Stiffness after total knee arthroplasty

The Journal of Arthroplasty Vol. 19 No. 4 Suppl. 1 2004 Stiffness After Total Knee Arthroplasty Richard S. Laskin, MD,* and Burak Beksac, MD† Abstract: Limitation of motion after knee arthroplasty can be the result of a multiplicity of factors. Among these are malpositioning of the components, especially in the sagittal plane; oversizing at the patellofemoral or tibiofemoral joint spaces; retaining posterior osteophytes; and persisting with a tight posterior cruciate ligament. Postoperatively, problems with physical therapy likewise can cause limitation of both extension and flexion. Specific patient factors also may affect the range of motion after surgery. Although most patients achieve a postoperative flexion that is highly correlated to that which was present preoperatively, factors such as pain, obesity, and deformities of adjacent joints may limit such motion. Key words: knee, arthroplasty, replacement, stiffness, motion. © 2004 Elsevier Inc. All rights reserved. In an ideal world, after a knee arthroplasty, a patient would achieve full extension and full flexion. In the real world, however, there are many patients who will achieve less than full range of motion [1– 3]. Reasons for this are the result of surgically related factors, postoperative physical therapy and pain factors, the underlying diagnosis [4,5], or some innate factors related not to the surgery, but to the patient and to the kinematics of the prosthesis. The knee should normally come close to full extension at heel strike. Ambulating with a knee that does not fully extend increases the muscular work of walking, particularly in the quadriceps, and decreases endurance. For any patient, full flexion occurs when the soft tissues of the thigh and calf compress against each other. Most patients do not use this amount of flexion to participate in their activities of daily living. Walking on a level surface is possible if the knee flexes only 45° to 55° during swing phase. Ascending or descending an 8-inch stair riser requires about 85° of flexion [6]. Standing from a chair is facilitated if the foot can be brought under the seat; this usually requires about 95° of flexion. Some activities require more flexion, however. Kneeling in prayer is comfortable if the patient has about 125° of flexion, whereas squatting for eating or for hygiene usually requires approximately 135° of flexion. Component Malposition Component malposition, especially in the sagittal plane, during surgery can lead to profound loss of flexion or extension. Optimally, the femoral component should be at right angles to the anatomic axis of the femur (ie, of the distal half of the femur) in the sagittal plane. If the component is very hyperflexed, it can lead to a potential block to full extension. If it is hyperextended, it can lead not only to anterior notching, but to a limitation of flexion. Intramedullary femoral alignment guides assist in proper positioning of the component to avoid these extremes [7–10]. However, they tend to introduce at least a few degrees of component flexion. Computer navigation gives a true mechanical axis of the entire limb in the sagittal plane. This From the *The Institute for Hip and Knee Replacement, Hospital for Special Surgery, New York, New York; and the †SSK Göztepe Education Hospital, Orthopaedics and Traumatology Department, Istanbul, Turkey. No benefits or funds were received in support of this study. Reprint requests: Richard S. Laskin, MD, 535 E. 70th St, New York NY 10021. © 2004 Elsevier Inc. All rights reserved. 0883-5403/04/1904-1010$30.00/0 doi:10.1016/j.arth.2004.02.008 41 42 The Journal of Arthroplasty Vol. 19 No. 4 Suppl. 1 June 2004 Fig. 1. The femoral component normally is placed at right angles to the anatomic axis of the distal femur in the lateral plane. Using a surgical navigation system that takes reference from the femoral head may result in a femoral component that is slightly hyperextended. mechanical axis is slightly hyperextended as referenced to the apparent mechanical axis that is determined using intramedullary non-navigated instruments (Fig. 1). If a femoral component is placed at right angles to the computer-navigated axis, it will be in hyperextension on the distal femur. The clinical relevance of this has not yet been determined; however, to obviate this situation, many surgical navigation systems arbitrarily suggest placing the femoral component in several degrees of flexion to the navigated axis rather than at right angles. Malposition of the tibial component in the sagittal plane likewise can cause a limitation of motion. If the tibial resection is upsloped (ie, higher posterior than anterior) (Fig. 2), the size of the joint space posteriorly is diminished, and flexion may be limited. The total downslope of the articular surface of a prosthetic knee is a combination of the downslope in the tibial resection and the downslope or posterior slope that is inherent in the tibial polyethylene. Some authors have suggested that the amount of tibial component articular downslope should be equal to that present preoperatively [5], whereas others use a standard amount of posterior slope for all patients. Problems with insufficient downslope are more obvious in the patient in whom the posterior cruciate ligament has been retained than in the patient in whom the ligament has been sacrificed or recessed. ated for the size of the implants inserted. Such over-replacement can lead to a diminution of both flexion and extension. Patello-femoral over-replacement occurs if the anterior flange of the femoral implant, rather than being in contact, is anterior to the cut anterior surface of the femur. This potential problem should be considered when using posterior referencing instrumentation. One must inspect the region of the anterior resection to see if the “footprint” or can- “Stuffing” “Stuffing” of a joint occurs if an insufficient flexion, extension, or patello-femoral space is gener- Fig. 2. An upslope of the tibial resection in the lateral plane can lead to a diminished flexion space posteriorly and limitation of flexion after surgery. Stiffness After TKA • Laskin and Beksac Fig. 3. In a valgus knee, the tibial sensor rests at the lowest point of the “good” tibial plateau medially. The lowest point is in the center of the plateau. cellous surface created approximates the undersurface of the femoral component. Additionally, the junction of anterior resection and residual anterior cortex should be palpated and inspected. One must decide whether a deeper cut, corresponding to a smaller-sized femoral component, can be appropriately made. The use of anterior referencing instruments may lead to less or less severe anterior over-replacement. However, such instrumentation only assures that the anterior proximal exit of the cut corresponds to the undersurface of the femoral component. Such instrumentation does not assure that the thickness of the resected bone, including saw kerf, equals the thickness of the anterior flange of the femoral component. Such sizing instruments relate to the anterior cortex in this region and not to the prominence of the trochlear surfaces and the specific geometry of the trochlear groove, which are those that the anterior aspect of the femoral component is replacing. With anterior referencing, it also may be possible to over-replace posterior femoral condylar bone, as mentioned several paragraphs below. Over-replacement of the patello-femoral space also can occur if insufficient bone and cartilage are removed for the thickness of the patellar implant inserted. The patella can be measured before resection and after insertion of the trial component to assess this situation. In many systems, a guide is available to ensure that the amount of bone and cartilage removed equals the thickness of the implant. In situations in which the patella is markedly eroded or abnormal in shape, the reference thickness for males and females as described by Scott [11] can be used. 43 Stuffing of the tibio-femoral space in extension occurs if insufficient bone is removed from the distal femur or proximal tibia. Most current surgical techniques stress removing an amount of bone from the distal femur at least equal to the thickness of the component being used. This practice helps to properly position the joint line. There is less uniformity of practice with the tibial resection. It is usually stated that the resection sensor should be placed on the lowest part of the articular surface of the “noninvolved plateau” of the knee, and that the tibial resection line should be placed between 6 to10 mm below this. For a valgus knee, the “noninvolved plateau” is medial. Because the medial tibial plateau is concave in both planes, the stylus should rest in the center (Fig. 3). This is not the case in the varus knee, however. Here, the “normal plateau” is lateral, and the lateral tibial plateau is convex from front to back. The lowest point on the lateral side is posterior, and at times, exposure problems may make access to this area difficult (Fig. 4). This situation may lead to placement of the stylus centrally and resection of insufficient bone. Consequently, when the trial components are inserted, the knee will not fully extend. Further resection of bone is then required. If the knee does not come to full extension at surgery, it probably will not do so after surgery. Over-replacement of the flexion space can occur if the femoral component chosen is larger than the femoral dimension anteroposteriorly, or, more likely, if the tibial resection line is upsloped. Both of these can limit flexion. There is an additional point of great importance, but separate from resection levels and instrumentation, with regard to accessing full extension. It is Fig. 4. In a varus knee, the tibial sensor rests at the lowest point of the “good” tibial plateau laterally. The lowest point is in the back of the plateau 44 The Journal of Arthroplasty Vol. 19 No. 4 Suppl. 1 June 2004 Fig. 6. The use of a femoral component that results in inadequate capping of the bone posteriorly can result in a limitation of flexion. Fig. 5. Posterior femoral and tibial osteophytes can limit both flexion and extension. the state of relaxation provided by the anesthetic technique. Anything other than full relaxation can, and commonly will, created artifactual flexion contracture. Although this is relatively unlikely in a standard subarachnoid spinal anesthesia, it can occur on occasion with poor relaxation using standard general anesthesia. This has not been seen with the use of epidural anesthesia at the author’s institution. Posterior Femoral or Tibial Osteophytes Decreased flexion and extension can occur secondary to problems at the posterior portion of the femur or tibia. Large osteophytes in that area must be removed (Fig. 5). If these osteophytes are not removed, the tibial polyethylene is blocked from its full sagittal excursion and there is a limitation of flexion. Such osteophytes can tense the posterior capsule in extension as well as in flexion and cause a paradoxical block to full extension. After the posterior femur and proximal tibia are resected, the knee should be distracted in 90° of flexion with a laminar spreader, and a curved osteotome and curette should be used to remove any remaining posterior osteophytes and to strip a small portion of the posterior capsule from the back of the femoral condyles. Another problem that can occur posteriorly is the result of using a femoral component that has inadequate posterior capping (Fig. 6). Residual bone in this area may cause impingement and limit flexion. This problem is accentuated if a femoral component that is smaller in the anteroposterior dimension than the normal anatomy of the femur is chosen. This combination decreases the posterior offset and can lead to a diminution of flexion (Fig. 7). Implant Design The posterior radius of curvature of the femoral component can have an important influence on the ability to flex the knee [12,13]. Kurosaka noted that the quadriceps force that is generated during active flexion is larger for implants with smaller posterior radii of curvature, and, therefore, that in the patient with a weak quadriceps force, there is a possibility that less flexion might be achieved on weight bear- Fig. 7. The use of a femoral component that is too small in the anteroposterior direction can lead to a decreased posterior offset and limited flexion. Stiffness After TKA • Laskin and Beksac ing. As the posterior radius of curvature of the femoral component is decreased, there is the potential for greater flexion. Such “high-flex knees” have the potential problem of causing abnormal polyethylene wear posteriorly during extremes of hyperflexion. A Tight Posterior Cruciate Ligament Retention of a posterior cruciate ligament that is too tight can cause a limitation of flexion [14,15]. In a study the senior author reported on in 1999, in the knee with a fixed angular contracture ⬎15° to 20°, the posterior cruciate often is a part of the deformity and should be released if full extension as well as coronal alignment is to be achieved. When the ligament was retained, there was a statistically significant decrease in flexion, as well as a high incidence of radiolucent lines and loosening over a 10-year period. Assessing Flexion and Extension at Surgery The knee should come to full extension with all the trial components in place. Traditionally, surgeons have suspended the extremity by the ankle and visually assessed extension. Unfortunately, the perception of full extension may be distorted in patients with obese extremities. One method described is to suspend the extremity by the ankle, dorsiflex the ankle and foot, and then push proximally on the sole of the foot. A flexion result with this maneuver implies that residual flexion position exists. However, a negative response does not assure the absence of a mild flexion contracture. It recently has been seen in combination with computer guidance that 5° to 10° of flexion contracture may not lead to the positive flexion result. This may result from the center of gravity of the involved extremity and friction at the tibio-femoral surfaces. If the knee does not fully extend at surgery (before deflation of the tourniquet), it probably will not fully extend after surgery. Correction for a residual flexion contracture should therefore be performed during the surgery and may include rebalancing of the soft tissues, resection of additional bone from the distal femur to increase the extension space size, or posterior capsular stripping. In the early 1980s, it was routine to perform a transverse section of the posterior capsule for cases with a large flexion contracture. It was felt that with the knee flexed, the popliteal vessels were displaced 45 posteriorly, protecting them during such capsular cutting. Unfortunately, this is not the case, and, indeed, Zaidi [16] demonstrated in 1995 that the artery actually transpositions anteriorly during acute knee flexion, making it more vulnerable. A safer alternative is to strip the capsule from its attachment on the posterior distal femur to release it, rather than cutting it. At surgery, with the trial implants in place, the hip should be held in flexion and the calf allowed to passively flex. The amount of flexion that will be obtained after surgery is within 5° to 10° of that demonstrated by this maneuver. Physical-Therapy Problems Postoperative problems with the patient’s physical-therapy program can lead to a diminution of flexion or extension or both. When a patient is in a continuous passive motion (CPM) machine, the knee axis often is not aligned with the flexion axis of the machine. If the machine is allowed to cycle throughout the night and day, the patient often will never come to full extension and may begin to develop a flexion contracture. This is especially true in the patient with a heavy thigh or calf. The routine use of a pillow under the knee at night, although comfortable for the patient with some swelling in the knee, may lead to the development of a flexion contracture as well. The authors use a CPM machine for several hours per day and a knee splint for the first few days after surgery as part of their therapy regimen. Problems in Patients with Bilateral Angular Deformities Patients who have severe bilateral gonarthrosis may undergo both knee surgeries under the same anesthesia, or in a staged manner. There are indications and contraindications for both scenarios. There is one potential problem in patients who have bilateral severe angular deformities if the surgeries are staged. After the first surgery, the operated leg is longer than the unoperated extremity. The patients then bend the operative knee to achieve leg-length equality and balance when they walk, and may develop a flexion contracture on the operated side when none existed at surgery. The solution is to use a shoe lift on the unoperated side as a temporary measure until it too undergoes surgery. As well, full extension must be emphasized while recumbent. 46 The Journal of Arthroplasty Vol. 19 No. 4 Suppl. 1 June 2004 Patient Factors References As a broad generality, most patients achieve flexion to within 10° to 15° postoperatively of what they achieved preoperatively [1]. There are 2 subgroups, however, that do not follow this 1/1 relationship directly. Those patients who have severe limitation of flexion preoperatively tend to get more flexion postoperatively, but never to the degree of those who do not have marked preoperative limitation. Furthermore, and somewhat more disturbing, there is a subset of patients who have high degrees of flexion before surgery who never achieve that degree of flexion postoperatively. This latter group of patients is the one that may benefit from some implant modification, or possibly therapy changes to maintain the ability to hyperflex the knee. Pain control is an important factor in achieving motion after surgery. The use of continuous epidural anesthesia supplemented by a femoral nerve block and a mini surgical incision has enabled the authors to obtain rapid, early flexion to 85° to 90° within the first 2 days after surgery in the majority of patients. Patients who have had problems with pain control before surgery may be well served by evaluation from a chronic pain management team before the surgery and the establishing of a plan for their care. Obesity, per se, as measured by the body mass index, is essentially unrelated to range of motion after surgery. There are some patients who have truncal obesity but have thin legs. Others have moderate truncal obesity but with heavy legs. It is this latter group that has difficulty regaining flexion after surgery. However, much of this is probably related to difficulties with proper placement and use of a CPM machine. For all postoperative total knee patients, the authors have stressed passive flexion exercises on stairs or steps to supplement the benefit achieved from a CPM machine. In summary, limitation of motion after surgery usually is secondary to a combination of factors related to surgical technique, prosthesis design, and inherent patient anatomy and physiology. It is incumbent on the orthopedic surgeon to minimize those factors related to surgical technique over which he or she has control, to maximize the potential for functional motion postoperatively. 1. Anouchi YS, McShane M, Kelly F, et al: Range of motion in total knee replacement. Clin Orthop 331: 87, 1996 2. Lizaur A, March L, Cebrian R: Preoperative factors influencing the range of motion after knee replacement for severe osteoarthritis. J Bone Joint Surg Br 79:626, 1997 3. Harvey IA, Barry K, Johnson R, Elloy MA: Factors that affect the range of movement of total knee arthroplasty. J Bone Joint Surg Br 75:950, 1993 4. Ritter MA, Stringer EA: Predictive range of motion after total knee replacement. Clin Orthop 275:115, 1979 5. Maloney WJ, Schurman DJ: The effects of implant design on range of motion after total knee arthroplasty. Total condylar versus posterior stabilized total condylar designs. Clin Orthop 278:147, 1992 6. Laubenthal KN, Smidt GL, Kettlekemp DB: A quantitative analysis of knee motion during activities of daily living. Phys Ther 52:34, 1972 7. Laskin RS: Alignment of total knee components. Orthopedics 7:62, 1984 8. Laskin RS, Turtel A: The use of an intramedullary tibial alignment guide in total knee replacement arthroplasty. Am J Knee Surg 123:2, 1989 9. Laskin RS: Total knee arthroplasty in the presence of large bony defects of the tibia and marked knee instability. Clin Orthop 248:66, 1989 10. Bono J, Laskin RS: Tibial intramedullary alignment in total knee arthroplasty. Am J Knee Surg 8:7, 1995 11. Chmell MJ, McManus J, Scott RD: The thickness of the patella in men and women with arthritis. Knee 2:239, 1995 12. Kurosaka M, Yoshiya S, Mizuno K, et al: Maximizing flexion after total knee arthroplasty. J Arthroplasty 14(Supp 1):59, 2002 13. Dennis DA, Komistek RD, Stiehl JB, et al: Range of motion after total knee arthroplasty. J Arthroplasty 13:749, 1998 14. Laskin R: Total knee replacement with posterior cruciate ligament retention in patients with a fixed varus deformity. Clin Orthop 331:29, 1996 15. Lombardi AV, Mallory TH, Fada RA, et al: An algorithm for the posterior cruciate ligament in total knee arthroplasty. Clin Orthop 392:75, 2003 16. Zaidi SHA, Cobb AG, Bentley G: Danger to the popliteal artery in high tibial osteotomy. J Bone Joint Surg Br 77:384, 1995