The relationship between pain with walking and self-rated health 12 months following total knee arthroplasty: a longitudinal study

Osteoarthritis (OA) of the knee is a common disorder that leads to pain and restricted mobility. Total knee arthroplasty (TKA) for OA aims to relieve pain and improve function when conservative treatment is no longer effective. Although the procedure is highly effective in terms of prosthesis-related outcomes [1] and quality of life [2], some patients continue to report a higher degree of functional limitations 1 year after TKA compared to healthy age matched controls [3]. Of note, in a recent systematic review [4], approximately 20% of patients reported little or no improvement in terms of pain relief.

The mechanisms that underlie this variability in TKA outcomes are likely to be multifactorial and complex [5]. For example, poorer self-efficacy with walking skills [6], more preoperative walking limitations, higher body mass index (BMI), slower 1 month gait speed, contralateral knee pain, and use of quadstick before surgery [7] were associated with poorer walking outcomes 6 months after TKA. Recently, our research group demonstrated that one in five patients reported no improvements in pain interference with walking 1 year after TKA [8]. In that study, pain interference with walking was measured at repeated intervals during the first year. In a growth mixture modeling, two subgroups of patients with different trajectories of pain with walking were identified. While one group of patients (i.e., Continuous Improvement group, 78%) improved steadily during the first year, a second group of patients (i.e., Recurrent Interference group, 22%) was characterized by initial improvements followed by recurrent pain interference with walking after the first 3 months, resulting in no improvements in terms of pain interference with walking 12 months after surgery. Patients in the Recurrent Interference group were characterized by higher preoperative pain, fatigue, and depression scores and poorer illness perceptions compared to Continuous Improvers. The current study builds on these results to further investigate how walking outcomes and a comprehensive set of predictors are related to a general health outcome (i.e., self-rated health (SRH) at 12 months after TKA.

Walking difficulties seem to influence patients’ future health. For example, walking difficulties were associated with higher mortality rates 10 years after TKA [9]. In addition, walking difficulty and shorter walking distance predicted higher health care costs, disability, and mortality [10] as well as poorer self-rated health [11] in the general population.

SRH is a global measure of general health status [12] and is considered a reliable predictor of a variety of outcomes including mortality and morbidity, recovery after disease, and use of health care resources in various patient populations [13‐15]. In addition, SRH is used as an outcome measure in population-based studies [16] and was associated with walking speed and walking difficulties [11]. Because SRH was responsive to changes in health and mental well-being during recovery after TKA [17], Peruccio and colleagues suggested that the use of this simple measure could identify patients who might benefit from targeted interventions to improve general health. However, to our knowledge, no studies have investigated the association between pain with walking, other pre- and perioperative factors, and SRH 1 year after TKA. Therefore, the aims of this study were to (1) describe changes in SRH from before until 12 months after TKA, (2) identify preoperative factors associated with poorer preoperative SRH, and (3) investigate the key predictors of poorer SRH 12 months after TKA.

Data collection for this longitudinal study was performed between October 2012 and September 2014 at a high volume surgical clinic in Norway. The clinic serves patients from all regions of Norway. Patients ≥18 years of age were eligible for inclusion if they were scheduled for TKA for OA, agreed to participate, and did not have a diagnosis of dementia. Patients undergoing unicompartmental or revision surgery were excluded. The study was approved by the Regional Medical Research Ethics Committee of Health South East of Norway (# 2011/1755).

This study aimed to investigate key predictors associated with poorer SRH 12 months after TKA. Findings from this study suggest that lack of improvement in walking during the first year following TKA was the strongest predictor of poorer SRH scores 12 months after surgery. After controlling for demographic, clinical, and personal factors, membership in the Recurrent Interference group was associated with a 12 point lower SRH score, well above the minimally clinically important difference of 8 points for this scale [29]. In addition, a number of pre- and perioperative factors (i.e., higher number of painful sites, higher BMI, perioperative use of ketamine as supplemental pain medication, higher preoperative depression scores and poorer preoperative SRH) were associated with poorer SRH scores after TKA. In contrast, poorer preoperative SRH was associated with higher levels of CRP, a higher number of painful sites, higher fatigue severity, and more illness concern related to the osteoarthritic knee.

The preoperative SRH scores in our sample were comparable to OA patients awaiting joint replacement and TKA patients [30, 31]. As expected, overall the sample’s SRH score improved significantly during the first year, with a large effect size (Cohen’s d = 0.77). An improvement of almost 14 points is well above the clinically important difference of 8 points for this scale [29], which suggests that this improvement is clinically meaningful. This result is in line with previous studies that suggested that TKA has a high success rate for the majority of patients [2]. The improvement in SRH found at 1 year after surgery is similar to a previous report [31].

Consistent with other studies [9, 32], the association between poorer trajectories of pain interference with walking and poorer SRH 1 year after TKA suggests that walking ability is an important predictor of future health. For example, postoperative walking disability was linked to higher mortality rates 10 years after TKA [9]. In the general population, poorer walking performance among older adults was associated with a higher risk for future incidence of cardiovascular disease, mortality, and mobility disabilities [32]. Patients in our study received physical therapy for up to 6 months after surgery. While rehabilitation programs for TKA patients tend to focus on muscle strengthening and range of motion of the joint, assuming that this approach will in turn improve patients’ ability to walk, these programs seem to have limited effect on walking ability [33]. Although physical activity levels tend to increase after TKA, patients do not reach the same functional levels as the healthy population, and postsurgical activity levels seem to be influenced mainly by patients’ physical activity levels prior to surgery [34]. However, a recent study of a rehabilitation program that used walking and weight bearing exercises had better results for walking speed six months after TKA compared to usual physical therapy [35]. In addition, a 3 month hiking program resulted in improved walking speed and quality of life [36]. Improvements in walking and pain are among the most important outcomes for patients [37], but a substantial number of patients experience reduced walking distance and difficulty climbing stairs [38]. These persistent problems may potentially lead to a less active life and reduced quality of life. Additional studies are needed to determine the effect of improved walking skills on SRH after TKA.

Interestingly, higher preoperative BMI was associated with poor SRH at 12 months after surgery, but not with SRH preoperatively. Obesity is considered a modifiable risk factor for OA [39] and weight gain is a risk factor for undergoing TKA [40]. With the global increase of obesity [41], the number of overweight patients undergoing TKA is likely to increase. Obesity is associated with a variety of short and long term complications after TKA [42]. Consistent with our finding, in a systematic review [2], obesity was associated with a poorer quality of life after TKA. A recent study found that patients with a BMI ≥30 kg/m² are at increased risk for poorer knee function, revision surgery within 5 years, superficial infection and deep venous thrombosis [43]. In addition, compared to non-obese patients, a systematic review found that patients with a BMI ≥40 kg/m² had significantly lower implant survivorship and lower function scores. However, this association was not seen in patients with a BMI ≥30 kg/m² [44]. Based on these findings, a cutoff BMI ≥40 kg/m² may guide patient selection for TKA. However, Kulkarni and colleagues argued that because all patients are unique, the decision to perform surgery should not rely solely on a patient’s BMI but consider each patient’s potential risks and benefits [42]. Patients in this category may need counselling about weight reduction and other treatment alternatives for painful OA of the knee. Obese patients with BMI ≥30 kg/m² needs to be informed about the risk of a poorer outcomes including SRH, prior to deciding to undergo TKA.

Patients with a higher number of painful sites before and 12 months after surgery were more likely to report poorer SRH, both prior to surgery and 12 months after TKA. This characteristic was the only factor common to both time points. Interestingly, number of comorbidities correlated with number of painful sites prior to surgery (.22, p = .008) but not with number of painful sites after 12 months (.10, p = .23). It is well known that patients with OA have a large number of painful sites often accompanied by pain sensitization [45] which can be linked to increased symptom severity. A higher number of painful sites as well as higher total symptom load was associated with persistent pain after TKA [46], lower functional scores [47], poor improvements in walking after TKA [21], poor overall health and sleep quality, female gender [48], and poorer health-related quality of life [49]. While preoperative widespread pain was associated with higher pain severity prior to and 12 months after joint replacement surgery [50], patients seem to gain the same amount of pain relief [50] and functional improvements [51] from their TKA compared to patients without widespread pain. Thus, multiple painful sites prior to surgery should be regarded as a risk factor for decreases in walking ability and poorer SRH 12 months after TKA.

Our study revealed that patients with higher preoperative depression scores were more likely to report poorer SRH at 12 months after TKA. This finding complements a systematic review of the literature stating that poorer mental health is associated with worse pain and functional outcomes 1 year after TKA [52]. However, in a recent study, depressed patients improved at the same or better rate as non-depressed patients and had similar satisfaction rates 1 year after TKA [53]. However, depressive symptoms are risk factors for poorer adherence with physical therapy [54]. Thus, identification of depressed patients may be important to individualize care plans and optimize their treatment outcomes.

Poorer SRH prior to surgery was significantly associated with poorer SRH 12 months after TKA. Similarly, poorer preoperative SRH was found to be predictive of poorer health outcomes and SRH 6 months following total hip and knee replacement surgery [55]. While in other samples, SRH was used as a predictor that captured a variety of future health outcomes [13‐15], its use as a predictor in joint replacement surgery is limited. Because this simple tool is easy and non-demanding for clinical use, it could be considered for inclusion as part of a screening tool to assess patients at risk for poor outcomes after TKA surgery. However, more research is needed to establish its usefulness in conjunction with other risk factors.

Except for number of painful sites, the factors associated with SRH prior to surgery were not the same as the factors associated with SRH 12 months after TKA. Prior to surgery, SRH scores were associated with higher fatigue scores and higher levels of the inflammatory marker CRP. A systematic review found that higher CRP levels in OA patients were significantly associated with pain, decreased physical function, but not with radiographic OA, suggesting that low-grade systemic inflammation may increase patients’ symptomatology [56]. Patients with hip and knee OA are characterized by multiple co-occuring symptoms including pain, fatigue, and depression [57], which may impact their activity levels [58]. Hawker et al [59] found that fatigue contributed to lower physical activity and increased pain in OA patients, which in turn may lead to poorer SRH. However, as these findings were not factors in the model for SRH 12 months after TKA, they may be linked mainly to the OA condition and do not significantly impact the trajectory of recovery after surgery.

This study has several limitations. While our sample was recruited from one single hospital, the surgical department is a high volume surgical clinic which admits patients from all regions of Norway, which we believe increases the generalizability of our findings. No data were available on previous injuries that may have caused the OA condition. Other factors not accounted for in our study may contribute to long-term outcomes after TKA (e.g., inflammation, reactions to metal and polyethylene). More specific details on other chronic diseases 12 months after surgery may have provided us with more detailed information about other factors that impact patients’ SRH. Finally, the use of objective measures of walking ability may have provided additional insight.

The prospective design and relatively large sample are advantages with our study. In addition, all patients received the same type of implant and a standardized regimen for pain management and mobilization. Finally all of the patients received physical therapy for 4–6 months after TKA.

In conclusion, patients with no improvements in terms of pain interference with walking report poorer SRH 12 months after TKA. In addition, poorer SRH after TKA was associated with higher BMI, a higher number of painful sites, higher depression scores and poorer preoperative SRH. Patients whose pain with walking over time does not improve need to be followed closely and may need additional physical therapy to improve their general health status. While a screening tool may enable clinicians to detect patients who need additional follow-up, more research is needed to develop and test such a screening tool. Finally, more research is needed to determine what type and dose of physical therapy is needed to improve TKA patients’ walking ability.