Introduction

Surgical approaches in total hip arthroplasty (THA) include anterior [17, 26, 33, 40, 43, 48, 50, 58, 60], lateral [1, 13, 27, 61, 63] and direct lateral and posterior approach [2, 7, 8, 18, 54, 64]. Early research [42] shows several advantages of the direct anterior approach (DAA) in THA that claims to be as effective but less invasive than the posterior approach (PA). However, the higher risk of femoral fracture and soft tissue damage cannot be underestimated. Due to the difficult femoral exposure and possible complications related to femoral preparation, this approach can result in a higher rate of undersized stems when compared to other approaches [6, 10, 22, 29, 34, 47]. The present authors believe that the femoral implant design (collar [3,4,5] or collarless stem [6, 10, 22, 29, 34, 47], long [3,4,5,6, 10, 22, 29, 34, 47] or short stem [8, 15, 20, 30, 38, 39, 41, 49]) in the collared femoral stem may relate to a lower rate of stem subsidence and limb length discrepancy (LLD) in mid-term to long-term follow-up when compared to collarless femoral stem. However, currently there is no consensus as to which femoral implant design is the most suitable for DAA in THA. Therefore, this systematic review and meta-analysis aim to assess and compare postoperative complications and revision rates due to neurapraxia, wound complications, LFCN and LLD after DAA using collared compared to collarless femoral stem and short femoral stem compared to long femoral stem in THA. These clinical outcomes consist of the postoperative complications (neurapraxia, wound infection, LFCN, hematoma, artery injury, cup malposition, embolism, fracture and implant loosening) and revision rates due to dislocation, periprosthetic fracture and implant migration.

Material and method

Medline and Scopus databases were used to identify relevant studies published in English since the date of inception to June 6, 2018. The PubMed and Scopus search engines were used to locate studies with the following search terms: [(DAA OR direct anterior approach) AND total hip arthroplasty]. Search strategies for Medline and Scopus are described in detail in “Appendix1”. References from the reference lists of included trials and previous systematic reviews were also explored.

Inclusion criteria

Clinical studies (e.g., observational, cross sectional, cohort or RCT) that reported clinical outcomes of DAA using collar compared to collarless femoral stem and short femoral stem compared to long femoral stem in THA were eligible if they met the following criteria:

  • Reported at least one of the following outcomes: complications (neurapraxia, wound infection, LFCN, hematoma, artery injury, cup malposition, embolism, fracture and implant loosening) and revision rates due to dislocation, periprosthetic fracture and implant migration.

  • Had sufficient data to extract and pool, i.e., the reported mean, standard deviation (SD), the number of subjects according to treatments for continuous outcomes, and the number of patients according to treatment for dichotomous outcomes.

The reference lists of the retrieved articles were also reviewed to identify publications on the same topic. Where there were multiple publications from the same study group on the same population, the most complete and recent results were used. Non-English studies were excluded.

Data extraction

Two reviewers (P.P. and J.K.) independently performed data extraction using standardized data extraction forms. General characteristics of the study [i.e., mean age, gender, body mass index (BMI), mean follow-up time, study design, type of approach (MIS or standard), implant design (short or long stem, collar or collarless) and fixation method (cemented or cementless)] were extracted. All dichotomous outcomes (postoperative complications and revision of femoral stem due to neurapraxia, wound complications, LFCN and LLD) were also extracted. Any disagreements were resolved by discussion and consensus with a third party (A.A.).

Outcomes of interest

The outcomes of interest included postoperative complications and revision of the femoral stem. These outcomes were measured as reported in the original studies which were postoperative complications (neurapraxia, wound infection, LFCN, hematoma, artery injury, cup malposition, embolism, fracture and implant loosening) and revision rates due to dislocation, periprosthetic fracture and implant migration were considered.

Statistical analysis

For dichotomous outcomes (complications and revision), the prevalence was pooled and calculated using the inverse variance method as follows [55]: \( \bar{p} = \frac{{\sum w_{i} p_{i} }}{{\sum w_{i} }} \) where p was the pooled prevalence, pi was the prevalence of complications of each study, wi was 1/var(pi), which was the weight of each study. Heterogeneity of prevalence across studies p was checked as follows: \( \sum w_{i} \left( {p_{i} - \bar{p} } \right)^{2} \) The Q statistic follows a \( \chi^{2} \) distribution with number of studies (k) − 1 degree of freedom (d.f.). The degree of heterogeneity was also quantified using the I2 statistic [23]. This value can range from 0 to 100%, the closer to 100%, the higher the heterogeneity. If heterogeneity was present, between studies variation was then estimated as follows: \( \tau^{2} = \frac{{Q - \left( {k - 1} \right)}}{{\sum w_{i} - \frac{{\sum w_{1}^{2} }}{{\sum w_{1} }}}} \) if Q k 1 or 0 otherwise. This was used to calculate a weight term that accounted for variations between studies \( w_{i}^{*} = \frac{1}{{\text{var} (p_{1} ) = \tau^{2} }} \), and then the pooled prevalence was estimated using the random effects model as follows: \( 95\% {\text{CI}} = \bar{p} ^{*} \pm \frac{1.96}{{\sqrt {\sum w_{i}^{*} } }} \). Meta-regression analysis was then applied to explore causes of heterogeneity [23, 56]. Coverable parameters, i.e., mean age, gender, body mass index (BMI), mean follow-up time, study design, type of approach (MIS or standard), implant design (short or long stem, collar or collarless) and fixation method (cemented or cementless) were considered in the meta-regression model. Power of the test for meta-regression was also assessed [51]. The unstandardized mean difference and odds ratio (OR) were estimated by indirect meta-analysis using a random effects model, otherwise a fixed effects model was applied. All analyses were performed using STATA version 14.0 [57].

Results

Three hundred and twenty-seven and 338 studies were identified from Medline and Scopus, respectively, as shown in Fig. 1. Two hundred and sixteen studies were duplicates, leaving 449 studies for review of titles and abstracts. Of these, 34 articles [3,4,5,6, 8, 10, 11, 15, 16, 19,20,21,22, 24, 25, 28,29,30,31, 34,35,36, 38, 39, 41, 44, 46, 47, 49, 53, 54, 62, 65, 66] were relevant, and the full papers were retrieved. Characteristics of these studies are given in Tables 1 and 2. Thirty-four studies [3,4,5,6, 8, 10, 11, 15, 16, 19,20,21,22, 24, 25, 28,29,30,31, 34,35,36, 38, 39, 41, 44, 46, 47, 49, 53, 54, 62, 65, 66] were included for the analysis of DAA in THA; 23 studies [3,4,5, 10, 11, 16, 19,20,21,22, 24, 25, 28,29,30, 35, 36, 38, 41, 49, 53, 62, 65] were retrospective cohort, four studies [6, 34, 46, 47] were prospective cohort, and seven studies [8, 15, 31, 39, 44, 54, 66] were RCTs. Thirty-one studies [6, 8, 10, 11, 15, 16, 19,20,21,22, 24, 25, 28,29,30,31, 34,35,36, 38, 39, 41, 44, 46, 47, 49, 53, 54, 62, 65, 66] and three studies [3,4,5] were included for analysis of collarless and collared femoral stem. Twenty-six studies [3,4,5,6, 10, 11, 15, 16, 19, 21, 22, 24, 25, 28, 29, 31, 34,35,36, 44, 46, 47, 53, 54, 62, 65, 66] were long femoral stem, and eight studies [8, 20, 30, 38, 39, 41, 49] were short femoral stem. Twenty-seven studies [3,4,5,6, 8, 10, 15, 16, 19,20,21, 25, 28, 30, 31, 35, 36, 39, 41, 44, 46, 49, 53, 54, 62, 65, 66] were cementless fixation, seven studies [11, 22, 24, 34, 38, 47] were fixation with both cemented and cementless, and one study [29] was cemented fixation. Twenty-six studies [4,5,6, 8, 10, 15, 16, 19, 21, 22, 25, 28,29,30,31, 34, 36, 38, 39, 41, 44, 46, 49, 53, 62, 66] were standard approach, and eight studies [3, 11, 20, 24, 35, 47, 54, 65] were minimally invasive surgery approach. Twelve studies were primary osteoarthritis (OA), seven studies were primary OA and osteonecrosis, one study was hip dysplasia, and seven studies did not mention the cause of pathology. Mean age, BMI, mean follow-up and percentages of female gender of long-stem DAA participants varied from 58.1 to 69.8 years, 22.9 to 30.7 kg/m2, 1 to 94 months and 32.6 to 92.8%, while short stem varied from 58.4 to 67.4 years, 26.6 to 31.1 kg/m2, 1.5 to 96 months and 39.1 to 60.5%.

Fig. 1
figure 1

Flow of study select

Full size image
Table 1 Characteristics of included studies of DAA in long stem
Full size table
Table 2 Characteristics of included studies of DAA in short stem
Full size table

Pooled prevalence of complications and revision between collar and collarless DAA

Overall, there were 6825 patients (6457 in the collarless group and 368 in the collar group). A total of 469 and 66 patients had complications and revision in the collarless group and no patient had complications and revision in the collar stem group (Supplement Table 1). The total complication and revision rate per patient was 5% (95%CI 3.3%, 7%) and 0.9% (95%CI 0.6%, 1.2%) in the all patients. The complication rate and revision rate were 5.7% (95%CI 3.8%, 7.7%) and 0.9% (95%CI 0.6, 1.2) in the collarless group. There was no prevalence of complication and revision in the collared stem group. By indirect meta-analysis, collared femoral stem provided a lower risk of having complications of 0.02 (95%CI 0.001, 0.30) when compared to collarless femoral stem. In terms of revision, there was no statistically significant difference in collared femoral stem compared to collarless femoral stem (Fig. 2, Table 3).

Fig. 2
figure 2

Comparisons of prevalence of complications and revisions between short versus long stem or collar versus collarless stem

Full size image
Table 3 Estimation of the pooled prevalence of complications and revisions of collar versus collarless DDA approaches
Full size table

Pooled prevalence of complications and revision between short- and long-stem DAA

Four thousand two hundred and eighty patients in long stem and 2545 patients in short stem were selected. The complication rate and revision rate were 10.2% (95%CI 9%, 11.4%), 0.7% (95%CI 0.3%, 1%) and 5.2% (95%CI 3.1, 7.2), 1.5% (95%CI 1%, 2%) in short and long femoral stem, respectively (Supplement Table 1). By indirect meta-analysis, the long femoral stem had a lower risk of having complications 0.57 (95%CI 0.48, 0.68) when compared to short femoral stem. In terms of revision, there was no statistically significant difference in long femoral stem compared to short femoral stem (Table 4).

Table 4 Estimation of the pooled prevalence of complications and revisions of short- versus long-stem DDA approaches
Full size table

Sources of heterogeneity

Meta-regression was applied for exploring the cause of heterogeneity by fitting a co-variable (i.e., age, percentage of female patients, BMI, follow-up time and type of disease and study quality), and meta-regression was applied to assess this. None of the co-variables could explain the heterogeneity (Table 5).

Table 5 Comparisons of prevalence of complications and revisions between short and long stem or collar and collarless stem
Full size table

Discussion

From the current available evidence, this systematic review and meta-analysis have shown the following: collared femoral stem provides a lower risk of having complications of 0.02 (95%CI 0.001, 0.30) when compared to collarless femoral stem. Long femoral stem had a lower risk of complications by 0.57 (95%CI 0.48, 0.68) when compared to short femoral stem. In terms of revision, there was no statistically significant difference in collared femoral stem compared to collarless femoral stem and long femoral stem compared to short femoral stem.

The issue of increased risk of revision for femoral-related complications with the anterior approach has been debated in the literature recently. Many studies [9, 14, 32, 37, 45, 59] reported that increased rate of revision for early femoral failure secondary to loosening or fracture in patients who had undergone total hip arthroplasty via the DAA approach compared to the posterior approach. An interesting finding of these studies was the effect of stem design on the rate of femoral loosening or fracture such as short stem [52], undersized stem [45] and collarless tapered stem [9, 52]. The hypothesis on the increased rate of loosening with the tapered wedge inserted via an anterior approach is that there is an increased risk of loosening due to failure to obtain adequate initial stability and subsequent ingrowth. In the anterior approach, broaching trajectory is not always linear; this may be secondary to patient body habitus, femoral exposure, or the patient’s proximal femoral anatomy. With a thin, collarless tapered stem design, nonlinear broach insertion and extraction can create an anterior metaphyseal gap that compromises initial stability leading to subsidence, distal potting and failure of proximal ingrowth [9]. Our study found that the collared stem lowered the risk of loosening or fracture when compared to the collarless stem, which is a similar result to some large retrospective studies [9]. A collared stem offers numerous theoretical advantages such as reduced subsidence, better rotational stability, and lower risks of calcar fracture propagation [9]. These theories were confirmed in a cohort study in which Demey et al. [12] found that collared Corail stems were able to withstand greater vertical and horizontal forces before the initiation of subsidence and subsequent fracture. As the femoral stem is at greatest risk of subsidence or early fracture before secondary fixation (osseointegration), it is possible that the improved immediate stability conferred by collared stems allows more rapid bony ingrowth to achieve secondary fixation. This may be of increased importance with the current emphasis on early functional recovery after THA. Moreover, Cidambi et al. [9] reported that the implantation of a collared HA-coated compaction broached stem or metaphyseal fit and fill stem is now routinely used at their center for primary THA via the DAA because of the aforementioned benefits. Therefore, our results confirmed that collared femoral stem should be used as implant of choice to decrease risk of femoral-related complications in DAA for THA.

The strength of this study is that we use adequate methodology of systematic reviews in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [106] as well as providing exploration and reduction in the heterogeneity of the studies using subgroup analysis and adequate statistical analysis.

Moreover, this study has conclusive evidence about long femoral stem and collar femoral stem should be selected to decrease complication and revision after DAA for THA.

There are some limitations in this study. First of all, the quality of studies for the meta-analysis was not high. Ideal evidence for systematic review is a randomized controlled trial (RCT), which is most commonly used in testing the efficacy of surgery. The quality of data available is relatively poor, with a predominance of non-comparative retrospective studies. There were no randomized controlled trials or prospective comparative studies. This could be a possible source of bias between groups due to the opportunity for selection, different baseline characteristics and likely for publication bias. Secondly, heterogeneity remains an important factor to be considered in the conduct and interpretation of meta-analysis, and the heterogeneity between studies was great. We applied the random effects meta-analysis to adjust for the differences between studies, and the possible causes of heterogeneity were explored if covariate data at baseline (e.g., age, percentage of female patients, BMI, follow-up time, type of disease and study quality) were available. The third limitation is that there is also a measurement bias, as the studies differed in their definition and reporting of complications. The fourth limitation is that indirect meta-analysis was used for calculating the mean difference and odds ratio between the two groups, due to the fact that most included studies were case series reports of only one technique. The fifth limitation is that there are other outcomes of interest that can be used to compare collared and collarless stem such as operation cost and quality of life. However, these factors could not be analyzed because of insufficient data. The sixth limitation is that among studies, the follow-up time varied, which could affect reports of reoperation rates. Finally, several studies did not make specific mention of certain complications. In this study, it was assumed that those complications were not present rather than not reported for the primary outcome analysis. This is a potential source of bias.

In conclusion DAA THA, collared femoral stem and long femoral stem had better complication rates when compared to collarless femoral stem and short femoral stem with direct and indirect meta-analysis methods. However, in terms of revision rates, there were no differences between all femoral stems (short versus long and collared versus collarless). Prospective randomized controlled studies are needed to confirm these findings as the current literature is still insufficient.