The bony geometry of the knee joint plays a large role in normal joint ROM and function. The distal femoral anatomy is such that the medial and lateral condyles are asymmetric with respect to their radii of curvature and overall diameters.¹ The nondiseased proximal tibia has approximately 5° to 10° of posterior slope.² Finally, the posterior condylar offset of the femur cannot be understated in its importance in allowing for adequate flexion. Without the posterior condyles being offset from the posterior femoral cortex, the posterior tibia would impinge on the posterior cortex of the femur in flexion, thereby preventing deep flexion.¹ In design of TKA implants and reconstruction of the knee, the offset from the posterior condyles must be recreated and not reduced significantly in size. In addition to bony clearance in deep flexion, this posterior condylar offset allows for soft tissues to clear up to approximately 150°, although this is increasingly seen as a significant source of motion restriction in the growing population of obese individuals. These patients who are experiencing thigh-calf impingement can pose specific challenges for surgeons performing TKA, as stiffness is a significant postoperative concern.

The ligaments and soft tissues of the knee are too many to review in detail here but suffice it to say they all play significant roles in the kinematics of the joint. Normal joint motion is afforded by the dynamic interplay between compliant capsular tissue, cruciate and collateral ligaments, and mobile musculature that can lengthen and contract to their full extent.

When the anterior cruciate ligament (ACL) and posterior cruciate ligament (PCL) function normally, they allow for the geometry of the distal femur and proximal tibia to interact in flexion-extension as well as rotation during the course of the gait cycle. With pathologic processes leading to either ACL or PCL deficiency, development of arthritis and subsequent contractures about the knee occur.³,⁴ For example, with PCL deficiency, there is decreased femoral rollback on the plateaus leading to earlier impingement from the tibia and therefore limited knee flexion.⁵

Perhaps more important than the cruciate ligaments are the collateral ligaments when it comes to knee joint contracture and stiffness. Coronal plane soft-tissue balancing may involve selective releases of the medial and/or lateral collateral ligaments (MCLs, LCLs). One study by Mihalko et al evaluated the soft-tissue release or bone resection maneuvers performed during TKA to achieve balance and full motion in patients with flexion contracture.⁶ They report that only MCL- or LCL-balancing procedures were necessary to correct FFD in 35.9% of patients—thereby implicating the collaterals as the most likely primary structure contributing to the original contracture. Nonetheless, the etiology of stiffness in knees is multifactorial, and despite properly functioning ACL, PCL, MCL, and LCL, tightness in the posterior capsule, extensor mechanism, and bony deformity can lead to coronal and sagittal plane deformities in the long run.⁷

Despite attempts at defining a normal knee joint arc of motion (AOM), it is important to point out that individual factors will all dictate a person’s everyday knee range in a nonarthritic state. These factors include genetic predisposition and patient factors such as age as well as external influences such as cultural environment, body habitus, and engagement in activities or exercise.⁸,⁹,¹⁰ Studies evaluating normal joint motion have shown that with age, flexion certainly decreases but should stay above 130°, while extension may slightly increase and remain between 1° and 2° by age 69 years.¹¹

As far as required ROM for normal activities of daily life, studies have reported that at least 90° of knee flexion and even up to 135° may be needed for comfortable execution of many activities including stair climbing, sitting on chairs, toileting, and getting into a bathtub safely.¹²,¹³ Particularly in non-Western cultures, individuals may need up to 165° of knee flexion in their everyday lives such as during squatting, kneeling, or sitting cross-legged during prayer or other activities.¹²

These deep flexion activities further exaggerate the posterior femoral rollback that was discussed earlier. Past 135° of knee flexion, several studies have shown that the lateral femur rolls posteriorly up to 31 mm from the center of the knee and even translates completely off of the lateral tibial plateau.⁵,¹⁴,¹⁵ Stability of the knee joint is maintained at these extreme angles by the lateral meniscus, which as mentioned earlier is mobile and translates with the femur as it rolls back, as well as by the musculature including the contraction of the quadriceps muscle.⁵ Without normal knee kinematics and resultant flexion contractures greater than 15°, the energy expended by the musculature during walking has been measured to be significantly increased in both healthy patients and those who underwent TKA.¹⁶

The amount of motion in a knee is significant as far as function is concerned; however, it is equally important when trying to discuss patient motion in the clinical and research settings. Classification of knee ROM has been performed in a myriad of ways in the literature with not much universal agreement.¹⁷ Sculco defined limitations of motion as mild, moderate, or severe based on the total ROM arc.¹⁸ His classification quantified severe ROM as an arc of less than 30° to 45°, moderate with an arc of 45° to 70°, and mild with a motion arc of 70° to 90°. This classification allowed for an assessment of contracture severity prior to undergoing TKA and even helped dictate the type and amount of soft-tissue releases needed to be performed at the time of operation.¹⁸

Scoring systems used in evaluating patients with both acute and chronic injuries of the knee are several and vary based on the goals of assessment.¹⁷,¹⁹ One of the main objective criteria in several of these scoring systems is the degree of flexion contracture and the arc of normal knee ROM. In the Knee Society Score (KSS), 25 points out of a possible 100 are achieved through amount of knee flexion (one for every 5° motion) and presence of flexion contracture.²⁰ The amount of knee motion has been further shown by Ritter to affect other aspects of KSS assessment including the walking ability score and stair climbing score.²¹ The Hospital for Special Surgery (HSS) Knee Score meanwhile delegates 18 points out of 100 for ROM component (one for every 8°).

The importance of developing a good classification system for knee joint motion is common to all good classification systems in orthopedics: that is, the system allows for a common language in the clinical and research settings and also provides either diagnostic or prognostic information, thereby ultimately guiding clinical treatment. Jain et al classified preoperative FFD into two groups and reported on the surgical considerations intraoperatively as well as the postoperative outcomes in ROM improvement after TKA for each group.²² This
allows for a discussion with the patient as to what ROM category they are in preoperatively and what to expect after a surgical intervention. Similarly, ROM classification after TKA has been shown by Quah et al to predict expected improvement in flexion contracture following surgery over the course of 2 years.²³ In their study, both groupings (FFD 5°-15° vs. >15°) improved significantly over 2 years; however, the first group had complete resolution of FFD, whereas the second group had residual mean contracture of 3° postoperatively.

Recreation of the normal knee joint motion is the goal of any knee surgery including TKA for arthritis. Restoration of biomechanical principals to normal knee kinematics has evolved over several decades, yet it still remains a challenge for arthroplasty surgeons. Much debate regarding the optimal method to restore joint anatomy and motion continues; however, there is no question that joint motion plays a large role in patient satisfaction and reported outcomes after treatment. Therefore, classification and scoring systems utilizing motion as a predictor for success and guide for treatment are useful and necessary tools.

The most common joint disorder of the knee in the United States is osteoarthritis.²⁴ Symptomatic knee osteoarthritis is said to affect 10% of men and 13% of women age 60 years or older, and the vast majority of TKA patients are diagnosed as having osteoarthritis.²⁵ Risk factors for the development of osteoarthritis of the knee include systemic and local factors such as genetic predisposition, obesity, age, female gender, and muscle weakness as well as sports participation, prior knee injury, and occupational exposures.²⁴,²⁵

Due to its overwhelming prevalence, osteoarthritis is then logically one of the most common reasons for flexion contracture in patients indicated for TKA.⁷ The preoperative FFD is due to a multitude of factors but usually includes a combination of stiff collateral and cruciate ligaments, tightened posterior capsule, and posterior osteophyte formation. A study by Fishkin et al evaluated the changes in the medial and lateral collateral ligaments with osteoarthritis by evaluating and comparing the stiffness in a group of osteoarthritic patients undergoing TKA with a group of osteoarthritic cadavers and a group of normal control nonarthritic cadavers. As expected, when strain was applied and assessed, the osteoarthritic groups (patients and cadavers) had statistically significantly increased stiffness measurements than the normal controlled group of cadaver knees (60.7 vs. 21.4 N/mm in medial compartment, 29.2 vs. 19.5 N/mm in the lateral compartment, P < .05).²⁶ The recommendation therefore in patients undergoing TKA with FFD due to osteoarthritis is to take special care to release and balance the collateral ligaments carefully in addition to performing adequate capsular release, posterior osteophyte excision, and bony resection.⁷

Rheumatoid arthritis (RA) is a classically described cause of severe preoperative FFD with additional deformities of valgus and external rotation of the knee joint.²⁷ With the advent of disease-modifying antirheumatic drugs (DMARDs), TKA performed for RA was thought to diminish significantly. However, epidemiological data from the Nationwide Inpatient Sample prove the rate of TKA being performed for RA to remain stable from 2002 to 2013.²⁸ The incidence in 2002 was 3.3% versus that in 2013 was actually 3.5%, suggesting a slight increase, likely due to the aging United States population as well as the overall increase in utility of the operation for all patients across the board.

Several studies have evaluated the degree of flexion contracture in patients with RA undergoing TKA and have documented the amount of correction attainable at final follow-up.²⁹,³⁰,³¹,³² Wang et al describe a series of 38 patients undergoing bilateral TKA for moderate to severe FFD with mean contracture of 38° and mean ROM of 49°. They were able to completely correct the contracture in 33 patients, with only 5 patients having residual contracture of 5° to 10° after TKA.²⁹ They and other authors advocate the use of careful soft-tissue release rather than bony overresection to avoid complication and need for significant constraint during primary TKA.²⁹,³²

Severe contracture occurs due to frank ankylosis of the knee joint, which can be caused by bony disease or significant fibrous tissue between the distal femur and proximal tibia. These patients are often resting in significant flexion or extension with minimal arcs of motion and have been reported on in the United States and in Asia as well.³³,³⁴,³⁵ Interestingly, because the patients have often lived with their severe contracture for so long, pain is often a secondary issue relative to functional deficits in these patients. The preoperative ROM is usually between 0° and 10° with severe flexion contractures of up to a mean of 105° in one study.³⁵ Kim and colleagues were able to achieve a correction to mean FFD of 6° after TKA with motion of 97° in their series of 27 patients.³⁵ The most important consideration in treating patients with bony
ankylosis with TKA is recognizing and avoiding potential complications. Reported complications include patellar tendon avulsion, patellar fracture, peroneal nerve palsy, hematoma, and deep joint infection.³³,³⁴ Infection must be ruled out in these patients as undiagnosed chronic septic arthritis of the native knee joint leading to severe contracture can be catastrophic if TKA is undertaken.

Hemophilia is a more rarely encountered reason for undergoing primary TKA; however, it is one of the most commonly reported reasons for flexion contracture in the arthroplasty literature.³⁶,³⁷,³⁸,³⁹,⁴⁰,⁴¹,⁴²,⁴³,⁴⁴ Hemophiliacs experience fixed flexion as a result of extrinsic tightness from quadriceps contractures and surrounding fibrous ankylosis over the course of years.³⁹,⁴¹ Patients with hemophilia undergoing TKA have inferior outcomes compared to those with primary osteoarthritis.³⁷,⁴² Traditionally this was due to a significantly higher complication rate including hemarthrosis, deep joint infection, MCL injury, and periprosthetic fracture.⁴²,⁴³

Careful surgical technique and management of preoperative expectations regarding ultimate ROM have led to encouraging results with TKA in hemophilia patients with FFD.⁴⁴

Kubes et al report on 72 TKAs done in patients with hemophilia types A and B and showed an improvement in mean flexion contracture from 17° preoperatively to 7° postoperatively.⁴² They report that TKA in hemophiliacs increases extension more so than flexion and that patients should be operated on prior to surpassing a flexion contracture of greater than 22° to obtain satisfactory improvement in extension. Similarly Atilla et al state based on a receiver operator curve analysis that a preoperative flexion contracture of 27.5° is an important threshold in predicting whether hemophilia patients will significantly improve their amount of extension.³⁶

Although the list of potential diagnoses (including several other inflammatory arthropathies that we would not detail in this chapter) leading to flexion contracture prior to TKA is lengthy, several miscellaneous causes of stiffness that are less reported in the literature should be mentioned. An obvious etiology is prior surgical intervention on the knee joint leading to scar formation and ultimately stiffness either in flexion, extension, or both. While arthroscopic procedures such as ACL reconstruction can result in significant swelling of the joint and eventual stiffness, open procedures are more likely to result in the scarring, leading to an eventual flexion contracture. Specifically, procedures on the extensor mechanism such as quadriceps or patellar tendon rupture repair and patellar fracture fixation can lead to shortened extensors, adhesions in the musculature or retinaculum, and ultimately a lack of full knee joint extension.⁷

Trauma to the knee joint resulting in either a fracture or soft-tissue injury will often result in deformity and contracture as well. This can be due to fracture malunion, scar tissue formation, or actual development of arthritis. Often referred to as posttraumatic arthritis, Blagojevic showed in a meta-analysis that previous knee trauma resulted in 3.86 odds ratio of developing osteoarthritis of the knee.⁴⁵ Finally, the presence of pain in the knee itself is often enough to cause joint stiffness, and if left long enough without motion, eventual flexion contracture of the knee. Much is written about pain as a consequence of osteoarthritis and the inflammatory cascade that occurs within the knee joint.⁴⁶ However, pain as a cause of stiffness is more difficult to prove yet intuitively makes clinical sense. Often seen with patients in the clinical setting both before and after surgery, pain in the knee leads to a refusal to move the joint and, as such, an eventual stiffness and contracture of the surrounding ligaments and muscles about the knee.

Due to the success in correcting flexion contracture deformity during the actual TKA procedure itself, treating stiffness before TKA is not often pursued. Unlike sports medicine procedures such as ACL reconstruction in which optimal postoperative outcomes are usually the result of close to normal preoperative knee joint ROM, patients undergoing TKA rarely are able to achieve normal knee kinematics before their TKA.

Nonetheless, there has been significant evaluation of physical therapy (PT) prior to TKA in efforts to improve postoperative outcomes. Termed “prehabilitation,” several studies have evaluated the efficacy of preoperative education and exercise treatment or evaluation on functional recovery and disability after TKA.⁴⁷,⁴⁸,⁴⁹,⁵⁰ One study by Mizner et al finds preoperative quadriceps strength to be a good predictor of postoperative functional ability up to 1 year after TKA.⁴⁷ Meanwhile a cohort study evaluated three groups for effects of preoperative exercise (Group 1 control, Group 2 strengthening and ROM group, Group 3 cardiovascular exercise group) on knee function scores after TKA found no difference between the three with regard to all measurement scales utilized.⁴⁹

Debate exists regarding the efficacy of prehabilitation on TKA outcomes as is seen from the results of a Cochrane review and systematic review in the literature.⁵¹,⁵² In one of the only randomized studies, Swank et al found patients with severe osteoarthritis undergoing usual care and exercise prior to TKA had better strength and function after surgery compared to those patients only undergoing usual care protocols.⁵³ However, even based on the multitude of studies, it is still difficult to draw any definitive conclusions or make recommendations for or against the use of preoperative PT and education due to the large variety of protocols, small numbers of patients, and lack of randomization in most of them.

Surgical options for degenerative arthritis leading to stiffness prior to TKA are truly limited to arthroscopic
interventions such as meniscal débridement and synovectomy. Two landmark papers including randomization to arthroscopic intervention or placebo-type surgery both showed no advantage to either arthroscopic lavage or débridement in patients with degenerative joint disease.⁵⁴,⁵⁵ This has led to significant change with regard to the indications for knee arthroscopy in alleviating joint pain and stiffness in patients with existing arthritic changes. Interestingly, however, one paper from Japan reports on the unique use of an arthroscopic posteromedial release in patients with medial osteoarthritis and flexion contracture.⁵⁶ They report in 58 knees with Kellgren and Lawrence grades of 3 or 4 that the mean improvement in functional and pain scores was significantly increased after intervention, suggesting a possible role for less invasive surgery in temporizing patients with FFD prior to TKA. Even so, it is difficult to endorse this intervention in this uncontrolled study, and so much of the management of FFD can and should be done with the use of TKA as will be discussed below.

The surgical technique in patients with severe deformity always begins with special attention to the approach starting with placement of the incision. Old surgical scars must be mapped out to determine if utilizable, and the location of the incision should be extensile in the event as additional exposure is required. In general, the most lateral skin incision should be used. The skin may be thick and adherent to the underlying extensor mechanism from resting in a flexed position for prolonged periods of time. The specific comparisons of a median parapatellar approach versus a midvastus or subvastus approach will not be discussed here, but even in the noncontracted patient, several randomized studies have failed to show a definitive difference in postoperative motion, knee scores, or final quad strength.⁵⁹,⁶⁰,⁶¹

Once an arthrotomy is undertaken, exposure of the tibiofemoral joints may require removal of patellar osteophytes, or if significantly contracted and adherent to the underlying femur, osteotomy of the patellar undersurface and a lateral release for aid with mobilization. When FFD is so severe that patellar mobilization is not possible in spite of these efforts, significant proximal and distal extension of the incision may be required for potential use of a quadriceps snip, V-Y turndown, or tibial tubercle osteotomy (TTO).¹⁸ The TTO can be used with soft-tissue attachment laterally and secured at the end of the TKA with wire fixation with good postoperative outcomes and high rates of bony union.⁶² We have no experience with a V-Y turndown. The need for a TTO is rare, as most knees can be safely exposed with a quadriceps snip and an extensive medial peel on the tibia, aided by tibial external rotation and an anterior drawer maneuver.

A discussion on posterior cruciate retaining versus substituting implant design will not be provided in this chapter. However, it should be noted that there remains debate as to the effects of the PCL on restoring normal knee joint kinematics after TKA.⁶³,⁶⁴ Nonetheless, in general, the retention or substitution of the PCL has not shown to significantly affect postoperative ROM, functional outcomes, proprioception, or strength in most studies to date.⁶⁵,⁶⁶,⁶⁷ In patients with severe FFD and contracture, we feel as though the PCL will likely be pathologic in function regardless and can be safely excised during the initial approach to further allow for improved exposure and subluxation of the tibia from the femur.

Regarding the distal femoral cut, there are varying philosophies as to initial intraoperative management in a patient with FFD. Initial additional distal femoral resection can lead to joint line elevation, patella baja, and mid-flexion instability—all of which can lead to problems of their own after complex TKA.² It has been shown that an additional 3.5 mm of distal femoral resection is needed to completely correct 10° of FFD.⁶⁸ On the contrary, not resecting enough distal femur initially will inevitably lead to flexion contracture after TKA in patients already presenting with severe preoperative FFD. Tanzer et al report a series in which they evaluate the natural history of flexion contracture after TKA where they avoided excessive bony resection of the distal femur. Patients had a mean preoperative FFD of 12.9° and immediate postoperative FFD of 14.8°, yet they found at final follow-up of just over 1 year, patients had FFD of 2.9° suggesting that flexion contracture can significantly improve after TKA and does not require excessive distal femoral resection.

Similarly, the tibial cut can greatly affect the knee range in both flexion and extension. Avoiding significant bony cuts will make a potential revision surgery easier and can avoid excessive use of constrained implants when possible. Recreation of tibial slope is one of the more important technical features and may be influenced by the surgeon’s choice of posterior cruciate substituting or retaining implant design.² In a three-dimensional model by Walker et al, tibial slope was found to be the most
important surgical variable in optimizing knee flexion with posterior tilt of 10° accounting for an improved flexion of 30°.⁶⁹