Randomised controlled trials

The study and participant characteristics of the 13 included RCTs^{110,112,113,115,119,123–129,145} are summarised in Table 9. More details can be found in Appendices 3 and 4. Briefly, four RCTs were conducted in the USA,^{113,115,125,127} one in the UK,¹¹² one in Australia,¹²³ two in Norway^110,126 two in the Republic of Korea^128,129 and three in Canada.^111,119,124 A total of 3175 participants were randomised across the 13 RCTs, with the number of participants in each study ranging from 100^124,128,145 to 557.¹²³ The mean age of participants across the RCTs ranged from 45¹²⁹ to 72^123,145 years. The proportion of women across the studies ranged from 24%¹²⁹ to 73%.¹¹⁰ The length of follow-up of the studies ranged from 3 months¹¹⁹ to 20 years.^119,129 The proportion of participants diagnosed with primary OA was reported for nine studies^{110,112,113,115,123,124,127–129} and ranged from 14%¹²⁹ to 96%.¹²³

TABLE 9

Overall study characteristics across the 13 RCTs comparing different types of THR

Comparison of THR interventions in the included RCTs was based on differences in hip replacement implant components (e.g. acetabular cup/shell, femoral stem and femoral head) according to their composition,¹²⁷ design,^115,128 bearing surface,^{113,115,124–126,145} fixation method^{110,112,119,129} and component size.¹²³ Table 10 shows the distribution of RCTs across the THR comparison categories.

TABLE 10

Distribution of 13 RCTs according to basis of THR comparison

Reported outcomes across the 13 RCTs varied. Most RCTs reported HHS^{110,112,113,115,119,124–129,145} and risk of revision.^{112,113,115,119,123–125,127–129} The follow-up of outcome assessments ranged from 3 months¹¹⁹ to 20 years.^119,129 Outcomes reported in the included studies can be found in Appendix 8. A summary of the functional/clinical and quality of life measures/tools used is provided in Appendix 9.

Systematic reviews

The five included systematic reviews^137–141 evaluated RCTs and non-RCTs of the clinical effectiveness of THR (see Appendix 3). The primary focus of these systematic reviews was the comparison of the effects of different cup fixation methods (cemented vs. cementless)^137–139 and materials used for implant articulations^140,141 on postoperative clinical/functional scores (HHS, OHS)^137,138,140 and risk of revision rate.^138,139 Searches in these systematic reviews were undertaken between July 2007¹⁴¹ and June 2011.¹³⁹ Further details on specific outcomes reported in the included systematic reviews can be found in Appendix 8.

Risk of bias in the randomised controlled trials

The risk-of-bias assessments for the 13 included RCTs comparing different types of THR are presented in risk-of-bias tables (see Appendix 2), the summary table (Table 11) and the risk-of-bias graph (Figure 10). Overall, four^{112,119,123,128} of the 13 RCTs reported an adequate method for random sequence generation and eight^{110,112,119,123–126,129} reported adequate treatment allocation concealment (low risk of bias). A greater proportion of the RCTs were rated as having a low risk of performance and detection bias for objective (e.g. mortality, dislocation) than for subjective (e.g. patient-administered functional scores) outcomes (92–100% vs. 15–23%, respectively). For at least eight of the RCTs, it was unclear whether or not awareness of THR type would influence the ascertainment of clinical/functional scores by patients/study personnel (performance bias)^{110,112,113,115,124,125,127–129,145} or outcome assessors (detection bias).^{112,113,115,124–126,128,129} Most RCTs failed to report the blinding status of the patients, study personnel and/or outcome assessors. Eight RCTs were judged as having a low risk of attrition bias. Five RCTs^{115,124,125,127,128} were judged as being at high risk for selective outcome and/or analysis bias. The risk of other biases (e.g. funding source, baseline imbalance in important characteristics, inappropriate analysis) for about one-third of the RCTs was judged to be high.

TABLE 11

Risk of bias summary for RCTs: review authors’ judgements about each risk of bias item – THR vs. THR

FIGURE 10

Risk of bias graph for RCTs: review authors’ judgements about each risk of bias item – THR vs. THR. ITT, intention to treat; NA, not applicable; PP, per protocol.

Methodological quality of the systematic reviews

The assessment of methodological quality of the five included systematic reviews comparing different types of THR is presented in Table 12 and the quality assessment sheets (see Appendix 2). Briefly, based on the number of methodological items that were satisfied, two systematic reviews^137,140 were judged to be of high quality (falling into the score range of 9–11) and two systematic reviews^138,141 were of medium quality (falling into the score range of 5–8). The one remaining systematic review¹³⁹ was judged to be of low quality (falling into the score range of 0–4). The specific unmet methodological items related to inappropriate analysis, absence of duplicate study selection, limited literature search, failure to address issues of publication bias and no information on conflicts of interest.

TABLE 12

Methodological quality assessment summary for systematic reviews: THR vs. THR

Functional/clinical measures

Twelve of the 13 included RCTs comparing different types of THR reported at least some results for the following functional scores measured at different postprocedure follow-up times: HHS (12 studies^{110,112,113,115,119,124–129,145}) WOMAC score (four studies^{119,124,129,145}), MACTAR score (one study¹¹⁹), Merle d’Aubigné and Postel score (one study¹¹⁹) and UCLA activity score (one study¹²⁹). None of these 12 studies reported measurements of the OHS.

Three of the five included systematic reviews comparing different types of THR reported at least some evidence on HHS^137,138,140 and OHS.¹³⁷ None of the three reviews reported any summary evidence for WOMAC, MACTAR, Merle d’Aubigné and Postel, and UCLA scores.

Randomised controlled trials (n = 12)

Mean HHS at follow-up (range 6 months–10 years) did not differ between the following interventions: cup fixation (two studies;^110,112 cemented vs. cementless), cup liner bearing surface (two studies;^113,145 cross-linked polyethylene vs. non-cross-linked polyethylene), cup and femoral stem fixation (one study;¹¹⁹ cemented vs. cementless) and femoral head-on-cup liner bearing surfaces (one study;¹²⁶ cobalt–chromium/oxinium-on-polyethylene vs. cobalt–chromium/oxinium-on-cross-linked polyethylene) (Table 13). The pooled MD for HHS in our meta-analysis of two studies^113,145 comparing cup liners made with cross-linked polyethylene compared with non-cross-linked polyethylene was 2.29 (95% CI –0.88 to 5.45), suggesting a non-significant benefit of cross-linked polyethylene cup liners (Figure 11).

TABLE 13

Harris Hip Score (range 0–100): RCTs

FIGURE 11

Harris Hip Score. XLPE, cross-linked polyethylene.

The evidence for the other comparisons based on cup shell design (porous coated vs. arc-deposited hydroxyapatite coated),¹¹⁵ femoral head bearing surface (oxinium vs. cobalt–chromium),¹²⁴ femoral head-on-cup liner bearing surfaces (ceramic-on-ceramic vs. metal-on-polyethylene or ceramic-on-polyethylene),^115,125 femoral stem composition (cobalt–chromium vs. titanium),¹²⁷ femoral stem design (short metaphyseal fitting vs. conventional diaphyseal fitting)¹²⁸ and femoral stem fixation (cemented vs. cementless)¹²⁹ was judged to be inconclusive.

Systematic reviews (n = 3)

One systematic review¹⁴⁰ reported the pooled MD for the HHS (Table 14). Pooled estimates for the comparison between metal-on-metal and metal-on-polyethylene bearing surfaces at two different follow-up times were not consistent: at 2 years metal-on-metal bearing surfaces gave a significantly higher HHS than metal-on-polyethylene, but at > 2 years there was no significant difference between the two types of THR. The remaining two systematic reviews presented only narrative summaries.^137,138 In summary, for the HHS the systematic review-based evidence was judged to be inconclusive.

TABLE 14

Harris Hip Score (range 0–100): systematic reviews

Western Ontario and McMaster University Osteoarthritis Index score

RCTs (n = 4)

Results from all four RCTs reporting postprocedural mean WOMAC scores indicated statistically non-significant differences between the THR groups compared with respect to cup liner bearing surface (cross-linked polyethylene vs. non-cross-linked polyethylene),¹⁴⁵ cup and femoral stem fixation (cemented vs. cementless),¹¹⁹ femoral head bearing surface (oxinium vs. cobalt–chromium)¹²⁴ and femoral stem fixation (cemented vs. cementless)¹²⁹ (Table 15). The MD in WOMAC score of –0.12 (95% CI –7.58 to 7.34) observed for one RCT¹⁴⁵ suggested no difference between cross-linked polyethylene and non-cross-linked polyethylene cup liners. Results for WOMAC score in the remaining three RCTs^119,124,129 were judged to be inconclusive because of incompletely reported data.

TABLE 15

Western Ontario and McMaster University Osteoarthritis Index (range 0–100): RCTs

Systematic reviews (n = 0)

No evidence was identified.

Other functional/clinical scores

Randomised controlled trials (n = 2)

In one RCT¹¹⁹ there was no difference in mean MACTAR scores (at 7 years: mean change difference 0.20, 95% CI –0.74 to 1.14) and Merle d’Aubigné and Postel scores (at 7 years: mean change difference –0.40, 95% CI –1.34 to 0.54) between patients who received a THR with cemented components and those who received a THR with cementless components (Tables 16 and 17). Results from one RCT¹²⁹ comparing femoral stem fixation (cemented vs. cementless) by the postoperative UCLA activity score were inconclusive because of incomplete data reporting (Table 18).

TABLE 16

McMaster Toronto Arthritis Patient Preference Questionnaire scores (range 0–30): RCTs

TABLE 17

Merle d’Aubigné and Postel scores (range 0–18): RCTs

TABLE 18

University of California Los Angeles activity scores (range 1–10): RCTs

Systematic reviews (n = 1)

The OHS was reported in one systematic review¹³⁷ comparing cup fixation methods (cemented vs. cementless), but the results were inconclusive (Table 19). This evidence was based on one RCT showing a statistically non-significant result.

TABLE 19

Oxford Hip Score (range 0–48): systematic review

Health-related quality of life

Only three RCTs^124,125,145 and one systematic review¹⁴⁰ reported any comparative evidence for measures of health-related quality of life.

Randomised controlled trials (n = 3)

In one RCT,¹⁴⁵ at follow-up times of 1 and 5 years, there was no difference in quality of life (on the mental and physical subscales of SF-12) between two groups of patients receiving cross-linked and non-cross-linked polyethylene cup liner bearings (Table 20).

TABLE 20

Short Form questionnaire-12 items (range 0–100): RCTs

In two other RCTs^124,125 there were no statistically significant differences in mean SF-12 mental and physical subscale scores between THR groups with different femoral head bearings (oxinium vs. cobalt–chromium)¹²⁴ and femoral head-on-cup liner articulations (ceramic-on-ceramic vs. ceramic-on-polyethylene).¹²⁵ This evidence was judged to be inconclusive (see Table 20).

Systematic reviews (n = 1)

One systematic review¹⁴⁰ reported evidence from two studies that compared SF-12 scores across different articulations (metal-on-metal vs. metal-on-polyethylene) (Table 21). The review did not provide any formal narrative or quantitative synthesis of the data. The evidence was considered to be inconclusive.

TABLE 21

Short Form questionnaire-12 items (range 0–100): systematic review

Revision

Evidence on revision was reported for 10 RCTs^{112,113,115,119,123–125,127–129} and five systematic reviews.^137–141

Randomised controlled trials (n = 10)

One RCT¹¹³ demonstrated a reduced risk of revision in patients who received cross-linked polyethylene compared with non-cross-linked polyethylene cup liners (RR 0.18, 95% CI 0.04 to 0.78) (Table 22). The evidence reported in the remaining nine RCTs showed statistically non-significant differences in the risk of revision between the different types of THR with wide CIs compatible with large size effects in both directions (i.e. favouring one or other of the treatment group). This evidence was deemed to be inconclusive (see Table 22).

TABLE 22

Revision rate: RCTs

Systematic reviews (n = 5)

Of the five systematic reviews reporting on revisions, two^137,141 provided pooled estimates for risk of revision (Table 23). According to one review,¹⁴¹ at 9 years post surgery the recipients of zirconium femoral heads were at similar risk for revision as the recipients of non-zirconium femoral heads (three pooled RCTs; risk difference 0.02, 95% CI –0.01 to 0.06). This evidence was considered conclusive in detecting no difference in revision rates between these two types of femoral head.

TABLE 23

Revision rate: systematic reviews

In another review¹³⁷ the risk of revision at 10 years after surgery did not significantly differ between cemented and cementless cup fixation THR groups (pooled RR 0.15, 95% CI 0.02 to 1.18). This result was considered inconclusive given the uninformative 95% CIs. Evidence from the remaining three reviews^138–140 was of a narrative nature, which precluded us drawing conclusions (see Table 23).

Mortality

Evidence on mortality was reported for six RCTs.^{110,113,119,123,128,145} None of the five systematic reviews reported on mortality.

Randomised controlled trials (n = 6)

Evidence from the six RCTs^{110,113,119,123,128,145} that reported mortality was inconclusive because of non-significant RR estimates and wide 95% CIs (Table 24). For example, based on a pooled RR estimate of 1.39 (95% CI 0.78 to 2.49),^113,145 5- to 10-year post-surgery mortality rates in the group receiving cross-linked polyethylene cup liners were not significantly different from those in the group receiving non-cross-linked polyethylene cup liners (Figure 12). Similarly, the rest of the studies showed non-significant results for mortality between THR groups defined by femoral stem and/or cup fixation (cemented vs. cementless)^110,119 and femoral head size (36 mm vs. 28 mm).¹²³ One RCT¹²⁸ reported no deaths for both treatment groups receiving femoral stems of different design.

TABLE 24

Mortality rate: RCTs

FIGURE 12

Mortality. XLPE, cross-linked polyethylene.

Femoral head penetration rate (measure of prosthesis movement)

Evidence on femoral head penetration rate was reported by three RCTs.^113,126,145 None of the five systematic reviews reported this end point.

RCTs (n = 3)

Two RCTs^113,145 demonstrated reduced femoral head penetration in favour of cross-linked polyethylene cup liners compared with non-cross-linked (conventional) polyethylene cup liners at 5–10 years of follow-up (Table 25). Similarly, in another RCT,¹²⁶ cross-linked polyethylene cup liners with either metal or oxinium femoral heads outperformed conventional polyethylene cup liners in reducing femoral head penetration during 2 years of follow-up.

TABLE 25

Femoral head penetration rate: RCTs

Complications

Evidence on the occurrence/absence of complications was reported by nine RCTs^{112,113,115,123–125,127,129,145} and three systematic reviews.^138–140 In most studies^{112,113,115,123–125,129,145} the reported complications were classified as postoperative. In one RCT¹²⁷ some of the complications were classified as perioperative.

Implant dislocation

Randomised controlled trials (n = 7)

Evidence on the occurrence/absence of implant dislocation was reported by seven RCTs^{110,112,115,123–125,127} (Table 26). Our pooled estimate of two studies^110,112 (Figure 13) indicated a reduced risk of implant dislocation at 10 years’ follow-up in recipients of cemented compared with cementless cups (pooled OR 0.34, 95% CI 0.13 to 0.89). Moreover, in one RCT¹²³ after 1 year of follow-up, the THR recipients with a larger size of femoral head experienced a lower risk of implant dislocation than those with a smaller size of femoral head (36 mm vs. 28 mm: RR 0.17, 95% CI 0.04 to 0.78). Evidence on implant dislocation for the remaining four RCTs^{115,124,125,127} was inconclusive because of incomplete data and non-significant results.

TABLE 26

Implant dislocation rate: RCTs

FIGURE 13

Implant dislocation.

Systematic reviews (n = 2)

Overall, no conclusions on implant dislocation could be drawn from the two systematic reviews, given the narrative evidence summary¹⁴⁰ and the mixed study designs¹³⁹ (Table 27). The pooled data from one review¹³⁹ was based on nine studies, most of which were not randomised and which indicated a lower risk of dislocation in the groups receiving cemented compared with cementless cups.

TABLE 27

Implant dislocation rate: systematic reviews

Osteolysis

Randomised controlled trials (n = 7)

Evidence on osteolysis was reported by seven RCTs^{112,113,115,125,127,129,145} (Table 28). In one RCT¹¹⁵ comparing different femoral head-on-cup liner bearing surfaces, recipients of ceramic-on-ceramic articulations had a reduced risk of osteolysis compared with recipients of metal-on-polyethylene articulations at 10 years post operation (RR 0.10, 95% CI 0.02 to 0.32).

TABLE 28

Osteolysis: RCTs

For seven RCTs, the evidence for osteolysis was inconclusive across the comparisons based on different methods of cup fixation (cemented vs. cementless),¹¹² cup liner bearing surface (cross-linked polyethylene vs. non-cross-linked polyethylene),^113,145 cup shell design (porous coated vs. arc-deposited hydroxyapatite coated),¹¹⁵ femoral head-on-cup liner bearing surface (ceramic-on-ceramic vs. ceramic-on-polyethylene),¹²⁵ femoral stem composition (cobalt–chromium vs. titanium)¹²⁷ and femoral stem fixation (cemented vs. cementless).¹²⁹

Systematic reviews (n = 2)

Overall, no conclusions could be drawn on the incidence of osteolysis from two low-quality systematic reviews^138,139 comparing cemented and cementless methods of cup fixation, given the narrative evidence summaries, mixed study designs and inconsistent findings (Table 29).

TABLE 29

Osteolysis: systematic reviews

Other complications

Randomised controlled trials (n = 7)

Seven RCTs reported other complications such as aseptic loosening (Table 30),^{112,113,119,124,127} femoral fracture (Table 31),^113,115,127 infection (Table 32),^{112,124,125,127} and deep-vein thrombosis (Table 33).¹²⁵ This evidence was judged to be inconclusive because of low event or zero event counts and CIs indicating great uncertainty.

TABLE 30

Aseptic loosening: RCTs

TABLE 31

Femoral fracture: RCTs

TABLE 32

Infection: RCTs

TABLE 33

Deep-vein thrombosis: RCTs

Systematic reviews (n = 1)

Of other complications, only aseptic loosening was reported in one low-quality systematic review¹³⁹ (Table 34). Pooled data from 11 studies, most of which were not randomised, pointed towards a greater risk of aseptic loosening with cemented compared with cementless cups; however, the evidence is inconclusive given the lack of numerical data and the evidence synthesis being based on mixed study designs.

TABLE 34

Aseptic loosening: systematic review

Randomised controlled trials

The majority of the evidence comparing THRs was rated as inconclusive by us (Table 36). In three RCTs there was evidence of a reduced risk of implant dislocation with the use of a cemented cup (vs. a cementless cup)^110,112 or a larger femoral head size (36 mm vs. 28 mm)¹²³ (high-grade evidence for the cup fixation comparison). In three other RCTs, patients who received a THR with a cross-linked polyethylene cup liner experienced a reduced (i.e. improved) femoral head penetration rate (moderate-grade evidence)^113,126,145 and risk for revision (very low-grade evidence)¹¹³ compared with recipients of conventional polyethylene cup liners. In one RCT¹¹⁹ the use of cementless fixation of the cup and femoral stem (vs. cemented fixation) was associated with a better implant survival rate. Moreover, the recipients of ceramic-on-ceramic articulations (vs. metal-on-polyethylene) experienced a reduced risk of osteolysis.¹¹⁵ For half of the studies,^{110,112,113,119,126,145} the mean post-THR clinical and functional scores (i.e. HHS, WOMAC score, SF-12 score, MACTAR score, Merle d’Aubigné and Postel score) measured at different follow-up times were similar between the different THR treatment groups (moderate-grade evidence for no difference in HHS across the comparisons for cup fixation and cup liner surface types).

TABLE 36

Summary of evidence regarding the differences between the different types of THR for each reported outcome: RCTs

Evidence from studies reporting the UCLA activity score,¹²⁹ mortality (very low-grade evidence),^{110,113,119,123,128,145} aseptic loosening,^{112,113,119,124,127} femoral fracture,^113,115,127 infection^{112,124,125,127} and deep-vein thrombosis¹²⁵ was all inconclusive. Also, the evidence reported in four studies was considered inconclusive for all outcomes (very low-grade evidence).^{124,125,127,128} Results were considered inconclusive by us because of partial reporting (missing data for effect estimates, CIs, SEs, SDs, p-values), great uncertainty (wide CIs), zero event counts and/or inconsistency in estimates.

Systematic reviews

The majority of evidence from the five systematic reviews comparing different types of THR^137–141 was considered inconclusive. This is because of unreported pooled results across RCTs (i.e. reporting only narrative syntheses), the reporting of inappropriate pooling methods (e.g. indirect naive comparison of single-group cohorts; pooling of studies of different design)^138,139,141 or the reporting of inconsistent summary findings¹⁴⁰ (Table 37). The evidence from one review¹⁴¹ indicated no difference in the risk for revision between two different articulations of zirconium-on-polyethylene and non zirconium-on-polyethylene.

TABLE 37

Summary of evidence regarding the differences between the different types of THR for each reported outcome: systematic reviews

Publication bias

The extent to which publication bias could have influenced the pooled treatment effect estimates (i.e. degree of funnel plot asymmetry) could not be explored because of an insufficient number of data points in the forest/funnel plots.

Heterogeneity, subgroup effects and sensitivity analysis

The data reviewed from RCTs were too sparse and heterogeneous (in terms of different types of THR) to allow exploration of whether or not the relative effect of any given THR differed by study-level methodological characteristics (i.e. risk of bias, type of data analysis) or patient-related characteristics (i.e. age, sex or functional status). None of the included RCTs reported within-study subgroup effects of the different THRs compared.

Randomised controlled trials

Study and participant characteristics of the three included RCTs^130–132 are summarised in Table 38. More details can be found in Appendices 3 and 4. Two RCTs^131,132 were conducted in Canada and one¹³⁰ was conducted in the UK. A total of 422 participants were randomised across the three RCTs, ranging from 104¹³¹ to 192¹³² participants. The mean age of participants ranged from 50¹³² to 56¹³⁰ years and the proportion of female participants across the studies ranged from 10.5%¹³¹ to 41%.¹³⁰ The total length of follow-up the studies ranged from 1 year¹³⁰ to 6 years.¹³² The proportion of participants diagnosed with primary OA was reported for two studies^130,132 and was 33%¹³² and 95%.¹³⁰

TABLE 38

Overall study characteristics across the three RCTs comparing THR and RS

The three RCTs reported on clinical/functional scores (e.g. HHS, OHS, UCLA activity score, WOMAC score), health-related quality of life and risk of revision. Follow-up of outcome assessments ranged from 3 weeks¹³⁰ to 5 years.¹³² Outcomes reported in the included studies can be found in Appendix 8.

Systematic reviews

Three systematic reviews^142–144 were included that evaluated the clinical effectiveness of THR compared with RS with respect to postoperative clinical/function (HHS, WOMAC score), risk of revision, mortality and complications. Searches for these systematic reviews were undertaken between March 2008¹⁴⁴ and January 2010.¹⁴³ Evidence was synthesised from both RCTs and non-RCTs (see Appendices 3 and 4). Further details on specific outcomes reported (or not reported) in the included systematic reviews can be found in Appendix 8.

Risk of bias in randomised controlled trials

The risk of bias assessment for the three included RCTs^130–132 comparing THR with RS is presented in risk of bias tables (see Appendix 2), the summary table (Table 39) and the risk of bias graph (Figure 14). Overall, two studies^130,132 reported an adequate method for random sequence generation and all three studies^130–132 reported treatment allocation concealment (low risk of bias). Two of the three studies^130,132 were rated as having a low risk of performance and detection bias for objective outcomes (e.g. revision, dislocation). The same two studies had a high risk of performance bias for subjective outcomes (e.g. patient-administered functional scores). Patients and study personnel were blinded in only one study.¹³¹ For two studies^130,132 the influence of attrition bias on objective outcomes was judged to be of low risk. All three studies were judged as being at low risk for selective outcome and/or analysis bias. Risk of other biases (e.g. funding source, balance/imbalance in important characteristics, inappropriate analysis) for one of the three studies was judged to be high.¹³¹

TABLE 39

Risk of bias summary for RCTs: review authors’ judgements about each risk of bias item – THR vs. RS

FIGURE 14

Risk of bias graph for RCTs: review authors’ judgements about each risk of bias item – THR vs. RS. ITT, intention to treat; NA, not applicable; PP, per protocol.

Methodological quality of systematic reviews comparing total hip replacement with resurfacing arthroplasty

The assessment of methodological quality of the three included systematic reviews^142–144 is presented in Table 40 and the data extraction sheets (see Appendices 3 and 4). Given the number of methodological items that were satisfied, one of the three reviews was judged as being of high quality (falling into the score range 9–11),¹⁴³ one was judged as being of medium quality (falling into the score range 5–8)¹⁴² and one was judged as being of low quality (falling into the score range 0–4).¹⁴⁴ The specific unmet methodological items related to inappropriate analysis, failure to address issues of publication bias and no information on conflicts of interest.

TABLE 40

Methodological quality assessment summary for systematic reviews: THR vs. RS

Functional/clinical measures

All three included RCTs comparing THR and RS reported some evidence for the following functional scores measured at 12–24 months after the procedure: HHS,¹³⁰ OHS,¹³⁰ WOMAC score,^131,132 UCLA activity score^131,132 and Merle d’Aubigné and Postel score.¹³²

In two RCTs there was no difference between the THR group and the RS group in mean postoperative OHS (12 months; MD –2.23, 95% CI –5.98 to 1.52),¹³⁰ Merle d’Aubigné and Postel score (24 months; MD 0.0, 95% CI –1.06 to 1.06)¹³² or WOMAC score (12 months; MD 2.20, 95% CI –1.57 to 5.97).¹³² One of these RCTs showed a significantly improved mean WOMAC score for the RS group compared with the THR group at 24 months of follow-up; however, this difference was not deemed to be clinically important (MD 3.30, 95% CI 0.01 to 6.58).¹³²

All three included RCTs comparing THR with RS reported some evidence for the following functional scores measured at 12–24 months after the procedure: HHS,¹³⁰ OHS,¹³⁰ WOMAC score,^131,132 UCLA score^131,132 and Merle d’Aubigné and Postel score.¹³²

Health-related quality of life

Two RCTs reporting quality of life measures showed statistically non-significant differences between the THR group and the RS group for both the SF-36 (p = 0.55 and p = 0.97 for mental and physical components, respectively)¹³¹ and the EQ-5D (MD –0.08, 95% CI –0.18 to 0.03).¹³⁰ These results were deemed to be inconclusive given the wide CI¹³⁰ and incomplete data reporting.¹³¹

Revision

The occurrence of implant revision was reported for only one RCT.¹³² There was no statistically significant difference between the THR group and the RS group for risk of revision at 6 months (RR 1.01, 95% CI 0.06 to 15.92), 24 months (RR 0.50, 95% CI 0.04 to 5.48) or 56 months (RR 0.54, 95% CI 0.10 to 2.91) post surgery. The 95% CIs around the effect estimates embraced the value 1.00 and therefore did not allow definitive conclusions to be made regarding the effectiveness of THR compared with RS.

Mortality rate

No evidence on mortality rates was identified from the RCTs.

Complications

Evidence on complications was reported for two RCTs.^130,132 Meta-analysis of the data on risk of infection from the two RCTs indicated that, at 12–56 months post operation, THR recipients were at an increased risk of infection compared with RS recipients (pooled OR 7.94, 95% CI 1.78 to 35.40) (Figure 15). In addition, evidence on the differences between groups for the risk of deep-vein thrombosis (Figure 16; pooled OR 0.60, 95% CI 0.15 to 2.42),^130,132 implant dislocation (Figure 17; pooled OR 3.97, 95% CI 0.79 to 19.90),^130,132 wound complications (RR 4.01, 95% CI 0.92 to 18.18)¹³⁰ and aseptic loosening (RR not estimable)¹³² was judged to be inconclusive by us.

FIGURE 15

Risk of infection.

FIGURE 16

Risk of deep-vein thrombosis.

FIGURE 17

Risk of implant dislocation.

A summary of the results for the difference outcomes is presented in Table 41.

TABLE 41

Summary of the results for THR vs. RS: RCTs

Functional/clinical measures

Two of the three included systematic reviews comparing THR with RS reported evidence on HHS,^142,143 WOMAC score,^142,143 Merle d’Aubigné and Postel score¹⁴² and UCLA activity score¹⁴² (Table 42). The evidence was inconclusive because of the lack of pooled MD estimates for all four scores as well as the inconsistent results for the mean HHS and WOMAC score.

TABLE 42

Summary results for RS vs. THR: systematic reviews

Health-related quality of life

No evidence was identified.

Revision

Both systematic reviews^142,143 found a higher risk of revision in patients receiving RS than in those receiving THR. One review meta-analysed data from four RCTs that compared risk of revision in RS and THR recipients, reporting a pooled RR estimate of 2.60 (95% CI 1.31 to 5.15) (see Table 42).¹⁴²

Mortality

Overall, evidence on mortality reported by both systematic reviews^142,143 was inconclusive because of great uncertainty in the effect estimates and the variability around them. For example, the pooled RR for mortality in one review¹⁴³ for the comparison between RS and THR was 1.10 (95% CI 0.10 to 17.8) (see Table 42).

Failure rate

One systematic review¹⁴⁴ reported an indirect naive comparison analysis (i.e. analysis without a common comparator) based on data from 15 studies of RS and 19 studies of THR (see Table 42). The analysis suggested a reduced risk of failure in the RS recipients compared with the THR recipients (3.70% vs. 11.60%). Given the well-recognised problems with validity of such methodology, this evidence was judged to be inconclusive.

Complications

Evidence on complications was reported by both systematic reviews^142,143 (i.e. implant dislocation, infection and component loosening) (see Table 42). The evidence consistently showed an increased risk for component loosening^142,143 but a reduced risk for implant dislocation¹⁴² among RS recipients compared with THR recipients. One review,¹⁴² which provided the risk of infection pooled across three studies, was not informative enough to draw any conclusions (RR 2.25, 95% CI 0.61 to 8.31).

Publication bias

The extent to which publication bias could have influenced the pooled treatment effect estimates (i.e. degree of funnel plot asymmetry) could not be explored because of insufficient numbers of data points in the forest/funnel plots.

Heterogeneity, subgroup effects and sensitivity analysis

The reviewed data from RCTs were too sparse (only three RCTs) to allow an exploration of whether or not the effect of any given THR relative to RS differed by study-level methodological (i.e. risk of bias, type of data analysis) or patient-related (i.e. age, sex or functional status) characteristics. None of the included RCTs reported within-study subgroup effects of the THR relative to RS (or vice versa).

Study design

• RCTs.
• Observational designs, cohort studies and registry-based studies.
• Decision-analytic modelling studies.
• Systematic reviews.
• Meta-analyses.

Population

• People with pain or disability resulting from end-stage arthritis of the hip for whom non-surgical management has failed.

Intervention

• Elective primary THR.
• Primary hip RS.

Comparator

• Different types of primary THR compared with RS for people in whom both procedures are suitable.
• Different types of primary THR compared with each other for people who are not suitable for hip RS.
• Studies reporting costs or utilities without a comparator were also included.

Record

• Full-text articles of completed or in-progress studies (protocols) published in English.

Outcomes

• Cost-effectiveness outcomes were costs (cost of resources/devices, quantitative use of resources reported) and clinical effectiveness measures or utility measures (utility, EQ-5D score or QALYs), incremental cost-effectiveness ratios (ICERs), uncertainty measures, the ceiling willingness-to-pay (WTP) ratios and probabilities of cost-effectiveness from cost-effectiveness acceptability curves (CEACs).

Study design (registries)

• Reporting of the results of joint replacement registry data collection.
• All study designs.
• Most recent publication in the series.

Population

• People with pain or disability resulting from end-stage arthritis of the hip for whom non-surgical management has failed.

Intervention

• Elective primary THR.
• Primary hip RS.

Comparator

• Different types of primary THR compared with hip RS for people in whom both procedures are suitable.
• Different types of primary THR compared with each other for people not suitable for hip RS.

Record

• Full-text articles of completed studies published in English and annual reports of national registries.

Outcomes

• All reported outcomes.

England and Wales

Jameson et al.³⁵³ conducted a retrospective cohort study and reported survival time to revision for RS procedures from 2003 to 2013. The study explored the risk factors independently associated with failure. Mean time to revision for each group was not reported. Data were taken from the NJR for England and Wales. The study concluded that women were at greater risk of revision than men (HR 1.30, 99% CI 1.01 to 1.76; p = 0.007), independent of age. Smaller femoral head components were also significantly more likely to require revision than medium (≤ 44 mm: HR 2.14, 99% CI 1.53 to 3.00; p < 0.001) or large heads (45–47 mm: HR 1.48, 99% CI 1.09 to 2.00; p = 0.001), as was surgery performed by low-volume surgeons (HR 1.36, 99% CI 1.09 to 1.71; p < 0.001).

McMinn et al.³¹⁸ examined mortality and revision rates among patients with OA undergoing THR, both cemented and uncemented procedures, or RS. The authors used data from the NJR database for the analysis [154,996 patients receiving cemented THR, 120,017 receiving uncemented THR and 8352 receiving RS (in particular, Birmingham hip RS)]. The baseline characteristics recorded include age (cemented mean 73.2 years, uncemented mean 66.7 years), sex (cemented: men 53,409, women 101,587; uncemented: men 50,529, women 69,488) and ASA grade. The analysis took into account the age of patients at primary surgery and their length of follow-up. Survival analysis was used to compare the cemented and uncemented procedures with adjustment for sex, age at primary surgery, ASA grade before the operation, complexity of the procedure and ‘both sides’ (surgery on both hips at the same time).

The multivariable survival analyses demonstrated a higher mortality rate for patients undergoing cemented THR than for those undergoing uncemented THR (adjusted HR 1.11, 95% CI 1.07 to 1.16). There was a lower revision rate for cemented procedures (unadjusted HR 0.53, 95% CI 0.50 to 0.57). The authors stated that these findings translate into small predicted differences in the population-averaged absolute survival probability at all time points. At 8 years post surgery the predicted probability of death in the cemented group was 0.013 higher (95% CI 0.007 to 0.019) than that in the uncemented group and the predicted probability of revision was 0.015 lower (95% CI 0.012 to 0.017). In multivariable analyses that included only men, there was a higher mortality rate in the cemented group and the uncemented group than in the RS group. RS had a similar revision rate to uncemented THR and both had a higher revision rate than cemented THR. The authors concluded that there was a small but significant increased risk of revision with uncemented THR compared with cemented THR, and a small but significant increased risk of death with cemented procedures.

A study from Smith et al.¹⁵ reported that, in women, RS resulted in worse implant survival than THR, regardless of head size. The predicted 5-year revision rates in 55-year-old women were 8.3% (95% CI 7.2% to 9.7%) for a 42-mm RS head, 6.1% (95% CI 5.3% to 7.0%) for a 46-mm RS head and 1.5% (95% CI 0.8% to 2.6%) for a 28-mm cemented metal-on-polyethylene stemmed THR. In men with smaller femoral heads, RS resulted in poor implant survival. Predicted 5-year revision rates in 55-year-old men were 4.1% (95% CI 3.3% to 4.9%) for a 46-mm RS head, 2.6% (95% CI 2.2% to 3.1%) for a 54-mm RS head and 1.9% (95% CI 1.5% to 2.4%) for a 28-mm cemented metal-on-polyethylene stemmed THR. Of the male RS patients, only 23% (5085/22,076) had a head size ≥ 54 mm. The authors concluded that RS resulted in similar implant survival to other surgical options in men with large femoral heads, and worse implant survival in other patients, particularly women.

Finland

Seppanen et al.³²⁹ analysed the risk of revision of 4401 RS procedures in the Finnish Arthroplasty Register compared with the risk of revision of 48,409 THRs performed during the same time period. The median follow-up time was 3.5 (range 0–9) years for RS and 3.9 (range 0–9) years for THRs. The study reported no statistically significant difference in risk of revision between RS and THR (risk of revision 0.93, 95% CI 0.78 to 1.10). The 4-year unadjusted Kaplan–Meier survival rate was 96% (95% CI 96% to 97%) for both the RS group and the THR group. Female patients had about double the risk of revision as male patients (risk of revision 2.0, CI 1.4 to 2.7).

Australia

Buergi et al.³¹¹ reported the use of RS based on the Australian National Joint Replacement Registry. A total of 7205 RS procedures were carried out between 1999 and 2005. The study concluded that, in the database, early revision rates were higher for RS than for THR. At 3 years, the revision rate after RS was 2.8% and that after THR was 2.0%.

Multinational

Corten et al.³¹³ compared RS survivorship reported by registries in Australia, England and Wales and Sweden with the failure of THR between 2006 and 2009. RS was associated with an overall increased failure rate compared with THR. The cumulative revision rates in the Australian registry were 3.7% for RS and 2.7% for THR. The 3-year revision rate for RS was 1.8% in England and Wales and 3.4% in Sweden.

A study using data from the Nordic Arthroplasty Registry compared the outcome of RS (n = 1638) with that of THR (n = 309,290) between 1995 and 2007.³⁴⁹ Results indicated that RS had a threefold increased revision risk compared with THR (RR 2.7, 95% CI 1.9 to 3.7). The difference was greater when RS was compared with cemented THR (RR 3.8, 95% CI 2.7 to 5.3). In men aged < 50 years the difference in revision risk was less (RS vs. THR: RR 1.9, 95% CI 1.0 to 3.9; RS vs. cemented THR: RR 2.4, 95% CI 1.1 to 5.3). However, the difference in revision risk was higher in women of the same age group (RS vs. THR: RR 4.7, 95% CI 2.6 to 8.5; RS vs. cemented THR: RR 7.4, 95% CI 3.7 to 15). In the Cox regression analysis, RS showed an increased risk of early aseptic revision compared with THR (RR 2.7, 95% CI 1.9 to 3.7; p < 0.001) and cemented THR (RR 3.8, 95% CI 2.7 to 5.3; p < 0.001).

The purpose of one recent study³⁵¹ was to evaluate the outcome of Birmingham hip RS using revision rates as reported in national joint replacement registry studies (categorised as from the UK, Australia, Asia and the USA). In total, 9806 RS procedures were analysed (reported as 44,294 observed component-years). The analysis revealed a significant difference in revisions per 100 observed component-years between studies authored by specialist clinical centres (defined by the number of patients treated, staff training 4and personal expertise) (0.27, 95% CI 0.14 to 0.40) and the register data (0.74, 95% CI 0.72 to 0.76). The average revision rate from register data was 3.41% (SD 1.79%).

Summary of resurfacing arthroplasty in registry studies

In summary, the eight studies that reported data from joint registries had mixed results. There is little evidence from long-term studies; generally, 5-year revision rates (or less) were reported. No two studies had the same comparators for analysis, which makes drawing conclusions from the eight studies difficult. The reported benefits of RS include preservation of the bone on the femoral side, greater physiological stress transfer at the proximal femur and lower risk of dislocation because of the larger femoral head compared with conventional THR.³⁵¹ However, the majority of studies included in this review found that RS had a higher revision rate than THR, particularly in female patients. Only one study found no significant difference between the procedures.³²⁹ No studies were included that reported RS implant survival as better than that for THR. One study of men only reported that RS had a similar revision rate to that of uncemented THR, but that both had a higher revision rate than that of cemented THR.³¹⁸

England and Wales

Jameson et al.³¹⁷ reported survival time to revision following primary cemented THR in 34,721 THRs recorded in the NJR for England and Wales between 2003 and 2010. The authors reported the 7-year rate of revision for any reason as 1.70% (99% CI 1.28% to 2.12%). The overall risk of revision was independent of age, sex, ASA grade, BMI, surgeon volume, surgical approach, brand of cement/presence of antibiotic, femoral head material (stainless steel/alumina) and stem taper size/offset.

Smith et al.¹⁶ assessed the use of metal-on-metal bearing surfaces in the NJR between 2003 and 2011. They reported that metal-on-metal THR failed at high rates and that this was linked to head size. Analysis of the 31,171 metal-on-metal THRs showed that larger heads failed earlier (cumulative incidence of revision: 3.2%, 95% CI 2.5% to 4.1% for 28-mm heads and 5.1%, 95% CI 4.2% to 6.2% for 52-mm heads at 5 years in men aged 60 years). The 5-year revision rates in younger women were 6.1% (95% CI 5.2% to 7.2%) for 46-mm metal-on-metal THR and 1.6% (95% CI 1.3 to 2.1) for 28-mm metal-on-polyethylene THR. This finding contrasted with findings for ceramic-on-ceramic bearing surfaces, for which larger head sizes were associated with improved survival (5-year revision rate: 3.3%, 95% CI 2.6% to 4.1% for 28-mm heads and 2.0%, 95% CI 1.5% to 2.7% for 40-mm heads for men aged 60 years).

Denmark

Johnsen et al.³³⁹ examined the association between patient-related factors and the risk of initial, short-term and long-term failure after primary THR using data from the Danish Hip Arthroplasty Registry (n = 36,984). The study concluded that in Denmark between 1995 and 2002 male sex and comorbidity index score (Charlson Comorbidity Index) were strongly predictive of THR failure. The Charlson Comorbidity Index includes 19 disease categories, which correspond to International Classification of Diseases, Eighth Edition (ICD-8) and International Classification of Diseases, Tenth Edition (ICD-10) codes used in the national registries. A total of 1132 primary THRs were revised (3.1% of the 36,984 procedures) during this time period.

A more recent study from Denmark³⁴¹ evaluated short-term (0–90 days) and longer-term (up to 12.7 years) mortality of patients undergoing primary THR compared with mortality in the general population. THR patients (n = 44,558) was matched at the time of surgery with three people from the general population (n = 133,674). The findings suggest that there was a 1-month period of increased mortality immediately after surgery among THR patients (adjusted mortality rate ratio 1.4, 95% CI 1.2 to 1.7); however, overall short-term mortality (0–90 days) was significantly lower (adjusted mortality rate ratio 0.8, 95% CI 0.7 to 0.9). THR surgery was associated with increased short-term mortality in subjects aged < 60 years and among THR patients without comorbidity. Long-term mortality was lower among THR patients than in the general population control group (adjusted mortality rate ratio 0.7, 95% CI 0.7 to 0.7).

Sweden

Lazarinis et al.³⁴³ analysed patient data (n = 8043) on cementless cups with or without a hydroxyapatite coating that had been recorded in the SHAR between 1992 and 2007. The primary end point was revision because of aseptic loosening; the secondary end points were cup revision for any reason and cup revision because of infection. The results reported that the hydroxyapatite coating was a risk factor for cup revision because of aseptic loosening (adjusted RR 1.7, 95% CI 1.3 to 2). Age at primary THR of < 50 years, paediatric hip disease, a cemented stem and the cup brand were also associated with a statistically significantly increased risk of cup revision due to aseptic loosening.

A more recent study from Sweden reported data from 1999 to 2010.³⁴⁴ The authors investigated revision rates of monoblock cups used in primary THR that were registered in the SHAR. Kaplan–Meier and Cox regression analyses with adjustment for age, sex and other variables were used to calculate survival rates and adjusted HRs of the revision risk for any reason. The cumulative 5-year survival rate with any revision as the end point was 95% (95% CI 91% to 98%) for monoblock cups and 97% (95% CI 96% to 98%) for modular cups (p = 0.6). The adjusted HR for revision of monoblock cups compared with modular cups was 2 (95% CI 0.8 to 6, p = 0.1). The authors concluded that there was not any clinically relevant difference in risk of revision between monoblock and modular acetabular cups in the medium term.

Australia

Luo et al.³¹⁴ analysed the effect of the AOANJRR on the cost of joint arthroplasty through identification of implants with higher than expected failure rates between 2003 and 2007. A total of 242,454 primary joint arthroplasties were performed in Australia at a cost of AU$4.1B. The authors state that if the poor-performing THRs had been conducted using average longevity designs, the number of THR revisions could have been reduced by 47%.

One study³¹⁵ investigated the relationship between the bearing surface and the risk of revision because of dislocation using 110,239 records in the AOANJRR from 1999 to 2007. The authors reported that 2621 (2.4%) primary THRs were revised for any reason; 862 (0.78%) THRs were revised because of dislocation. Ceramic-on-ceramic bearing surfaces had a lower risk of revision for dislocation than metal-on-polyethylene and ceramic-on-polyethylene bearing surfaces at 7 years’ follow-up. The authors reported a significantly higher rate of revision for dislocation with ceramic-on-ceramic bearing surfaces than with metal-on-polyethylene bearing surfaces when smaller head sizes (≤ 28 mm) were used in younger patients (< 65 years) (HR 1.53, p = 0.041) and also with larger head sizes (> 28 mm) in older patients (≥ 65 years) (HR 1.73, p = 0.016).

Italy

Di Tanna et al.²⁶¹ report data from the Emilia-Romagna Regional Registry on Orthopaedic Prosthesis from 2000 to 2007. This registry collects information on all orthopaedic interventions performed in Emilia-Romagna, Italy. The study assessed the cost-effectiveness of cementless prostheses compared with hybrid prostheses in 41,199 THRs and concluded that there were differences in the revision rate and impact on costs between the two groups. The authors concluded that, considering two cohorts of 100 subjects, 243 revisions would be expected in the cementless group compared with 300 in the hybrid group. This was equal to a 19% difference and a number needed to treat of 18.

A second paper reporting on the Emilia-Romagna Regional Registry on Orthopaedic Prosthesis³²⁰ conducted survival analysis using the Kaplan–Meier method to analyse survival rates for THRs in Italy between 2000 and 2006 (35,042 THRs, 5878 revisions). The reported cumulative survival rate for THR at 7 years was 96.8% (95% CI 96.4% to 97.1%). Multivariate analysis demonstrated that THR survival was affected by pathology, for example the presence of RA. Women comprised 66.4% of patients and > 54.0% of patients were overweight (BMI > 25 kg/m²). Mean age at primary surgery was 66.9 years (range 16–101 years) and at revision was 70.0 years (range 22–98 years).

Finland

Eskelinen et al.³²³ evaluated the population-based survival of cementless THR in patients aged < 55 years using data from the Finnish Arthroplasty Register. All cementless stems studied showed a survival rate of > 90% at 10 years.

Makela et al.³²⁷ analysed population-based survival rates for cemented and cementless THRs in patients aged ≥ 55 years in Finland between 1980 and 2006. The 15-year survival rate for cementless THR (80%) was comparable with the rates for the cemented groups [86% in cemented group 1a (cemented, loaded-taper stem combined with a cemented, all-polyethylene cup) and 79% in cemented group 2 (a cemented, composite-beam stem with a cemented, all-polyethylene cup)] when revisions for any reason were used as the end point. The authors concluded that both cementless stems and cementless cups, analysed separately, had a significantly lower risk of revision for aseptic loosening than cemented implants.

The same authors reported revision outcomes in primary OA.³²⁸ The 15-year survival rate of group 1 cementless THR (implants with a cementless, straight, proximally circumferentially porous-coated stem and a porous-coated press-fit cup) performed in 1987–96 (62%, 95% CI 57% to 67%) and group 2 cementless THR (implants with a cementless, anatomic, proximally circumferentially porous-coated stem, with or without hydroxyapatite, and a porous-coated press-fit cup with or without hydroxyapatite) performed during the same time period (58%, 95% CI 52% to 66%) was worse than that of cemented THR (71%, 95% CI 62% to 80%), although the difference was not statistically significant. The risk of revision for aseptic loosening of group 1 cementless THR (0.49, 95% CI 0.32 to 0.74) was lower than that of cemented THR (p = 0.001).

Slovakia

One study³⁴⁵ reported findings from Slovakia from 2003 to 2010, including a total of 4970 primary THRs and 457 revisions. Cement was used for all components in 35.45% of all arthoplasties, 53.25% were cementless and 11.28% were hybrids. By 2010, the revision rate reached 9.20%, representing an annual increase of 1.1%. The revision rate in the whole observed period from 2003 to 2010 was 9.15%.

Norway

Espehaug et al.³³² studied differences by county and regional health authority over a 20-year period (1989–2008) using data from the Norwegian Arthroplasty Register. The authors observed an increase in the number of THR operations, from 109 operations per 100,000 inhabitants in the years 1991–5 to 140 in 2006–8. Variations were found across the four regions studied.

A second study from Norway³³³ reported the risks of revision after THR during a 21-year period among hip replacements reported to the Norwegian Arthroplasty Register. The risks of revision during the time periods 1993–7, 1998–2002 and 2003–7 were compared with that of the reference period 1987–92. There was an overall reduced risk of revision in the time periods 1993–7, 1998–2002 and 2003–7 compared with the risk of revision in the reference period. The improved results were due to a reduction in the incidence of aseptic loosening of the femoral and acetabular components in all time periods and in all subgroups of prostheses. The best results were obtained with the use of cemented prostheses. Analyses of revision for any cause were carried out for all prostheses together and separately for cemented, hybrid, reverse hybrid and cementless prostheses. The major cause of revision was aseptic loosening of one or both implant components.

One study used data from the Norwegian Arthroplasty Register (data from 1987 to 2008)³³⁵ to compare the difference in risk of THR revision from infection and change in risk over time. Data was from 1987 to 2008.³³³ Of the 84,492 THRs, 534 (0.6%) were revised for infection. Women had a significantly lower risk of revision for infection than men (RR 0.41, 95% CI 0.34 to 0.48). The cumulative 5-year survival rate was 99.5% in RA patients and 99.4% in OA patients (RR 0.98, 95% CI 0.65 to 1.48 for RA vs. OA patients) with revision for infection as the end point. The risk of revision for infection from 6 years postoperatively was higher in patients with RA.

USA

One study reported registry data from the USA.³³⁷ It examined patient and surgical factors associated with deep surgical site infection (SSI) following THR using data from the Kaiser Permanente Total Joint Replacement Registry between 2001 and 2009. A total of 30,491 THRs were included in the analysis, of which 17,474 (57%) were performed on women. The incidence of SSI was 0.51% (155/30,491), equating to a total of 155 deep SSIs, which occurred at a mean of 72 days (median 28, SD 93.3 days) after the procedure. Patient factors associated with SSIs included female sex, obesity and ASA grade ≥ 3.

Multinational

Sadoghi et al.³⁵⁰ compared primary THRs between different countries in terms of THR number per inhabitant, age and procedure type and compared survival curves including all THRs using data from nine registries. On average, the annual number of primary THRs per 100,000 inhabitants was found to be 133 for all ages, 26 for those aged < 55 years, 269 for those aged 55–64 years, 520 for those aged 65–74 years and 531 for those aged ≥ 75 years. The fixation method varied by country, for example in Sweden 67% of THRs are cemented whereas in Emilia-Romagna (Italy) 89% are cementless. Cementless fixation was more popular in Australia, Denmark, Emilia-Romagna, New Zealand and Portugal (50%) and cemented fixation was used more in Sweden and Norway (50%). Cemented and cementless fixations were used equally in England and Wales and Slovakia. The use of hybrid fixation was more uniform across countries and ranged from 8% in Portugal to 34.5% in New Zealand. Denmark showed the lowest survival rate within the first 15 years; however, THRs performed between 2006 and 2009 in Norway had similar low survival rates. All survival curves calculated in the study (except for Danish data) varied by < 1% within the first 9 years. Multivariate or subgroup analyses were not performed to compare the survival curves. The use of primary RS was not reported separately in the registries from Norway and Slovakia. Use of RS in the other countries varied from 1% in Portugal to between 2% and 3% in Denmark, Emilia-Romagna, New Zealand and Sweden to approximately 5% and 6% in Australia and England and Wales, respectively.

Graves et al.³⁴⁷ performed an investigation of the use of metal-on-metal THRs in the National Arthroplasty Registries of Australia, England and Wales and New Zealand. All registries reported an increased revision rate associated with larger femoral head size when metal-on-metal bearing surfaces were used.

The Nordic Registry includes the joint registries of Denmark, Sweden and Norway. One study³⁴⁸ aimed to compare demographics, choice of implant, fixation techniques and results between the countries, including a total of 280,201 THRs performed between 1995 and 2006. The study reported that 9596 THRs (3.4%) had later been revised. RS accounted for ≤ 0.5% of procedures in all countries. The 10-year survival rate was 92% (95% CI 91.6% to 92.4%) in Denmark, 94% (95% CI 93.6% to 94.1%) in Sweden and 93% (95% CI 92.3% to 93.0%) in Norway.

A second study reporting data from the Nordic Registry compared the survival of cemented THRs with metal femoral heads made from various materials (cobalt–chromium, aluminium and zirconium).³³⁴ The study reported prosthesis survival and relative revision risks adjusting for age, sex and diagnosis between 1987 and 2010. In total, 132,000 cases of THR were included in the analysis. At 12 years the survival rate was 88.1% for cobalt–chromium heads and 74.8% for zirconium heads. Aluminium femoral heads provided no advantage over cobalt–chromium heads for prosthesis survival. The authors concluded that cemented polyethylene THR with aluminium heads had a similar survival rate as the same THR with ceramic-on-ceramic heads when any revision was the end point.

Summary of the total hip replacement studies

The 22 THR studies reported the analysis of registry data from nine countries. These studies examined various aspects of the THR procedure, including revision and survival rates; different implants and combinations of implant bearing surfaces; and outcome measures such as reason for failure and patient differences associated with failure. Four of the 22 THR studies used registry data from multinational databases. Sadoghi et al.³⁵⁰ provided an extensive review of registries worldwide. They stated that fixation methods varied by country, with the cemented THR being most popular in Sweden and Norway and the cementless THR being most common in Emilia-Romagna (Italy) but also popular in Australia, Denmark, New Zealand and Portugal. Cemented and cementless fixations were used equally in England and Wales and Slovakia. In terms of survival rates, THRs carried out in Denmark showed the lowest survival rate within the first 15 years.