Comparing total hip arthroplasty and hemiarthroplasty for the treatment of displaced femoral neck fracture in the active elderly over 75 years old: a systematic review and meta-analysis of randomized control trials

Hip fractures are a worldwide health problem, with the incidence increasing over time. The prevalence of hip fractures in Asian countries will rise from over 1 million per year in 1990 to over 6 million by 2050 [1]. The residual lifetime risk of suffering a hip fracture is about 5.6% for men and 20% for women [2]. The femoral neck fracture (FNF) is a major type of hip fracture, whose treatment includes internal fixation, HA, or THA [3]. Internal fixation is a preferred management option for young people or the elderly who are intolerant of prosthesis surgery [4]. THA and HA are widely used in DFNF elderly [3]. In general, HA has advantages of the shorter operation time, less blood loss, less technical demand, less economic burden, and a lower dislocation rate [5‐9]. THA is associated with better hip function, less acetabulum erosion, and a lower revision rate [5‐10].

With the increase of longevity and activity level, today’s elderly have a higher demand for adequate hip function and a higher risk of acetabulum erosion after arthroplasty than before, which seems to favor the THA procedure [11‐13]. Some RCTs [12‐16] suggested that THA could provide a superior hip function without increasing the mortality rate and general complication rate in patients with a mean age over 75 years old. Current guidelines [17, 18] suggested that THA would provide greater benefit in selected elderly patients because of less pain and lower revision rate, such as patients who were able to walk independently, not cognitively impair, and medically fit for the procedure.

However, the frequency of HA procedures increases with age, whose cutoff is at 76, due to, in part, the contradictions between the more extensive surgery of THA and the lower surgical tolerance of elderly patients [19, 20]. Some authors [6, 21‐23] found that there was no significant difference in hip function and life quality between the two procedures in patients with a mean age over 75 years, with better prosthetic survivorship and a trend of fewer hip complications in the HA group. A systematic review [8] indicated that both procedures are reasonable in patients over 80 years because of the nonsignificant differences in hip function, reoperation rate, and mortality rate.

There are still concerns whether the more substantial surgical trauma and higher dislocation rate would trade off the advantages of THA due to frailty and lower physical demands in the elderly over 75 years [6, 8, 19, 21]. Therefore, the choice of THA or HA in DFNF in this population remains uncertain. The previous meta-analyses did not include strict age restrictions, and some of these studies also included patients who were unable to walk independently or received a polycarbonate–urethane (PCU) THA that would lead to high heterogeneity, therefore making it complicated to apply these findings to patients with older age [7, 8, 24‐27]. The purpose of this study is to compare the effectiveness and safety of THA and HA in the active elderly over 75 years by using the latest evidence from previously performed RCTs.

This review was performed according to the Systematic Reviews and Meta-Analyses (PRISMA) statement [28]. This study was registered at the International Prospective Register of Systematic Reviews (CRD42019139135). The date of registration is 16 June 2019.

The choice of THA or HA in DFNF in the active very elderly (over 75 years) remains uncertain [7, 8, 13, 14, 16, 24‐27]. Our study revealed that THA was superior in terms of acetabulum erosion, pain HHS, total HHS, and EQ-5D, with a trend of a lower mortality rate within 6 months after surgery. However, the THA group had a longer average operative time and a higher dislocation rate, with a trend towards higher general complication rates, and longer duration of hospital stays. Moreover, there was no significant difference in terms of reoperation rate, postoperative infection, peri-prosthetic fractures, and VTE prevalence across the groups.

Compared to other meta-analyses [7, 8, 24‐27], our study limited patients’ age to over 75 years old, which was more specific than the previous studies. We excluded two RCTs due to the inclusion of patients with compromised mental states [35, 36], and two RCTs because they used a THA procedure utilizing the PCU acetabulum, which was pliable, and therefore might increase the erosion rates and negatively affect stability. This method also led to differences in operative time, hip function, and complication rates between PCU-THA and traditional THA [37, 38]. To our knowledge, this is the first meta-analysis comparing THA and HA in the active elderly over 75 years. The strict inclusion criteria reduced heterogeneity significantly in essential outcomes, such as pain and total HHS when compared to other meta-analyses [7, 8, 24].

We found no significant difference in the reoperation rate between two groups, which was supported by all the RCTs included [6, 9, 12‐16, 21, 22]. However, meta-analyses without strict inclusion criteria on age, THA component, or mental states suggested that reoperation rates were higher in the HA groups [7, 8, 24, 26, 27, 39]. Kannan et al. [23] found THA with an increased reoperation rate compared to HA according to the national registries, which would be lower with increasing age. The negative finding of our study may result from less physical demands and lower surgical tolerance in this population. After adding in the patients with painful symptoms, the reoperation rate was significantly lower in the THA group after 5 years follow-up, which had the following implications. First, the acetabular wear of HA would lead to painful symptoms despite the lower physical demand. Second, patient frailty may make it less likely to perform reoperation. Third, patients over 75 years in the THA group might have a better quality of life in the long term [35].

Although all the RCTs included except van den Bekerom et al. [6] found no significant difference in dislocation rate between two procedures in patients over 75 years [9, 12‐16, 21, 22], we found a higher dislocation rate in the THA group (5.3% vs. 0.64%), which was supported by other studies [6‐8]. A recent systematic review [8] suggested that the dislocation rate in patients over 50 years was 8.1% and 2.7% in the THA and HA group, respectively. We speculate that the lower trend of dislocation rate in patients over 75 years is mainly due to the lower level of activity. We believe the dislocation rate of THA, which is 3.9% within 6 months and 5.3% within 5 years, is acceptably low [14]. 82.4% of the patients with dislocation could be treated conservatively [13, 14, 16, 21, 22]. Moreover, our results suggested that the dislocation risk of THA was higher within 6 months after surgery. Avery et al. [13] found that all dislocations in the THA group occurred within 1 month after surgery. van den Bekerom et al. [6] reported that 62.5% of the dislocations occurred in the hospital-stay duration. Additionally, we found no dislocation occurs in the 84 patients with a follow-up of more than 5 years (mean 129 m), which was not mentioned in other meta-analyses [7, 8, 25‐27]. We theorize that first, dislocation may be a risk factor for mortality in patients over 75 years [21]. Second, patients with long-term follow-up have significantly lower levels of activity, which may contribute to the absence of dislocation.

In our series, the THA group had better total HHS outcomes in the short (1 year) and medium term (5 years). These results were supported by previous studies [12, 14‐16, 22, 24, 25, 39]. Lewis et al. [8] found no significant difference in total HHS and pain HHS in patients over 80 years. However, they included some RCTs using PCU-THA, which might lead to high heterogeneity [37, 38]. Whether the superior hip function of THA in this population still exists in the long term is controversial, whereas there is not enough data for our study. Avery et al. [13] and Tol et al. [21] found no significant difference in HHS between the two groups at a mean follow-up of 9 and 12 years. Our trial found that the THA group had better pain HHS after both in the short term and the middle term, with low heterogeneity. The heterogeneity might be associated with the differences in surgical techniques and rehabilitation programs because the two RCTs had similar inclusion criteria, patients’ characteristics, and prosthesis types [21, 22]. These results were different from the previous meta-analyses with younger patients [7, 8]. Besides, we found no difference in the functional HHS across the groups. Previous RCTs also reported the same result in 1, 5, and 12 years of follow-up [6, 21]. Therefore, we theorize that although the lower physical demand of patients over 75 years, THA can still provide less pain even in the short term due to the fewer acetabular wear, which may mainly contribute to the superior in total HHS and the quality of life [24]. The EQ-5D_index score was higher in the THA group with a mean follow-up of 18 months, indicating that the improvement of hip function by THA could still significantly improve the quality of life despite the lower activity level [14].

This study had some limitations. First, we included a small number of studies because of the strict inclusion criteria, and as a result, some essential data endpoints (average blood loss and long-term HHS) could not be analyzed. Second, nearly all the included studies [13, 16, 21, 22] had defects in the procedure for the blinding of participants and outcome assessment, and one trial was funded by the insurance company [22], which may have skewed the results. Third, we could only perform subgroup analyses according to age because of not enough data for the subgroup analyses based on comorbidities and ASA score. Fourth, the included studies were published between 2006 and 2017, and during this period, there were many advancements in orthopedics, such as prosthetic equipment and hip function evaluation methods. Further research with an older population or subgroup analyses based on other factors reflecting patient frailty, such as comorbidities and ASA score, is needed.

These findings suggest that THA may be a preferred management option for active elderly patients over the age of 75 years. First, the more extensive surgery of THA will not lead to detectable increases in mortality rate and general complications. Second, older patients, who have lower physical demand, can still benefit from THA in terms of hip function and quality of life. The major concern of THA is the higher risk of dislocation, especially in the first 6 months, which is acceptably low in this population.