End-stage knee osteoarthritis with and without sarcopenia – and the effect of knee arthroplasty

Background Sarcopenia often accompanies osteoarthritis but the relationship between them is still unclear and no strong consensus can be reached. Previous studies revealed that OA in the hip or knee is associated with declines in muscle mass and strength. This study aims to examine the status of sarcopenia in individuals with symptomatic end-stage OA of the knee, the effect of this condition on their peri-operative TKA rehabilitation and functional outcome. Methods This prospective study was conducted between 2015 to 2018 at our hospital. Patients with end-stage osteoarthritis of the knee on queue for total knee arthroplasty were recruited into the study. Primary outcome measures were DEXA in lean muscle mass, muscle strength, knee motion and function. Secondary outcomes measures were Quality of Life (QOL) measurements in pain, psychological and physical health.


Background
Sarcopenia is de ned as the progressive decline in muscle mass and strength.This is a common condition in the aging population resulting in signi cant functional impairment and inactivity [1][2][3].The prevalence of sarcopenia increases with age, reaching an astounding 50% among the population aged 75 or above in the United States [4].Sarcopenia is often associated with frailty, falls, fractures and disability in this susceptible population [5][6][7].Furthermore, the disease is a strong predicting risk factor for mortality and morbidity among the older patients in the nursing home [8].
Notably, sarcopenia often accompanies osteoarthritis (OA).However, the relationship between sarcopenia and OA is still unclear and no strong consensus can be reached [9,10].One postulation is that sarcopenia and OA is a co-existing condition [11,12], while other studies have identi ed sarcopenia as one of the risk factors for OA progression [13].Conversely, there is a study speculating that OA may inhibit the progression of sarcopenia since it was observed OA patients had higher BMP4-positive tissues, which indicate the presence of satellites cells that may increase muscle regeneration capabilities [14].There is currently emerging evidence that OA may contribute to the development of sarcopenia among the elderly.
Multivariate regression demonstrated that the risk of sarcopenia is higher in patients with OA [15].Furthermore, cross-sectional studies revealed that OA in the hip or knee is associated with declines in muscle mass and strength [16][17][18][19].Patients with OA of the knee had marked decreased muscle mass in their lower limbs, but their upper limbs or trunk were normal [16].The cause of this is likely due to their decline in physical function and sedentary lifestyle to avoid joint pain and stiffness.The total reduction in energy expenditure, together with ageing-related gains in adipose tissue lead to many patients developed overweight or even obesity.The excess load will further exacerbate their knee OA progression, and it is the combination of these factors that been regarded to create and perpetuate a vicious cycle between obesity, sarcopenia and osteoarthritis [20,21].
Patients with end-stage OA of the knee will eventually pursue total knee arthroplasty (TKA) as the only viable option.TKA has been proven to relieve pain and regain patients' mobility.It has been widely accepted that TKA will greatly increase social, physical and quality of life [22][23][24][25].However, the effect of sarcopenia on end-stage OA patients, their rehabilitation and functional recovery following TKA have not been previously studied.
This study aims to examine the status of sarcopenia in individuals with symptomatic end-stage OA of the knee, the effect of this condition on their peri-operative TKA rehabilitation and functional outcome.

Sample Size
The estimated study sample size is 50.This calculation is based upon an average of 180 TKAs take place annually in our hospital.Taking the 180 end-stage knee OA patients as study population, level of signi cance as 0.05, allowable error as 0.25 sample standard deviation, the sample size would be 45.Expecting a 10% withdrawal rate, a total of 50 subjects were required.Instead, researchers were able to nalise the recruitment of 58 end-stage OA knee patients upon their ful lment of study prerequisites for this research.

Eligibility Criteria
The inclusion criteria are: (1) aged over 50 with end-stage knee OA; (2) scheduled for TKA; (3) agreed to given written consent and be able to comply with study assessments.Exclusion criteria include (1) history of connective tissue disorders or myositis condition; (2) previous cases of alcoholism or drug abuse; (3) breastfeeding or pregnant women; (4) presence of serious pathologies, steroid-based systematic therapy in progress or got interrupted of less than 1 month, or signi cant hematologic disease; and (5) presence of signi cant cognitive impairment.Physical Measurements Patients demographic were recorded upon enrollment.Bodyweight and height were measured using a standard stadiometer and their Body Mass Index (BMI) was calculated (bodyweight in kg/[height in m] 2 ).Body composition at baseline and follow-ups were measured using dual-energy X-ray absorptiometry (DXA) (Horizon, Hologic, Bedford, MA).Total appendicular skeletal muscle mass (ASM) was calculated by the sum of lean mass measured in the 4 limbs, with the operator adjusting the cut lines of the limbs as described by Heyms eld et al. [27] Knee range was measured by 2 repeated measurements using a goniometer and with the best result reported.Grip strength was measured as the average of 3 repeated grip measurement on a dynamometer using the dominant hand.The six-meter gait speed test was used to measure gait speed by using the best time in seconds to nish a 6-m walk along a straight line using usual walking speed and the average value was used for analysis.
De nition of Sarcopenia Sarcopenia was de ned according to the Asian Working Group for Sarcopenia (AWGS) algorithm [28].A person who has low muscle mass, low muscle strength, and/or low physical performance was categorized as having sarcopenia.Low muscle mass was de ned as ASM index (ASM/height 2 ) < 7.0 kg/m 2 for men and < 5.4 kg/m 2 for women; low muscle strength was de ned as grip strength < 26 kg for men and < 18 kg for women; and low physical performance as gait speed < 0.8 m/s for both men and women.Outcomes Assessments were consecutively conducted within one month before TKA (baseline), 6 months (posttreatment), and 12 months postoperatively.Primary outcome measures were regularly examined via DXA in lean muscle mass, handgrip dynamometer at upper extremity strength, range of motions (ROM) in terms of knee joints motion ( exion/extension) and 6-meter gait speed test at lower extremity functions.DXA measurement values were used to produce the Lean Mass Index (LMI) which is de ned as the ratio of total lean mass (soft tissue only, excluding bone) to height-squared and the Appendage Lean Mass Index (ALMI) which is de ned as the ratio of lean mass on limbs to height-squared.
Secondary outcomes measures were measured by Quality of Life (QOL) measurements in psychological and physical health.Pain, stiffness and physical functions of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) indicated scores ranging from zero to 100, with higher scores representing greater associable disability functions.Medical Outcomes Study Short Form 12 Health Survey Version 2 (SF-12v2) were administered to approximate the general health status in subjects as means to compute a Physical and Mental Health Composite Score (PCS & MCS) that range from a scale of zero to 100, indicated by the higher the score the better level of health.International Physical Activity Questionnaire (IPAQ) instructed an internationally comparable record of health-related physical activity, used to monitor changes of the amount or types of exercise performance level over the research period.Physical activity levels in terms of IPAQ were categorised as "low", "moderate" and "high" and the categorization followed the standard criteria [29,30].The contraposition of SF-12v2 and IPAQ indexes across research timelines allowed meaningful interpretation of bodily and psychological functional uctuation over the effect of TKA on sarcopenia symptom.

Statistical analysis
Demographic statistics on age, sex, BMI, and length of hospital stay were reported in terms of mean ± SD or frequencies where appropriate (Table 1 1.

Primary outcomes (Muscle Mass & strength)
After undergoing total knee arthroplasties, there was a statistically signi cant improvement in walking speed in both sarcopenic and non-sarcopenic patients as evident by reduced time in the six-meter gait speed test (10.24± 5.35 (baseline) to 7.69 ± 2.68 (12 months), p < 0.01) (Table 2) (Supplemental Fig. 1).These were accompanied by improvement trend in muscle mass in both sarcopenic and non-sarcopenic patients at 12 months (The Lean Mass Index (LMI) in sarcopenic: 12.93 ± 1.27 (baseline) to 13.27 ± 1.3 (12 months), p = 0.14; LMI in non-sarcopenic: 14.96 ± 1.83 (baseline) to 15.42 ± 2.01 (12 months), p = 0.06) (Table 3).After controlling for possible confounders, it was found that sarcopenic female that were overweight or obese had statistically signi cant improvement in both ALMI (age ≤ 75: p = 0.04 and age > 75: p = 0.05) and LMI (age ≤ 75:p = 0.04 and age > 75:p = 0.04) after total knee arthroplasties in 12 months (Table 4) (Supplemental Fig. 2).Nevertheless, despite the increase in muscle mass after TKA, both the ALMI and LMI in sarcopenic subjects remained lower than non-sarcopenic subjects at twelve months with statistical signi cances (Table 5).There was no change in handgrip power before and after TKA and subsequent follow-up (p = 0.97) (Table 2).However, there was a signi cant improvement in exion and extension range of the operated knee (p = 0.03 and p < 0.01).Secondary outcome measures (Quality of Life) (Table 2) of life measured with WOMAC self-reporting tool revealed progressive signi cant improvement across subscales scoring at the pain, stiffness and physical function (p < 0.01).SF12v2 being the interpretable scale for health indicated signi cant PCS within this study as values increase with times under post-surgical care from TKA (p < 0.01).In conjunction with this trend, percentage distributions of IPAQ ratings showed increased engagement of high-intensity activities (Supplemental Fig. 3).

Adverse events
No adverse event was noted during this study.

Discussion
Our study illustrates a high prevalence of sarcopenia among patients with end-stage OA of the knee.There were 58 patients entered at baseline, of which 19 (32.8%) had sarcopenia and 39 (67.2%) were not.The prevalence of sarcopenia in Asia ranged from 6.7-18.6% in older men and 0.1-23.6% in older women according to various reports from Japan, Taiwan, Hong Kong and Korean [31][32][33][34].However, it has also been found that the prevalence of sarcopenia among community-dwelling elderly with OA is near three times that of those without OA and this possibly explains the relatively high prevalence of sarcopenia among our OA subjects [35].
This study demonstrated that total knee arthroplasties can bene t patients with severe knee OA with or without co-existing sarcopenia by improving knee function and symptoms, in turn enhancing their gait speed and potentially lean muscle mass.It is the de cit in gait speed and lean muscle mass which are the core components that de ned sarcopenia.According to the latest review in Lancet on sarcopenia, physical activity is regarded as the primary treatment of sarcopenia while there is currently no speci c drugs approved for the treatment of sarcopenia [36].Our study illustrates the importance of identifying sarcopenic patients with concomitant joint disease and managed accordingly to facilitate them having physical activity as the treatment of sarcopenia.At the end of this study, ve sarcopenic patients at baseline turned non-sarcopenic, leading to a total of 44 patients without sarcopenia (75.9%).However, our results also showed that knee arthroplasty alone cannot allow sarcopenic subjects to pick up the overall difference in average lean muscle mass compared to non-sarcopenic subjects.This highlights the importance of managing sarcopenia through a multimodal approach, for example, a combination of high protein diet, knee arthroplasties, and followed by supervised exercise program which by then should be more effective as the physical limitation by knee osteoarthritis has been alleviated.
Having a supervised exercise program is especially important for these groups of patients as they are adapted towards a sedentary lifestyle to cater the pain and weakness induced by osteoarthritis and/or sarcopenia and they tend to continue the same comfortable, sedentary lifestyle if without proper encouragement and training [36].In our study, the patients would only receive standard physiotherapy designed for rehabilitation of knee arthroplasty surgery.The physiotherapy aims at regaining the knee range and walking ability of patients after surgery but does not target building skeletal muscle strength and mass as in those resistance exercise program for sarcopenia.This allowed us to observe the isolated effect of knee arthroplasty on muscle mass and function.However, further studies are warranted to investigate any additive effect of supervised resistance exercise program after knee arthroplasty on sarcopenia.Having said that, some of these OA patients were elderly with low motivation and possibly content with a pain-free knee without further interest to participate in subsequent endurance muscle training.As such some passive physical intervention or "exercise mimetics" like neuromuscular electrical stimulation or whole-body vibration can be considered for those elderly who are unwilling or unable to do physical exercise given their frailty.[37] [38] In fact, whole-body vibration has been shown to increase knee extensor strength and decrease lower leg swelling after TKA and thus worth further investigation on their combined effect on sarcopenia [39].
Previous studies have reported on the negative impact of sarcopenia on surgical outcome.For example, sarcopenia has been identi ed as risk factors for morbidity and mortality in colorectal surgery and gastric cancer surgery, and also a risk factor for prosthetic infection after joint arthroplasty [40][41][42].In our study, no increase in infection rate nor other complications were found, nevertheless, the occurrence of late infection and late complications are beyond our study period.One important difference between the current study and the previous researches on sarcopenia with surgery is that those surgeries mainly induce a catabolic status in the patients while knee arthroplasty induces catabolism in early phase followed by anabolism due to patient regaining their mobility and ability to exercise.This phenomenon could also be a possible explanation of the signi cant improvement of lean mass in overweight or obese sarcopenic female in our study as they lost fat and weight during the initial catabolism after arthroplasty and built up muscle, made possible by better walking ability and less bodyweight hindering movement, during their subsequent rehabilitation [20,35,43].In essence, knee arthroplasty helps break the vicious cycle of immobility, obesity and sarcopenia by returning these individuals to normal lower limb function.
There are certain limitations to our study.As mentioned previously, all patients receive standard physiotherapy in the early phase for post-op rehabilitation.Afterwards, we did not restrict or prescribe further exercise to patients and each of them may engage in variable degrees of exercise.This could contribute to variable improvement in muscle mass among our patients.Similarly, although we encourage our patients to have high protein intake according to dietitian advice, we could not control the exact patients' diets at home and those having a relatively higher protein diet may have better muscle mass building than their counterpart [36].

Conclusions
To conclude, our study showed that sarcopenia among patients with end-stage OA of the knee is not uncommon.Total knee arthroplasty can provide a signi cant reduction in pain, stiffness and function.A positive trend of lean muscle mass increase was observed and was signi cant in the sarcopenic female that were overweight or obese.Further studies focusing on this group of patient and employment of multimodal therapy of knee arthroplasty with a supervised exercise program is warranted.

Declarations
Ethics approval and consent to participate: -Ethical was obtained from the ethics review board of the Joint NTEC/CUHK Ethics Committee (Research Ethics Committee approval number: 2015.539).

Consent for publication:
-Written informed consent was obtained from every participant.
This study was conducted in compliance with the Declaration of Helsinki and was approved by The Joint Chinese University of Hong Kong -New Territories East Cluster Clinical Research Ethics Committee (Ethics approval number: 2015.539).Study registration was made with the US ClinicalTrials.gov(NCT03579329).This prospective study was conducted at the Prince of Wales Hospital, Hong Kong from 1st November 2015 to 30th May 2018.Consecutive patients visiting the Orthopaedics Specialist Outpatient Clinic with symptomatic end-stage OA of the knee on queue for scheduling TKA procedures are entitled to participate in the study.Radiographic severity of knee OA was assessed and documented based on the Kellgren and Lawrence classi cation [26].Clinical diagnosis of knee OA was based on medical history and clinical examination of knee joints.

Table 2
Longitudinal comparisons of SF12v2, WOMAC, IPAQ, and Range of motions of all patients PCS: Physical component score; MCS: mental component score; IPAQ: International Physical Activity Questionnaires

Table 3
Longitudinal comparisons of Appendage Lean Mass Index and Lean Mass Index in patients with or

Table 4
Longitudinal comparisons of Appendage Lean Mass Index and Lean Mass Index in patients with and without sarcopenia * Statistical signi cance using ANOVA Appendage Lean Mass Index (ALMI): Appendage lean mass/Height 2 ;: Lean Mass Index (LMI): Total lean mass/Height 2 * Statistical signi cance using ANOVA Appendage Lean Mass Index (ALMI): Appendage lean mass/Height 2 ;: Lean Mass Index (LMI): Total lean mass/Height 2

Table 5
Cross-sectional comparisons of Appendage Lean Mass Index and Lean Mass Index between patients with and without sarcopenia in the 3 time points