Abstract
Aims
The primary aim was to assess whether preoperative health-related quality of life (HRQoL) was associated with postoperative mortality following total hip arthroplasty (THA) and knee arthroplasty (KA). Secondary aims were to assess whether patient demographics/comorbidities and/or joint-specific function were associated with postoperative mortality.
Methods
Patients undergoing THA (n = 717) and KA (n = 742) during a one-year period were identified retrospectively from an arthroplasty register. Patient demographics, comorbidities, Oxford score, and EuroQol five-dimension (EQ-5D) were recorded preoperatively. Patients were followed up for a minimum of seven years and their mortality status was obtained. Cox regression analysis was used to adjust for confounding.
Results
During the study period, 111 patients (15.5%) undergoing THA and 135 patients (18.2%) undergoing KA had died at a mean follow-up of 7.5 years (7 to 8). When adjusting for confounding, the preoperative EQ-5D was associated with postoperative mortality, and for each 0.1 difference in the utility there was an associated change in mortality risk of 6.7% (p = 0.048) after THA, and 6.8% (p = 0.047) after KA. Comorbidities of connective tissue disease (p ≤ 0.026) and diabetes (p ≤ 0.028) were associated with mortality after THA, whereas MI (p ≤ 0.041), diabetes (p ≤ 0.009), and pain in other joints (p ≤ 0.050) were associated with mortality following KA. The preoperative Oxford score was associated with mortality, and for each one-point change in the score there was an associated change in mortality risk of 2.7% (p = 0.025) after THA and 4.3% (p = 0.003) after KA.
Conclusion
Worse preoperative HRQoL and joint specific function were associated with an increased risk of postoperative mortality. Both HRQoL and joint-specific function decline with longer waiting times to surgery for THA and KA and therefore may result in an increased postoperative mortality risk than would have been expected if surgery had been undertaken earlier.
Cite this article: Bone Jt Open 2022;3(12):933–940.
Take home message
Health-related quality of life and joint specific function were associated with postoperative mortality risk, and therefore optimizing preoperatively may have benefits for the patient.
Introduction
Total hip arthroplasty (THA) and knee arthroplasty (KA) are associated with an improvement in health-related quality of life (HRQoL) following surgery.1 The improvement in HRQoL over the patient’s lifetime results in THA and KA being some of the most cost-effective interventions in medicine due to the quality of life gained according to the cost of treatment.1 However, conversely a delay to surgery is proven to be associated with a clinically significant worsening of a patients HRQoL.2 The postoperative effects of this deterioration while waiting for THA or KA are not clear. Previous studies have shown that the deterioration while awaiting surgery is not regained following surgery (i.e. they do not have a greater improvement to attain the same HRQoL, they would have had should they and had surgery earlier).1 Therefore, the deterioration while waiting would seem to have longer lasting effects on the patients HRQoL.
COVID-19 disrupted healthcare services and has contributed to millions of patients waiting for planned hospital treatment. This has resulted in patients waiting longer for THA and KA and the effects of this are not benign.2,3 Approximately a third of patients awaiting THA and a quarter of patients awaiting KA are in a health state “worse than death”, with significantly worse HRQoL than that observed prior to COVID-19 pandemic.2 To the authors knowledge, only one study has assessed the effect for preoperative HRQoL on postoperative mortality in patients following THA, which demonstrated a worse HRQoL was associated with increased risk of mortality at a mean follow-up of 2.4 years.4 There is also data from hip fracture patients that their preoperative HRQoL was predictive of postoperative mortality risk at up to five years post-injury.5,6 There are currently no available data in relation to preoperative HRQoL and mortality following KA. Therefore the longer-term consequences, beyond five years, of preoperative HRQoL on mortality following THA or KA are not known. Furthermore, the effective of joint specific function prior to surgery on mortality are also not known. It is recognized that there is a deterioration in HRQoL and joint-specific function with increasing waiting time to THA and KA, and if this is associated with an increased postoperative mortality risk this would have repercussions on waiting time targets and prioritization for patients.7
The primary aim of this study was to assess whether preoperative HRQoL was associated with postoperative mortality following THA and KA. Secondary aims were to assess: 1) patient demographics/comorbidities; or 2) whether joint-specific function were associated with postoperative mortality. The null hypothesis was that a clinically significant deterioration in HRQoL was not associated with an increased mortality risk.
Methods
Ethical approval was obtained for data collection (Scotland B REC and the number 20/SS/0125 A), and the study was registered and approved at the Royal Infirmary of Edinburgh and University of Edinburgh, UK. A retrospective single centre study was undertaken. Patients undergoing THA or KA during 2014 were identified from an arthroplasty outcome register held at the study centre. There were 717 patients undergoing THA and 742 underwent KA, and had preoperative patient-reported outcome measures (PROMs). Only patients undergoing primary surgery were included. Simultaneous and sequential bilateral procedures, and those undergoing a revision THA and KA during the study period were excluded. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for reporting observational studies were followed (Supplementary Material table i).8
Patient demographics, BMI comorbidities, and PROMs were collected prospectively at the time of the patients preassessment for their surgery (two to four weeks prior). Comorbidities recorded were myocardial infarction (MI), chronic obstructive pulmonary disease (COPD), connective tissue disease (CTD), diabetes, stroke, liver disease, back pain, and other joint pain. These were prospectively recorded at the preassessment clinic by the patient. There were simply recorded as categorical (yes/no) and were not graded on severity.
Outcomes
HRQoL was assessed using the EuroQol five-dimension (EQ-5D) general health questionnaire (primary outcome measure), which evaluates five domains: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression.9 The three-level version of the EQ-5D questionnaire was used (EQ-5D-3L), with responses to the five domains recorded at three levels of severity (no/slight problems, moderate/severe, or unable/extreme problems).10 Permission was obtained from the EuroQol Research Foundation (Netherland) to use the UK version of the EQ-5D-3L. This index is on a scale of -0.594 to 1, where 1 represents perfect health and 0 represents death. Patients scoring less than zero for the EQ-5D score were defined to be in a state WTD.11 The minimal clinically important difference (MCID) in the EQ-5D score after THA is 0.08; therefore, a difference in the score of 0.08 or more was defined as clinically important.12 The EQ visual analogue scale (VAS) was also completed that assesses how good or bad the patient’s health is on that day, and ranges from 100 (best health) to 0 (worst health).
Joint-specific function was assessed using Oxford Hip Score (OHS)13 or Oxford Knee Score (OKS),14 according to the joint being replaced. The Oxford score consists of 12 questions (specific to the joint being assessed) that are scored from 0 (worst) to 4 (best), with a total score ranging from 0 (worst) to 48 (best).15 The MCID in the OHS and OKS is five points.16 Joint-specific pain was also measured using a VAS that ranged from 100 (no pain) to 0 (worst worst).
Mortality
Patient mortality status was obtained from the local (study centre) hospital electronic records which is the sole provider for national health care for the catchment population. This was done using each patient’s unique Community Health Index (CHI) number. Furthermore, to ensure capture of patients that may have moved out with the catchment area the National Records of Scotland were contacted to assess the patients mortality status.17
Statistical analysis
Statistical analysis was performed using Statistical Package for Social Sciences (SPSS) software version 17 (IBM, USA). Parametric tests were used to assess continuous variables for significant differences between groups (alive vs dead) using an unpaired t-test (age, BMI, EQ-5D, EQ-VAS, OHS, OKS, and pain VAS) were assessed for differences. Dichotomous variables were assessed using a Chi square test between group comparisons (sex and comorbidity). Kaplan-Meier time to event methodology was used to assess patient survival following TKA or KA. Log rank (Mantel-Cox) test was used to assess differences in survival between THA and KA. Cox regression analysis was used to assess the independent association of factors influencing patient mortality when adjusting for confounding variables. Due to the co-linearity (r > 0.7) of the EQ-5D (primary outcome) and the Oxford scores (secondary outcome), two separate models were undertaken. Only variables that were significant or were treading towards significance (p < 0.1) for mortality on unadjusted analysis were included in the models (Table I and Table II). A p-value < 0.05 was defined as statistically significant.
Table I.
Variable | Overall (n = 717) | Alive (n = 606) | Dead (n = 111) | Difference/odds ratio | p-value |
---|---|---|---|---|---|
Age (years: mean, SD) | 66.6 (12.9) | 64.9 (12.7) | 76.0 (9.5) | 11.1 (8.6 to 13.7) | < 0.001† |
Sex, n (%) | |||||
Male | 320(44.6) | 271 (44.7) | 49 (44.1) | 1.02 (0.7 to 1.5) | 0.911* |
Female | 397(55.4) | 335 (55.3) | 62 (55.9) | ||
Mean BMI, kg/m2 (SD) | 29.8 (17) | 30.3 (12.7) | 27.5 (9.5) | 2.8 (0.9 to 6.4) | 0.144† |
Comorbidities, n (%) | |||||
MI | 37 (5.2) | 31 (5.1) | 6 (5.4) | 1.2 (0.5 to 2.9) | 0.727* |
COPD | 39 (5.4) | 28 (4.6) | 11 (9.9) | 2.5 (1.2 to 5.1) | 0.014* |
CTD | 85 (11.9) | 63 (8.8) | 22 (19.8) | 2.3 (1.3 to 3.9) | 0.003* |
Diabetes | 56 (7.8) | 40 (6.6) | 16 (14.4) | 2.6 (1.4 to 4.9) | 0.002* |
Stroke | 25 (3.4) | 21 (3.5) | 4 (3.6) | 1.1 (0.4 to 3.3) | 0.847* |
Liver disease | 16 (2.2) | 13 (2.1) | 3 (2.7) | 1.4 (0.4 to 5) | 0.630* |
Back pain | 390 (54.4) | 328 (54.1) | 62 (55.9) | 1.3 (0.8 to 2) | 0.246* |
Other joint pain | 516 (72) | 434 (71.6) | 82 (73.9) | 1.4 (0.8 to 2.4) | 0.285* |
HRQoL | |||||
EQ-5D (mean, SD) | 0.348 (0.330) | 0.359 (0.330) | 0.287 (0.330) | 0.072 (0.005 to 0.139) | 0.035† |
EQ-VAS (mean, SD) | 66.1 (23.2) | 66.9 (23.0) | 62.0 (23.6) | 5.0 (0.3 to 9.7) | 0.038† |
Joint-specific function | |||||
OHS (mean, SD) | 20.0 (9.1) | 20.3 (9.1) | 18.3 (9) | 2.0 (0.1 to 3.8) | 0.036† |
Pain-VAS (mean, SD) | 49.6 (24.4) | 50 (24.3) | 47.3 (24.8) | 2.7 (2.2 to 7.7) | 0.278† |
-
*
Chi-squared test.
-
†
Independant t-test
-
COPD, chronic obstructive pulmonary disease; CTD, connective tissue disease; HRQoL, health-related quality of life ; MI, myocardial infarction; OHS, Oxford Hip Score; SD, standard deviation; VAS, visual analogue scale.
Table II.
Variable | Overall (n = 742) | Alive (n = 607) | Dead (n = 135) | Odds ratio (95% CI) | p-value |
---|---|---|---|---|---|
Mean ge, yrs ( SD) | 69.2 (9.8) | 67.9 (9.4) | 75.5 (9) | 7.6 (5.9 to 9.4) | < 0.001† |
Sex, n (%) | |||||
Male | 313 (42.2) | 256 (42.2) | 57 (42.2) | 1 (0.7 to 1.5 | 0.992* |
Female | 429 (57.8) | 351 (57.8) | 78 (57.8) | ||
Mean BMI, kg/m2 (SD) | 30.8 (7.6) | 30.9 (6.2) | 30.7 (12) | 0.1 (1.4 to 1.6) | 0.846† |
Comorbidities, n (%) | |||||
MI | 47 (6.3) | 30 (4.9) | 17 (12.6) | 2.9 (1.5 to 5.5) | 0.001* |
COPD | 33 (4.4) | 21 (3.5) | 12 (8.9) | 2.8 (1.4 to 6.0) | 0.004* |
CTD | 116 (15.6) | 85 (14) | 31 (23) | 2.0 (1.2 to 3.2) | 0.005* |
Diabetes | 116 (15.6) | 84 (13.8) | 32 (23.7) | 2.0 (1.3 to 3.2) | 0.003* |
Stroke | 30 (4) | 16 (2.6) | 14 (10.3) | 4.6 (2.2 to 9.7) | < 0.001* |
Liver disease | 14 (1.9) | 10 (1.6) | 4 (3) | 1.9 (0.6 to 6.2) | 0.275* |
Back pain | 321 (43.3) | 264 (43.5) | 57 (42.2) | 1.0 (0.7 to 1.5) | 0.961* |
Other joint pain | 533 (71.8) | 447 (73.6) | 86 (63.7) | 0.7 (0.4 to 1.1) | 0.088* |
HRQoL | |||||
EQ-5D (mean, SD) | 0.395 (0.309) | 0.407 (0.308) | 0.340 (0.306) | 0.067 (0.009 to 0.124) | 0.023† |
EQ-VAS (mean, SD) | 68.3 (22.2) | 69.4 (22.3) | 63.6 (21.4) | 5.8 (1.6 to 9.9) | 0.006† |
Joint-specific function | |||||
OHS (mean, SD) | 20 (7.9) | 20.5 (7.8) | 17.8 (8.1) | 2.7 (1.2 to 4.1) | < 0.001† |
Pain-VAS (mean, SD) | 52.5 (24.4) | 52.1 (24.3) | 54.4 (25.1) | 2.3 (1.6 to 7.4) | 0.314† |
-
*
Chi-squared test.
-
†
Independant t-test.
-
COPD, chronic obstructive pulmonary disease; CTD, connective tissue disease; HRQoL, health-related quality of life; MI, myocardial infarction; OHS, Oxford Hip Score; SD, standard deviation; VAS, visual analogue scale.
A priori power calculation was performed using the known MCID difference in the EQ-5D of 0.08 and standard deviation of 0.3 (effect size 0.267),12 an α of 0.05, using a 1:6 ratio (assumed 15% mortality at seven- to eight-year follow up would require a cohort of 710 (101 deceased and 609 alive) to achieve a power of 80%.18,19
Results
During the study period, 111 patients (15.5%) undergoing THA and 135 patients (18.2%) undergoing KA had died at a mean follow-up of 7.5 years (7 to 8) (Figure 1). There was no difference in survival between THA and KA (p = 0.181, log rank). The mean age of those undergoing THA was 66.6 years, of which 397 (55.4%) were female, with a mean BMI of 29.8 kg/m2 (Table I). The mean age of those undergoing KA was 69.2 years, of whom 429 (57.8%) were female, with a mean BMI of 30.8 kg/m2 (Table II).
Fig. 1
Primary aim: HRQoL and postoperative mortality
A significantly lower (worse) preoperative EQ-5D (mean difference (MD) 0.072, 95% confidence interval (CI) 0.005 to 0.103; p = 0.035) and EQ-VAS (MD 5.0, 95% CI 0.3 to 9.7; p = 0.038) were associated with postoperative mortality in patients undergoing THA (Table I). When adjusting for confounding, preoperative EQ-5D was associated with postoperative mortality (hazard ratio (HR) 0.52, 95% CI 0.27 to 0.99; p = 0.048), and for each 0.1 difference in the utility there was an associated change in mortality risk of 6.7% (Table III). Similarly, a significantly lower preoperative EQ-5D (MD 0.067, 95% CI 0.009 to 0.124; p = 0.023) and EQ-VAS (MD 5.8, 95% CI 1.6 to 9.9; p = 0.006) were associated with postoperative mortality in patients undergoing TKA (Table II). When adjusting for confounding, preoperative EQ-5D was associated with postoperative mortality (HR 0.52, 0.27 to 0.99; p = 0.047), and for each 0.1 difference in the utility, there was an associated change in mortality risk of 6.8% (Table IV).
Table III.
Variable | B | Hazard ratio* | 95% CI | p-value |
---|---|---|---|---|
Age | 0.099 | 1.104 | 1.076 to 1.132 | 0.000 |
Sex | -0.278 | 0.758 | 0.502 to 1.143 | 0.186 |
BMI | -0.021 | 0.979 | 0.947 to 1.013 | 0.222 |
Comorbidities | ||||
COPD | 0.412 | 1.511 | 0.752 to 3.033 | 0.246 |
CTD | 0.583 | 1.792 | 1.071 to 2.998 | 0.026 |
Diabetes | 0.645 | 1.906 | 1.083 to 3.354 | 0.025 |
EQ-5D | -0.652 | 0.521 | 0.273 to 0.993 | 0.048 |
-
*
Hazard ratio for mortality: 0.1 change = 6.74%.
-
CI, confidence interval; COPD, chronic obstructive pulmonary disease; CTD, connective tissue disease ; EQ-5D, EuroQol five-dimension.
Table IV.
Variable | B | Hazard ratio* | 95% CI | p-value |
---|---|---|---|---|
Age | 0.027 | 1.028 | 0.692 to 1.526 | 0.893 |
Sex | 0.092 | 1.096 | 1.071 to 1.123 | 0.000 |
BMI | 0.016 | 1.016 | 0.994 to 1.038 | 0.160 |
Comorbidities | ||||
MI | 0.704 | 2.021 | 1.080 to 3.784 | 0.028 |
COPD | 0.646 | 1.907 | 0.886 to 4.108 | 0.099 |
CTD | 0.425 | 1.529 | 0.946 to 2.473 | 0.083 |
Diabetes | 0.625 | 1.868 | 1.187 to 2.939 | 0.007 |
CVA | -0.142 | 0.868 | 0.204 to 3.697 | 0.848 |
Other joint pain | -0.403 | 0.668 | 0.444 to 1.006 | 0.050 |
EQ-5D* | -0.658 | 0.518 | 0.271 to 0.991 | 0.047 |
-
*
Hazard ratio for mortality: 0.1 change = 6.80%.
-
CI, confidence interval; COPD, chronic obstructive pulmonary disease; CTD, connective tissue disease; CVA, cerebrovascular accident ; EQ-5D, EuroQol five-dimension; MI, myocardial infarction.
Table V.
Variable | B | Hazard ratio* | 95% CI | p-value |
---|---|---|---|---|
Age | 0.098 | 1.103 | 1.076 to 1.131 | 0.000 |
Sex | -0.309 | 0.734 | 0.485 to 1.111 | 0.144 |
BMI | -0.019 | 0.981 | 0.950 to 1.013 | 0.243 |
Comorbidities | ||||
COPD | 0.468 | 1.596 | 0.804 to 3.169 | 0.181 |
CTD | 0.604 | 1.829 | 1.095 to 3.055 | 0.021 |
Diabetes | 0.633 | 1.883 | 1.071 to 3.312 | 0.028 |
OHS | -0.027 | 0.973 | 0.951 to 0.997 | 0.025 |
-
*
Hazard ratio for mortality: 1 point = 2.74%.
-
COPD, chronic obstructive pulmonary disease; CTD, connective tissue disease; OHS, Oxford Hip Score.
Secondary aim: Patient demographics/comorbidities and postoperative mortality
Comorbidities of COPD, CTD, and diabetes were significantly associated with an increased mortality risk for those undergoing THA (Table I), whereas MI, COPD, CTD, diabetes and stroke were associated with mortality in patients undergoing KA (Table II). When adjusting for confounding, patients with CTD or diabetes were more likely to die following THA (Table III and Table V), whereas patients with a history of MI, diabetes, or pain in other joints were likely to die following KA (Table IV and Table VI).
Table VI.
Variable | B | Hazard ratio* | 95% CI | p-value |
---|---|---|---|---|
Age | -0.056 | 0.946 | 0.634 to 1.411 | 0.784 |
Sex | 0.088 | 1.092 | 1.067 to 1.118 | 0.000 |
BMI | 0.013 | 1.013 | 0.990 to 1.037 | 0.282 |
Comorbidities | ||||
MI | 0.656 | 1.927 | 1.028 to 3.612 | 0.041 |
COPD | 0.717 | 2.049 | 0.971 to 4.325 | 0.060 |
CTD | 0.340 | 1.405 | 0.865 to 2.283 | 0.169 |
Diabetes | 0.600 | 1.822 | 1.158 to 2.866 | 0.009 |
CVA | 0.024 | 1.024 | 0.247 to 4.245 | 0.974 |
Other joint pain | -0.443 | 0.642 | 0.427 to 0.966 | 0.034 |
OKS | -0.042 | 0.959 | 0.932 to 0.986 | 0.003 |
-
*
Hazard ratio for mortality: one point = 4.28%.
-
CI, confidence interval; COPD, chronic obstructive pulmonary disease; CTD, connective tissue disease; CVA, cerebrovascular accident; MI, myocardial infarction; OKS, Oxford Knee Score.
Secondary aim: joint-specific function and postoperative mortality
A significantly lower (worse) preoperative OHS (MD 2.0, 95% CI 0.1 to 3.8; p = 0.036) was associated with postoperative mortality in patients undergoing THA (Table I). When adjusting for confounding preoperative OHS was associated with postoperative mortality (HR 0.97, 95% CI 0.951 to 0.997; p = 0.025), and for each one-point difference in the score there was an associated change in mortality risk of 2.74% (Table V). Similarly, a significantly lower preoperative OKS (MD 2.7, 95% CI 1.2 to 4.1; p < 0.001) was associated with postoperative mortality in patients undergoing TKA (Table II). When adjusting for confounding preoperative OKS was associated with postoperative mortality (HR 0.96, 95% CI 0.93 to 0.99; p = 0.003), and for each one-point difference in the score there was an associated change in mortality risk of 4.28% (Table VI).
Discussion
This study has shown that preoperative HRQoL was independently associated with postoperative mortality up to eight years following THA and KA, with a worse HRQoL being associated with an increased risk of mortality. Patient comorbidities were also shown to be independently associated with postoperative mortality, although only diabetes was a common risk factor for following THA and KA. A novel aspect of the study was that preoperative joint specific function was independently associated with postoperative mortality, with a worse Oxford score being associated with an increased risk of mortality. The crude mortality rate following THA was 15.5% and 18.2% following KA at a mean follow-up of 7.5 years, although this was not significantly different and may be due to the older age of those undergoing KA.
A novel aspect of the current study was demonstrating the association of preoperative HRQoL with mortality risk postoperatively. Although this may be an unsurprising connection, there is a paucity evidence to support this association. To the authors knowledge, the only study assessing HRQoL and mortality after primary THA or KA is that reported by Cnudde et al.4 They reported 42,862 THA from the Swedish hip arthroplasty register at a mean follow-up of 2.4 years. However, rather than assess the effect of the validated EQ-5D utility on mortality, they assessed the effect of each of the five dimensions: mobility, self-care, usual activity, pain/discomfort, and anxiety/depression on mortality. They found that patients with moderate or severe problems had a higher mortality risk. However, this is not how the EQ-5D utility is reported, and it is weighted according to each of the responses given.10 This current study demonstrated that a change in the EQ-5D utility was independently associated with mortality risk up to eight years following THA and KA. Although not specific to arthroplasty, population level data supports the association of worsening subjective health with increased mortality risk.20,21 More specifically, a worse EQ-5D utility when used to measure HRQoL has been shown to be associated with increased risk of hospitalisation and mortality in an elderly Italian population.22 The association of the EQ-5D with mortality has also been shown in the frailer hip fracture population, being a moderate predictor of mortality at 12 months and also at five years,5 but at this later time point, it was not significant when accounting for confounding factors.6
In the knowledge of the effect of HRQoL on mortality after THA or KA, this may have repercussions on prioritising patients for surgery, before their HRQoL deteriorates further, and whether it is possible to optimize patients prior to surgery and improve their HRQoL and therefore lower the risk of postoperative mortality. Due to the prolonged waits for THA and KA due to the disruption of COVID-19 on planned healthcare services, patients waiting six months or more have been shown to have a clinically significant deterioration in their HRQoL.2 Patients are not only are having to endure a deteriorating HRQoL while awaiting their essential surgery, but the results of the current study suggest this may have longer-term repercussions on their mortality. It is recognized that the preoperative EQ-5D of patients currently awaiting THA (0.119) or KA (0.074) have a worse utility compared to patients prior to the COVID-19 pandemic,2 which would equate to an approximate increase mortality risk of 8.1% and 5.0%, respectively. To put this into context, these increased mortality risks are similar to that observed at five years in patients with stage two breast cancer.23 A lower HRQoL is also associated with a longer hospital length of stay and increased perioperative complications,24,25 which may have financial and patient morbidity consequences. Furthermore, patients may not achieve the same level of HRQoL following their THA or KA; they may have done prior to their deterioration and therefore not achieve the same outcome if they had their surgery earlier.1 In the knowledge of these potential ramifications to patients awaiting THA or KA, it would seem inappropriate to prioritize routine arthroplasty a level four in light of the increased mortality risk associated with worsening HRQoL. As priority level four currently suggests that a patient waiting longer than three months will not come to harm.7,26 The ideal resolution may be that all patients receive their surgery within three months of being placed on the waiting list, but this is not likely to be the case for several years.27,28 An additional option is to try and avoid deterioration in, and maybe improve, HRQoL by offering patients waiting for surgery a prehabilitation package of care, which has been shown to improve HRQoL.29,30 This may even help optimize the outcomes of their surgery, which has been observed in surgical cancer patients and can be undertaken remotely.31
MI, diabetes, and CTD were shown to be associated with approximately double the mortality risk during the first eight years following THA or KA, which is consistent with previous studies.32,33 Two systematic reviews by Berstock et al32,33 assessed mortality up to 90 days following THA or KA, and demonstrated a similar hazard ratio for mortality associated with comorbidities of diabetes and cardiovascular disease. There are limited data reporting factors associated with longer-term mortality after THA or KA.18,34 Buchele et al34 assessed a cohort of 809 patients up to 20 years following THA or KA, and demonstrated diabetes and cardiac insufficiency to be associated with increased mortality risk. However, they did not assess the effect of CTD as a group of conditions, but did show that patients with a comorbidity of gout had a higher risk to mortality. More recently, Yapp et al18 demonstrated inflammatory polyarthropathy was associated with a hazard ratio of 1.8, similar to the current study for CTD, for increased risk of mortality at a median follow-up of 7.4 years after KA.
The second novel aspect of the current study was the association of joint-specific function with mortality. To the authors knowledge, this has not been described previously. It was not possible to include the EQ-5D and the Oxford scores in the same regression models due to the observed collinearity between the measures, and therefore the independent effect of each on mortality could not be confirmed. However, it would seem intuitive that deteriorating joint-specific function would be associated with deconditioning of the patient due to increased pain and dysfunction.35 A clinically significant deterioration on the Oxford score of five points,16 while awaiting surgery would be associated with an increased mortality risk of 14.5% for those awaiting a THA and 23.4% for those awaiting a KA.
The results of this study should be interpreted in the knowledge of the limitations. The cohort used to assess effect of HRQoL on mortality was relatively small compared to registry data; however, the mortality rates observed in the current study were similar to larger studies.18,19 The study used the patient’s unique CHI number to ensure follow of the patients not only within the catchment population, but also across the Nation. The effect of social deprivation and smoking were not assessed in the current study which have previously been shown to influence mortality in both the short and longer term.18,32,33 The cause of death was also not assessed as this was not readily available for all patients. To investigate the cause of death associated with those with the worse HRQoL, it may help understand the reason(s) for the increased risk of mortality. Finally, assessment of change in the patients HRQoL following surgery was not taken into account, and whether this was also associated with postoperative mortality should be explored in future studies.
In conclusion, worsening HRQoL and joint-specific function were both associated with an increased risk of postoperative mortality up to eight years following surgery. Both HRQoL and joint specific function worsen with longer waiting time to surgery for THA and KA, and this may result in an increased mortality risk following their surgery and should be further investigated.
References
1. Jenkins PJ , Clement ND , Hamilton DF , Gaston P , Patton JT , Howie CR . Predicting the cost-effectiveness of total hip and knee replacement: A health economic analysis . Bone Joint J . 2013 ; 95-B ( 1 ): 115 – 121 . Crossref PubMed Google Scholar
2. Clement ND , Scott CEH , Murray JRD , Howie CR , Deehan DJ , IMPACT-Restart Collaboration . The number of patients “worse than death” while waiting for a hip or knee arthroplasty has nearly doubled during the COVID-19 pandemic . Bone Joint J . 2021 ; 103-B ( 4 ): 672 – 680 . Crossref Google Scholar
3. Morris JA , Super J , Huntley D , Ashdown T , Harland W , Anakwe R . Waiting lists for symptomatic joint arthritis are not benign: prioritizing patients for surgery in the setting of COVID-19 . Bone Jt Open . 2020 ; 1 ( 8 ): 508 – 511 . Crossref Google Scholar
4. Cnudde P , Nemes S , Mohaddes M , et al. Is preoperative patient-reported health status associated with mortality after total hip replacement? Int J Environ Res Public Health . 2017 ; 14 ( 8 ): E899 . Crossref PubMed Google Scholar
5. Parsons N , Griffin XL , Achten J , Costa ML . Outcome assessment after hip fracture: is EQ-5D the answer? Bone Joint Res . 2014 ; 3 ( 3 ): 69 – 75 . Crossref PubMed Google Scholar
6. Knauf T , Bücking B , Bargello M , et al. Predictors of long-term survival after hip fractures? 5-year results of a prospective study in Germany . Arch Osteoporos . 2019 ; 14 ( 1 ): 40 . Crossref Google Scholar
7. Clement ND , Skinner JA , Haddad FS , Simpson AHRW . Prioritization versus rationing of healthcare – elective surgery is not optional surgery . Bone & Joint Research . 2022 ; 11 ( 5 ): 301 – 303 . Crossref Google Scholar
8. von Elm E , Altman DG , Egger M , et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies . Bull World Health Organ . 2007 ; 85 ( 11 ): 867 – 872 . Crossref PubMed Google Scholar
9. Brooks R . EuroQol: the current state of play . Health Policy . 1996 ; 37 ( 1 ): 53 – 72 . Crossref PubMed Google Scholar
10. Dolan P . Modeling valuations for EuroQol health states . Med Care . 1997 ; 35 ( 11 ): 1095 – 1108 . Crossref PubMed Google Scholar
11. Scott CEH , MacDonald DJ , Howie CR . “Worse than death” and waiting for a joint arthroplasty . Bone Joint J . 2019 ; 101-B ( 8 ): 941 – 950 . Crossref Google Scholar
12. Larsen K , Hansen TB , Søballe K . Hip arthroplasty patients benefit from accelerated perioperative care and rehabilitation: a quasi-experimental study of 98 patients . Acta Orthop . 2008 ; 79 ( 5 ): 624 – 630 . Crossref PubMed Google Scholar
13. Dawson J , Fitzpatrick R , Carr A , Murray D . Questionnaire on the perceptions of patients about total hip replacement . J Bone Joint Surg Br . 1996 ; 78 ( 2 ): 185 – 190 . PubMed Google Scholar
14. Dawson J , Fitzpatrick R , Murray D , Carr A . Questionnaire on the perceptions of patients about total knee replacement . J Bone Joint Surg Br . 1998 ; 80 ( 1 ): 63 – 69 . Crossref PubMed Google Scholar
15. Murray DW , Fitzpatrick R , Rogers K , et al. The use of the Oxford Hip and Knee Scores . J Bone Joint Surg Br . 2007 ; 89 ( 8 ): 1010 – 1014 . Crossref Google Scholar
16. Beard DJ , Harris K , Dawson J , et al. Meaningful changes for the Oxford Hip and Knee Scores after joint replacement surgery . J Clin Epidemiol . 2015 ; 68 ( 1 ): 73 – 79 . Crossref PubMed Google Scholar
17. No authors listed . National Records of Scotland: Preserving the past, recording the present, Informing the future . https://www.nrscotland.gov.uk/ ( date last accessed 24 October 2022 ). Google Scholar
18. Yapp LZ , Clement ND , Moran M , Clarke JV , Simpson AHRW , Scott CEH . Long-term mortality rates and associated risk factors following primary and revision knee arthroplasty : 107,121 patients from the Scottish Arthroplasty Project . Bone Joint J . 2022 ; 104-B ( 1 ): 45 – 52 . Crossref Google Scholar
19. Pedersen AB , Baron JA , Overgaard S , Johnsen SP . Short- and long-term mortality following primary total hip replacement for osteoarthritis: A Danish nationwide epidemiological study . J Bone Joint Surg Br . 2011 ; 93-B ( 2 ): 172 – 177 . Crossref PubMed Google Scholar
20. Burström B , Fredlund P . Self rated health: Is it as good a predictor of subsequent mortality among adults in lower as well as in higher social classes? J Epidemiol Community Health . 2001 ; 55 ( 11 ): 836 – 840 . Crossref PubMed Google Scholar
21. Mossey JM , Shapiro E . Self-rated health: a predictor of mortality among the elderly . Am J Public Health . 1982 ; 72 ( 8 ): 800 – 808 . Crossref PubMed Google Scholar
22. Cavrini G , Broccoli S , Puccini A , Zoli M . EQ-5D as a predictor of mortality and hospitalization in elderly people . Qual Life Res . 2012 ; 21 ( 2 ): 269 – 280 . Crossref PubMed Google Scholar
23. No authors listed . Office for National Statistics cancer survival in England: adults diagnosed . https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/datasets/cancersurvivalratescancersurvivalinenglandadultsdiagnosed ( date last accessed 24 October 2022 ). Google Scholar
24. Piuzzi NS , Strnad GJ , Ali Sakr Esa W , Barsoum WK , Bloomfield MR , Brooks PJ . The main predictors of length of stay after total knee arthroplasty: patient-related or procedure-related risk factors . J Bone Joint Surg Am . 2019 ; 101-A ( 12 ): 1093 – 1101 . Crossref PubMed Google Scholar
25. Ferguson R , Prieto-Alhambra D , Peat G , et al. Does pre-existing morbidity influences risks and benefits of total hip replacement for osteoarthritis: A prospective study of 6682 patients from linked national datasets in England . BMJ Open . 2021 ; 11 ( 9 ): e046712 . Crossref PubMed Google Scholar
26. No authors listed . Clinical guide to surgical prioritisation during the coronavirus pandemic . Royal College of Surgeons of England . https://www.rcseng.ac.uk/coronavirus/surgical-prioritisation-guidance/ ( date last accessed 24 October 2022 ). Google Scholar
27. Oussedik S , MacIntyre S , Gray J , McMeekin P , Clement ND , Deehan DJ . Elective orthopaedic cancellations due to the COVID-19 pandemic: where are we now, and where are we heading? Bone Jt Open . 2021 ; 2 ( 2 ): 103 – 110 . Crossref PubMed Google Scholar
28. Yapp LZ , Clarke JV , Moran M , Simpson A , Scott CEH . National operating volume for primary hip and knee arthroplasty in the COVID-19 era: a study utilizing the Scottish Arthroplasty Project dataset . Bone Jt Open . 2021 ; 2 ( 3 ): 203 – 210 . Crossref PubMed Google Scholar
29. Simpson A , Howie CR , Kinsella E , et al. Osteoarthritis preoperative package for care of orthotics, rehabilitation, topical and oral agent usage and nutrition to improve outcomes at A year (OPPORTUNITY); A feasibility study protocol for A randomised controlled trial . Trials . 2020 ; 21 ( 1 ): 209 . Crossref PubMed Google Scholar
30. Simpson A , Howie CR , Kinsella E , et al. Osteoarthritis preoperative package for care of orthotics, rehabilitation, topical and oral agent usage and nutrition to improve outcomes at A year (OPPORTUNITY); A feasibility study protocol for A randomised controlled trial . Trials . 2020 ; 19 ( 1 ): 209 . Crossref PubMed Google Scholar
31. van Gestel T , Groen LCB , Puik JR , et al. Fit4Surgery for cancer patients during covid-19 lockdown: a systematic review and meta-analysis . Eur J Surg Oncol . 2022 ; 48 ( 6 ): 1189 – 1197 . Crossref Google Scholar
32. Berstock JR , Beswick AD , Lenguerrand E , Whitehouse MR , Blom AW . Mortality after total hip replacement surgery: A systematic review . Bone Joint Res . 2014 ; 3 ( 6 ): 175 – 182 . Crossref PubMed Google Scholar
33. Berstock JR , Beswick AD , López-López JA , Whitehouse MR , Blom AW . Mortality after total knee arthroplasty: a systematic review of incidence, temporal trends, and risk factors . J Bone Joint Surg Am . 2018 ; 100-A ( 12 ): 1064 – 1070 . Crossref PubMed Google Scholar
34. Büchele G , Günther KP , Brenner H , et al. Osteoarthritis-patterns, cardio-metabolic risk factors and risk of all-cause mortality: 20 years follow-up in patients after hip or knee replacement . Sci Rep . 2018 ; 8 ( 1 ): 1 – 8 . Crossref Google Scholar
35. Cisternas AF , Ramachandran R , Yaksh TL , Nahama A . Unintended consequences of COVID-19 safety measures on patients with chronic knee pain forced to defer joint replacement surgery . Pain Rep . 2020 ; 5 ( 6 ): e855 . Crossref PubMed Google Scholar
Author contributions
N. D. Clement: Conceptualization, Data analysis, Methodology, Writing – original draft.
R. F. L. Patton: Investigation, Methodology, Data curation, Writing – review & editing.
D. J. MacDonald: Data Curation, Writing – review & editing.
A. D. Duckworth: Conceptualization, Data analysis, Writing – review & editing.
Funding statement
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
ICMJE COI statement
N. D. Clement reports being and editorial board member for The Bone and Joint Journal and Bone and Joint Research. A. D. Duckworth declares unrelated research grants from the NIHR, OTA, and SORT-IT; unrelated book royalties from Taylor and Francis and Elsevier; and hip fracture lecture at BOA 2021 for AgNovos Healthcare; being an editorial board member for The Bone and Joint Journal, Bone and Joint Research, and Bone & Joint 360; associate editor for Trials, OTA, and JBJS Case Connector; and member of the OTA Research Committee and OTS Research Committee.
Acknowledgements
The authors would like to thank the patients that have contributed their time to this project and completing their questionnaires, the orthopaedic team at Edinburgh Orthopaedics and the University of Edinburgh for their support.
Open access funding
The authors report that they received open access funding for this manuscript from the Department of Orthopaedics at the Univeristy of Edinburgh, Edinburgh, UK.
Follow N. D. Clement @EdinOrthopaedic
Follow A. D. Duckworth @DuckworthOrthEd
Supplementary material
Table showing the STROBE statement: Checklist of items that should be included in reports of cross-sectional studies.
© 2022 Author(s) et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/