Is there an optimal age for total knee arthroplasty?: A systematic review

Purpose The purpose of this systematic review was to elucidate the optimal age for patients undergoing total knee arthroplasty (TKA), to optimize the balance between the benefits and risks by analyzing patient-reported outcome measurements (PROM), revision rate, and mortality according to age. Materials and methods A rigorous and systematic approach was used and each of the selected studies was evaluated for methodological quality. Data were extracted according to the following: study design, patients enrolled, patient age at the time of surgery, follow-up period, PROM, revision rate, and mortality. Results Thirty-nine articles were included in the final analysis. The results were inconsistent in the PROM analysis, but there was consensus that PROM were good in patients in their 70s . In the revision rate analysis, there was consensus that the revision rate tends to increase in TKA in younger patients, but no significant difference was observed in patients > 70 years of age. In the mortality analysis, there was consensus that the mortality was not significantly different in patients < 80 years of age, but tended to increase with age. Conclusion This systematic review shows that the PROM were good when TKA was performed in patients between 70 and 80 years of age; the best PROM could be achieved around 70 years of age, and no significant difference in the revision or mortality rates was observed between 70 and 80 years of age; however, mortality tended to increase with age. Therefore, the early 70s could be recommended as an optimal age to undergo TKA. Supplementary Information The online version contains supplementary material available at 10.1186/s43019-020-00080-1.


Introduction
Total knee arthroplasty (TKA) is generally accepted as a cost-effective and successful treatment option for endstage knee osteoarthritis (OA) [1]. The prevalence of OA is expected to increase in the future and the use of TKA will be expanded along with increased life expectancy, emphasis on quality of life, and implant development. Therefore, there is a possibility that this will result in an increased need for TKA. This also raises the possibility of increased uptake of TKA in younger and older patients ("extreme" age groups) [2,3]. Therefore, TKA in the extreme age groups could proportionally increase as the volume of TKAs performed increases [4].
TKA can reduce pain and improve patient-reported outcome measures (PROM) and ability to perform activities of daily living. However, TKA can also be accompanied by unexpected complications such as bleeding, acute kidney injury, postoperative delirium, venous thromboembolism, pneumonia, cardiovascular complication, and infection [5][6][7][8][9][10][11][12][13]. High mortality and morbidity are more frequently observed in older patients [14]. Another important consideration in TKA is the longevity of the implants. Long-term survivorship of a TKA implant up to 20 years after surgery was reported as 97.8% [15]. Considering the average age of the patients, there can be increased need of revision TKA in a younger patient due to aseptic loosening, implant wear, and other reasons related to longevity.
Generally, surgeons are concerned about the outcomes after TKA such as PROM, pain reduction, and patient satisfaction. Moreover, surgeons are also concerned about complications. Risk of revision and mortality are the most important considerations in decision-making when performing TKA. In particular, when performing TKA in the extreme age groups, surgeons are concerned about the risk of revision in younger patients and medical comorbidity and mortality in older patients. In recent studies, it is reported that TKA is a good treatment option for knee OA in the extreme age groups, that is for patients age > 90 years or < 55 years [16][17][18][19].
However, the impact of age on patient satisfaction is still debated even though the incidence of TKA uptake among younger (< 55 years) and geriatric (> 80 years) patients is increasing [18,20]. Some studies have shown good treatment results for TKA even when performed at extreme ages. However, the results of these studies were not analyzed by age [16][17][18]21]. Therefore, a better understanding of the effect of age on TKA outcomes, considering the balance between the benefits and risks of TKA, can improve the outcome and facilitate better control of patient expectations [22]. The purpose of this systematic review was to elucidate the optimal age for performing TKA that optimizes the balance between the benefits and risks of TKA, by analyzing PROM results, revision rate, and mortality according to age.

Search strategy
To verify the research question, a rigorous and systematic approach conforming to the preferred reporting items for systematic review and meta-analysis (PRISMA) guidelines was used [23]. In phase 1 of the PRISMA search process, selected databases were searched, including the MEDLINE, EMBASE, and Cochrane database (31 March 2019). This systematic review of the available literature was performed using the keywords: "total knee arthroplasty", "total knee replacement", "age factor", "aged", "young", "extreme age", "old", "octogenarian", "nonagenarian", "treatment outcome", "revision", "mortality", in several combinations. The citations in the included studies were screened, and unpublished articles were also checked with a manual search using Google Scholar. The bibliographies of the relevant articles were subsequently cross-checked for articles not identified in the search. In phase 2, abstracts and titles were screened for relevance. In phase 3, the full text of the selected studies was reviewed according to the inclusion criteria and methodological appropriateness was determined using a predetermined question. In phase 4, the studies were systematically reviewed, if appropriate.

Eligibility criteria
Studies meeting the following criteria were included: (1) studies on TKA, (2) articles written in English, (3) articles with full text available, (4) human in vivo studies, (5) articles including PROM or revision rate or mortality, and (6) comparative study of results according to age. The exclusion criteria were the following: (1) not related to TKA, 2) no direct comparison according to age category, (3) published before 2000, (4) not a clinical study (review article), (5) TKA not performed for treatment of OA, (6) simultaneous evaluation of TKA and total hip arthroplasty, and (7) not a comparison study by age.

Data extraction
Each of the selected studies was evaluated for methodological quality by two independent authors. Data were extracted using the following standardized protocol: first author, publication year, publication journal, study type, number of cases, age of the patient at the time of surgery, follow-up period, PROM, revision rate, and mortality, among others. The extracted data were then crosschecked for accuracy, and any disagreements were settled by a third author.

Quality assessment
The methodological quality of the studies was assessed using the modified Coleman criteria (Additional file 1) [21]. The modified Coleman criteria have a scaled potential score ranging from 0 to 100. Scores of 85-100 are considered excellent, 70-84 good, 55-69 fair, and < 55 poor. The criteria are used to assess the quality of surgical studies.

Search
The initial electronic search yielded 3337 articles. After removing duplicate studies, and applying the inclusion and exclusion criteria, 39 articles were included in the final analysis. Some articles were studies based on registry data, some involved retrospective cohorts, some enrolled prospective cohorts, and some were case-control studies. The PRISMA flow chart is shown in Fig. 1.

Quality
The quality of all articles was assessed using the modified Coleman criteria [21]. The studies included and the modified Coleman criteria scores are presented in Table 1. The average modified Coleman criteria score of the studies we analyzed was 56.7, and the scores were good in 3 of the studies, fair in 20 studies, and poor in 16 studies.

PROM
Twenty-two of the studies reviewed provided data based on PROM outcomes. The results were inconsistent and are presented in Table 2. Among the 22 studies, age was not related to PROM in 9 studies but differed according to age in another 9 studies.
In studies where no differences in PROM were reported, patients age 75, 80, and 85 years were used as age-related references [33,35,40,44,48,50,53]. PROM were compared in regression analysis in another two studies, and the authors reported no age-related differences [31,47].
Among the nine studies in which PROM differed according to age, two studies reported that better PROMs were achieved after TKA in older patients [29,54] and five studies reported that younger patients had better PROM after TKA [26,28,36,57,60]. However, the baseline age of the patients was 80 years in two of the studies, and in one study the patients' limitation of activity increased fourfold over the age of 80 years, and it was difficult to compare the differences in PROM in patients age < 80 years. In two studies only, regression analysis showed that physical activity decreased as age increased [28,36]. Pitta et al. [25] reported that the best PROM was achieved at 68 years of age, and Elmallah et al. [34] reported that the effects of age on Knee Society scores, the Short Form-36 findings, and the lower extremity activity scale were different.
Four studies reported outcomes in the extreme age groups. In the very oldest patients, only one study reported on PROM after TKA: there was no difference between nonagenarians and younger patients in the degree of improvement in PROM [37]. In the very youngest patients, three studies reported PROM after TKA; in all three there was a relatively smaller improvement in clinical outcome in patients < 55 years of age [19,27,32].

Revision rate
Nine studies in this review provided data on revision rates, and the results are presented in Table 3. Eight studies reported that younger patients were more likely to undergo revision until death, and one study reported no difference in revision rates according to age [24,30,38,41,45,49,55,56].
The baseline age of the patients was 65 years in four studies, and high revision rates were reported in the younger group in these studies [38,41,45,49]. Similar results were reported in another study in which the baseline age of the patients was 55 years. Bayliss et al. [30] also reported that the younger age group had higher revision rates, with the lowest implant survival rates seen in patients in their 50s at the time of index surgery and decrease in revision rates seen after 70 years of age. Meehan et al. [45] reported that the revision     [56] also reported that the revision rate was 5 times higher in patients < 55 years of age and 2 times higher in patients 55-64 years of age compared to that noted in patients > 65 years of age. Wainwright et al. [55] reported that patients < 50 years of age at the time of surgery have a greater chance of requiring revision surgery than of dying, and those around the age of 58 years have a 50:50 chance of requiring revision. In addition to group comparisons, Namda et al. [49] reported a 38% reduction in revision rates with every 10-year increase in age.
Only one study reported no significant difference in revision rates between patients < 55 and ≥ 55 years of age [35]. However, the median follow-up period was 12 years, therefore, the revision rates thereafter could not be confirmed.

Mortality
Mortality outcomes are presented in Table 4. Fourteen studies in this review provided data on mortality. Among the 14 studies, 12 reported high mortality rates in older patients, whereas 2 studies reported no difference in mortality rates according to age or that younger patients had increased mortality.
Mortality among patients > 90 years of age was reported in three studies [37,39,43]. Two studies reported higher mortality rates in this age group than that in the control group [39,43], but in another study, mortality rates were higher in the nonagenarian group; however, these were in accordance with life expectancy projections identified by the Office for National Statistics [37].
In four studies, there was an increase in mortality rates with age [42,52], but there was a sharp rise at around 85 years of age [46,51]. In three studies, the mortality rate in patients ≥ 80 years of age was higher than that in the control groups [48,59,60]. In another study, the mortality rate in patients ≥ 65 years of age was higher than that in patients < 65 years of age [20].

Discussion
The purpose of this study was to elucidate the optimal age to perform TKA when considering PROM, revision rate, and mortality factors. Based on several studies, the principal findings were as follows: (1) there was no significant difference in the PROM before the age of 80 years, and it is best to perform TKA around 70 years of age; (2) there was no significant difference in the TKA revision rate in patients older than 70 years, but the rate tended to decrease with age; and (3) there was no significant difference in mortality at the age of 80 years, but it tended to increase with age. Therefore, it is considered that TKA should be performed in patients in their early 70s because the PROM would be relatively good, the revision rate would not increase, and the risk of mortality would not be high.
In terms of PROM, the effects of age were inconsistent. Some studies reported that age was not related to PROM, while some reported that older patients have better PROM, and other studies reported the opposite. In addition, some studies reported a relationship between age and PROM, but it was not a linear relationship. However, the age standard was around the 70s in studies that reported that older patients have better PROM and around the 80s in studies that reported that younger patients have better PROM. In a study that found no linear relationship, the best PROM were reported in patients in their 70s. Therefore, there was consensus that PROM were good between 70 and 80 years of age and the best PROM could be achieved in patients in their 70s. Even among the very oldest patients, those > 90 years of age also had good PROM; however, this age range was not considered to be optimal for performing TKA [17,18]. A previous meta-analysis showed good results even among patients < 55 years of age, but this study did not compare the results according to age.
Most studies were consistent on revision rates, showing mostly that the younger the patient at the time of TKA, the greater is the probability of revision during their lifetime, and most studies compared the revision rate based on 65 or 70 years of age as the reference standard. Therefore, there was consensus that the revision rate tends to increase in younger patients, but there is no significant difference in patients > 70 years of age.
Most studies were consistent on mortality, showing mostly that the older the patient at the time of TKA, the higher is the risk of mortality. Most studies that reported high mortality rates in older patients used the 90s or 80s as the standard age. Only one study compared mortality at the age of 65 years. Some studies did not show an increased mortality rate among these patients in contrast to that in the general population; one study reported that younger patients have higher risk of mortality compared to older patients, but this may be due to selection bias [37,58]. Therefore, there was consensus that the mortality rate was not significantly different at 80 years of age, but tended to increase with age. In summary, TKA performed between the ages of 70 and 80 years has the best outcome. With respect to mortality, it would be better to perform TKA when the patients are younger. Therefore, the authors of these studies believe that from 70 to 80 years of age is the optimal range for undergoing TKA.
There are many factors that are influenced by age when performing TKA. Older age is the predictive factor for postoperative pneumonia and for postoperative delirium after TKA [8,[61][62][63]. Some studies report that the risk of infection is high in older patients [45,64]. The transfusion rate and ICU care are also age-related factors [9,12,13,65]. These complications should also be considered in determining when to perform TKA. In general, however, PROM, revision rate, and mortality are the most common considerations in determining when to perform TKA by considering the patients age. The strength of this study is that the authors only considered patient benefit, revision risk, and mortality when evaluating the outcomes of TKA according to the patient's age. This study has several limitations. First, there have been many studies using registry data; however, only some small cohort or comparative study was included. However, the results of these studies were mostly consistent. Second, meta-analysis was not performed due to differences in the age-related criteria used in the studies. Nonetheless, this did not influence the results significantly, because we did not analyze the exact age, but the age-related trends. Third, the possibility of errors due to different follow-up duration and PROM measurement indices in each study cannot be ignored. Fifth, life expectancy differs in each country, and thus, comparison of the results based on specific country may not be possible. Furthermore, we did not take into account the increase in life expectancy, which is another limitation of this study.

Conclusion
This systematic review shows that the PROM were good when TKA was performed in patients between 70 and 80 years of age; the best PROM could be achieved around 70 years of age, and no significant difference in the revision rate and mortality rate was observed between 70 and 80 years of age; however, mortality after TKA tended to increase with age. Therefore, the early 70s could be recommended as an optimal age to undergo TKA.
Additional file 1. Modified Coleman Criteria used of quality assessment of studies.