Patient-reported outcome measures, the holy grail of outcome assessment: Are they powerful enough to show a difference in knee arthroplasty alignment? A call for more comprehensive and objective data collection

In the last 40 years over which total knee arthroplasty (TKA) has been carried out, mechanical alignment (MA) has been the commonly accepted gold standard for alignment in both clinical and basic science studies [7, 15]. In the last decade, on the basis of phenotype discussions and a better understanding of the patients' individual differences, a shift to a more personalized alignment has been recognized, aiming for closer restoration of the prearthritic knee anatomy and the native knee kinematics [3, 12, 21]. This trend towards more personalization has led to the introduction of many different alignment methods and surgical techniques.

The theme ‘Orthopedics is all about anatomy, plus a little bit of common sense’ is one of the most quoted sentences in sports orthopaedics but is also applicable to knee arthroplasty. There are different knee constitutions [12, 16], and not every knee that requires TKA is considered to have pathological alignment. Since the current literature has confirmed that MA changes the constitutional alignment in the majority of knees [11], modifying their original anatomy, other philosophies have recently emerged [13]. Stephen Howell's kinematic alignment (KA) was a revolutionary approach that changed the mindset of knee surgeons aiming to restore the prearthritic knee anatomy [14], with the hope of improving the outcome of patients after TKA [6].

At present, KA is expected to achieve the desired alignment targets much more accurately and its proponents hope to achieve a success rate of total hip arthroplasty in TKA outcomes. Still, the risk of extreme postoperative alignments when unrestricted KA principles are followed is real. The question is where to draw the line between normal and pathological and to which degree the constitutional coronal alignment should be restored [13]. Restricted kinematic alignment (rKA) strives to set these standards, as it restores the prearthritic knee phenotype, except for extreme native alignments, which are adjusted to previously set criteria of safe zones [30]. This set of criteria could help the surgeon to determine what is normal or acceptable deformity and should be restored to prearthritic values, to respect the constitutional soft tissue envelope. On the other hand, abnormal constitutions should be identified, and outliers should be adjusted inside safe zones to avoid deleterious effects on implant survivorship. It is also of paramount importance to recognize the threshold for the pathological knee alignment, which should not be restored using personalized arthroplasty philosophy and should be corrected [11]. The recommendations for implementing KA principles have recently been defined more clearly and could lead to more standardized results [20].

The data supporting KA are currently not so convincing as to result in a new gold standard: on the other hand, historically, in the beginning, most new techniques did not show immediate superiority in terms of the gold standard, becoming widely used after a period of adaptation.

The literature in support of this shift has grown exponentially. The current authors used the search engine Embase on 21 July 2024, utilizing the search words ‘kinematic alignment’ and ‘total knee arthroplasty’: 513 articles on these topics were identified between 2008 and 2024, mainly from European authors [22, 23].

However, a few major concerns remain: (1) Are patient-reported outcome measures (PROMs) and other subjective measurements sufficient indicators of TKA outcomes to support shifting from a traditional approach like MA to a more personalized one like unrestricted KA or rKA? (2) Should the orthopaedic community focus on acquiring more objective data (i.e., spatiotemporal and kinematic parameters as determined by gait analysis and three-dimensional [3D] platforms) before widely endorsing the use of alternative surgical strategies? (3) What are the current protocols for acquiring more objective data following TKA?

PROMs have been historically used to assess the quality of life (QoL) and the level of activities of daily living (ADL) of patients who have knee osteoarthritis (OA): this information has also been used to justify the clinical indication for joint replacement surgery and ultimately to monitor the surgical outcome [17, 19]. In recent years, especially in North America, PROMs have been utilized not only to monitor and compare provider performance but also to adjust providers’ and medical institutions’ reimbursement following TKA [25]. Many studies have contested the validity of PROMs as tools to reflect objective measures of function: Hill et al. [10] showed a weak association between PROMs and standard functional tests (SFTs) among patients affected by knee OA or among patients who underwent TKA. Gojło et al. [8] suggested that, based on data of postoperative evaluations and patient improvement observed in PROMs, surgeons might have an overrated picture of patients' health after joint arthroplasty. Eckhard et al. [4] reported, in TKA patients, an alarming ceiling effect of several PROMs, including the Knee Injury and Osteoarthritis Outcome Score (KOOS) in its pain, symptoms, ADL and QoL subscales, the total Western Ontario and the McMaster Universities Osteoarthritis Index (WOMAC) and finally the KOOS joint replacement short form score. The definition of an optimal time frame between the surgical intervention and the acquisition of PROMs has also been a matter of debate: Ekhtiari et al. [5] showed that PROMs rapidly plateau by 6 months following TKA, with no further clinically significant improvements afterwards; interestingly, this trajectory was independent of patients' demographic data and comorbidities.

Because of this uncertainty, there is an increasing perception that PROMs are not sensitive enough and that they do not assess just the knee, but the overall status of the patient as well. PROMs should therefore be used with at least one performance-based outcome measure (PBOM) and impairment-based outcome measure (IBOM), as more objective tools [1, 19]. Examples of these objective measures of physical function include multiple spatiotemporal parameters (cadence, stance/swing times, step time and step length) and sagittal kinematic variables (trunk, hip and knee ROM) [10]. Gait analysis represents the ideal tool to obtain objective data following TKA, especially at present, thanks to the increasing use of 3D motion capture technologies [28, 31]. The classical approach of infrared camera-based systems coupled with fixed-point patient markers [26] has been recently modified by the combination of dynamic electromyography (EMG) and gyroscopes and accelerometers associated with smartphones [29].

The applications of these technologies during gait analysis highlighted major differences between subjective reports of physical function and pain (i.e., KOOS scores) and objective gait parameters: Boekesteijn et al. [3] showed that KOOS scores greatly improved within the first 2 months, while spatiotemporal gait parameters mainly improved between 2 and 15 months after surgery. In a similar comparative study, Graff et al. [9] investigated the potential relationships between four validated and widely used patient-reported questionnaires (Knee Society Score or KSS; Oxford Knee Score or OKS; KOOS and the 12-Item Short Form Health Survey or SF12) and three objective outcome measures (muscle strength, knee laxity and the Timed Up and Go Test [TUG]): the TUG test was the only objective measure to demonstrate a statistically significant correlation with PROMs.

The need for objective data to drive our surgical decision in TKA is obvious. However, the literature is still sparse in terms of recommended protocols for the acquisition of objective data following TKA [24]. Recently, a radiologic-based protocol has been introduced [2], to be used in a standard hospital setting, for pairing classical PROMs with more objective, kinematic data: this protocol includes the execution of basic radiographs (single-leg stance in extension, lunge, squat and kneeling) as recommended by previous studies [27] and utilization of fluoroscopy (setting the fluoroscopic system with X-ray pulses up to 20 ms) during chair-rise, stair ascent and rapid open-chain knee flexion–extension cycles to reproduce the entire gait cycle: to the current authors, this protocol still requires a multidisciplinary approach and a high level of expertise, both of which are not available in many standard hospital settings. Vij et al. [31], in a systematic review of current applications of gait analysis after TKA, showed that the literature is sparse regarding gait analysis studies reporting both preoperative and postoperative kinematic and clinical data: the same authors recommended including defined motion analysis parameters (knee adduction moment, knee adduction impulse, total ROM throughout the entire gait cycle, varus angle, cadence, stride length and gait velocity) in clinical outcome studies.

In an extremely exciting time where there is increasing enthusiasm for changing multiple dogmas elaborated by our mentors and masters but also at a time where most knee surgeons still consider MA as the gold standard [21], extreme caution needs to be exercised before a universal personalized approach in TKA alignment can be considered to lead to better outcomes and less dissatisfaction among patients. Some of the advanced technologies that are currently and enthusiastically used, especially by the new generation of knee surgeons [18], should be applied to obtain more objective data from our patients following TKA surgery. Like in all emerging techniques, there are early adopters, who very much want to prove the success of their preferred technique. It is time to consider the more objective outcomes of different alignment philosophies more carefully in an unemotional, basic science manner.

The authors declare no conflict of interest.

Our manuscript has not been submitted, published or under consideration for publication in the same or substantially similar form in any other journal. All colleagues listed as authors on the byline are qualified for authorship and have read and approved the article.