen by the subjects, and different findings might be obtained if the subjects squatted down to full knee flexion, with their weight on their feet. Nevertheless, our results are in agreement with the limited data available in the literature [10,11]. Cornwall et al. [10] using MRIs, radiographs and fluoroscopy of volunteer subjects at knee flexion angles larger than 140°, reported that the posterior cruciate ligament plays a role in providing contact in the lateral tibiofemoral joint that differs largely from that seen in the flexion range of 0–140°. Dye et al. [11] employed high speed cine computed tomography to visualize isolated com-

Table 1 Patellar flexion angles with respect to femur and tibia at the three tested positions for all five subjects Subject 1 (45 years)

Subject 2 (30 years)

Subject 3 (29 years)

Subject 4 (26 years)

Subject 5 (28 years)

Position Position Position Position Position Position Position Position Position Position Position Position Position Position Position 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 Knee flexion 160.5 angle (deg.) Angle alpha1 61 (deg.) Angle 99 beta2 (deg.) 1

93.5

155.6

134.2

97

154.8

141.5

100.5

163.9

132

119

151.6

146.3

103

26.5

58

47

27.5

65.5

54

30

63

51.5

36.5

66

57

32

67

105

95.5

69.5

103

87

70.5

102

80.5

82.5

97

90

71

Angle alpha is the angle between patellar axis and tibial axis. Angle beta is the angle between patellar axis and femoral axis (patello-femoral angle).

2

306

M.S. Hefzy et al. / Medical Engineering & Physics 20 (1998) 302–307

partments of undissected cadaveric knees during a full and continuous range of motion. They reported that the lateral compartment of the tibio-femoral joint moved more posteriorly in flexion than the medial compartment. In addition, our data show that the patella flexion angle (with respect to the femur) always lags knee flexion and reaches average values of 72.1°, 88.3° and 101.2° when the knee is flexed on average to 102.6°, 138.5° and 157.3°, respectively. These data are in agreement with those of Van Eijden et al. and Hefzy et al. [12,13] who reported that the patellar flexion angle increases with knee flexion to reach values of 65° and 90° when the knee is flexed to 90° and 120°, respectively. These data suggest that the rate of increase in patellar flexion angle (with respect to the femur) decreases past 120° of knee flexion. This is in agreement with the data reported by Van Eijden et al. [12] as shown in Fig. 5. Our data also show that the patellar flexion angle with respect to the tibia reaches average values of 30.5°, 52.4° and 62.7° (for the three respective positions). These data suggest that past 138.5° of knee flexion, and in deep flexion, the patella is highly flexed with respect to the tibia. These results provide considerable further information on the kinematics and contact characteristics of the patello-femoral and tibio-femoral joints in deep flexion. More importantly, our data indicate that in this position, contact of the tibia on the femur is asymmetric posteriorly on the tibia and occurs on the most proximal

points of the posterior condyles. These findings help to explain the limitations of existing TKRs. Furthermore, provocative data have been reported describing the damage pattern of OA in the Middle East which is different than that of other parts of the world. Li et al. [14] reported that most of their Saudi OA patients did not show features of severe articular cartilage degeneration at the patello-femoral joint (PFJ). On the other hand, the anatomical pattern of gonarthrosis in Saudi Arabia was compared with that in Sweden with samples coming from tertiary health care units in urban areas in each country [15]. The same definition of OA was used in the data analysis of both populations and a higher percentage of PFJ involvement of 80% was found in the Saudi patients compared with 48% of the Swedish. In England, McAlindon et al. [16] reported a smaller percentage of PFJ OA. Yet, their study was the first to show prevalence of PFJ OA in the English community. Consequently, they concluded that OA studies can no longer neglect the PFJ. Our data show that in deep knee flexion, the patella is in contact only with the condyles, thus suggesting a possible reduction in patello-femoral contact area and a subsequent increase in contact stresses. This is in agreement with the data reported in the literature indicating that a large increase in the patello-femoral contact stresses occurs during squatting [17] when the knee is highly bent, and might explain the observed involvement of the patello-femoral joint in Saudi OA patients [15] whose daily living activities include sitting cross-legged, squatting and kneeling in deep flexion.

Acknowledgements The authors wish to acknowledge the contribution of Abdel Mohsen Al-Baddah, Department of Mechanical Engineering at King Saud University, in writing the computer code used in the analysis of the bi-planar Xrays. Also, the authors acknowledge the assistance of Laurie Harrison, Biomechanics Laboratory, and Anita Lee, Biomedical Physics and Medical Imaging for their assistance in obtaining the radiographic views.

References

Fig. 5. Patellar flexion angles (averages and standard deviations) with respect to the femur and tibia in deep knee flexion: + , Hefzy’s data [13]; × , Van Eijden’s data [12].

[1] Buckwalter JA, Lohmander S. Current concepts review: operative treatment of osteoarthrosis. Current practice and future development. JBJS 1994;76–A(9):1405–18. [2] Gill T, Schemitsch EH, Brick GW, Thornhill TS. Revision total knee arthroplasty after failed unicompartmental knee arthroplasty or high tibial osteotomy. Clin Orthop Rel Res 1995;321:10–18. [3] Mankin HJ, Buckwalter JA. Editorial. Restoration of the osteoarthritic joint. JBJS 1996;78–A(1):1–2. [4] Stern SH, Singer LB, Weissman SE. Analysis of hospital cost in total knee arthroplasty. Clin Orthop Rel Res 1995;321:36–44.

M.S. Hefzy et al. / Medical Engineering & Physics 20 (1998) 302–307

[5] Insall JN. Presidential address to the knee society: choices and compromises in total knee arthroplasty. Clin Orthop Rel Res 1988;226:43–8. [6] National Hospital Discharge Survey. National Center for Health Statistics, 1991. [7] Faro LMC, Huiskes R. Quality Assurance of Joint Replacement. Legal Regulation and Medical Judgement. Acta Orthop Scand Suppl 1992;250(63):1–33. [8] Ahlberg A, Moussa M, Al-Nahdi M. On geographical variations in the normal range of joint motion. Clin Orthop Rel Res 1988;234:229–31. [9] Shoji H, Yoshino S, Komagamine M. Improved range of motion with the Y/S total knee arthroplasty system. Clin Orthop Rel Res 1987;218:150–63. [10] Cornwall B, Okuno M, Reid G, Forrest W, Lyons W. Biomechanics of the knee in deep flexion. Poster #57, 28th Annual Meeting of the Canadian Orthopaedic Research Society, Winnipeg, Man., 12–13 June 1994. [11] Dye SF, Shapeero LG, Lipton MJ, Galvin EG. Quantitative assessment of functional morphology by means of cine computed tomography. Proceedings of the Third Annual Meeting of the

[12]

[13]

[14]

[15] [16]

[17]

307

Orthopaedic Research Society, San Francisco, CA, 19–22 January 1987. Van Eijden TMGF, Kouwenhoven E, Verburg J, Weijs WA. A mathematical model of the patellofemoral joint. J Biomech 1986;19(3):219–29. Hefzy MS, Jackson WT, Saddemi SR, Hsieh Y-F. Effects of tibial rotations on patellar tracking and patello-femoral contact areas. J Biomed Eng 1992;14:329–43. Li J, Scudamore RA, Khan B, Cooke TDV, Wyss U, Bryant T. Elements of lower-limb alignment in Caucasian and Arab patients having osteoarthritis with varus knees. Orthop Trans 1994;18:676–7. Ahlberg A, Moussa M. The anatomical pattern of gonarthrosis in Saudi Arabia. Int Orthop (SICOT) 1989;13:177–8. McAlindon TE, Snow S, Cooper C, Dieppe PA. Radiographic patterns of osteoarthritis of the knee joint in the community: The importance of the patellofemoral joint. Annals of the Rheumatic Diseases 1992;51:844–9. Dalqvist NJ, Mayo P, Seedhom BB. Forces during squatting and rising from deep squat. Eng Med 1982;11:69–76.