Collared versus collarless hydroxyapatite-coated stems for primary cementless total hip arthroplasty; a systematic review of comparative studies. Is there any difference in survival, functional, and radiographic outcomes?

Introduction: This systematic review aims to critically assess the literature comparative studies investigating collared and collarless Corail stem in primary total hip arthroplasty (THA) to find differences in revision rates, radiographic and clinical outcomes, and postoperative complications between these two types of the same stem. Methods: Eligible studies were found by searching PubMed, Science Direct/Scopus, and the Cochrane Database of Systematic Reviews from conception till May 2023. The PRISMA guidelines were followed. The investigation encompassed randomized controlled trials, case series, comparative, cohort, and observational studies that assessed at least one comparative outcome or complication between collared and collarless Corail stems. Results: Twelve comparative studies with 90,626 patients undergoing primary THA were included. There were 40,441 collared and 58,543 collarless stems. The follow-up ranged from 12 to 360 months. Our study demonstrated no significant difference in stem revision relative risk (RR = 0.68; 95% confidence interval (CI), 0.23, 2.02; p = 0.49), number of radiolucent lines (RR = 0.3; 95% CI, 0.06, 2.28; p = 0.29) and overall complication risk (RR = 0.62; 95% CI, 0.22, 1.76; p = 0.37) between collared and collarless stems. The collared stems demonstrated significantly lesser subsidence (mean difference: 1.01 mm; 95% CI, −1.77, −0.25; p = 0.009) and risk of periprosthetic fractures (RR = 0.52; 95% CI, 0.29, 0.92; p = 0.03). Conclusion: The comparative studies between collared and collarless stem groups showed similar survival and overall complication rates and functional outcomes. The similar revision rates between groups make the impact of higher subsidence for collarless stems uncertain. The lower risk of periprosthetic fractures in the collared stems group must be clarified further but could be related to increased rotational stability.


Introduction
Total hip arthroplasty (THA) stands as a successful procedure for hip osteoarthritis and over time, technological advancements, including different types of cementless fixation and design philosophies, have expanded the implant options, indications, and age-group target [1][2][3][4][5].The Corail stem Ò (DePuy Synthes) is a non-cemented, fully hydroxyapatite-coated femoral stem made of forged titanium alloy for hip reconstruction.It was introduced in the late 1980s and has excellent long-term published outcomes [6].The basic stem design, combining surface finish and full hydroxyapatite coating, prevents the release of metal ions and promotes osseointegration with the endosteal surfaces [7].The Corail stem has collarless and collared options.The collarless one has standard and high offset stem options for proper soft tissue tensioning and femoral offset restoration.The collared stem is available in standard and

OPEN ACCESS REVIEW ARTICLE
coxa-vara offset versions and is mainly used for patients with poor bone quality.
The Norwegian Joint Registry recently supported that the collared Corail stem has shown reduced revision risk, significantly adjusted health gain, and fewer revisions due to pain, periprosthetic fractures (PPFs), and dislocations compared to the standard collarless stem [8].Proponents of a collared Corail stem [9,10] support that this stem benefits the initial implant stability, allows faster postoperative full-weight bearing, protects against stem migration, and beneficially distributes the vertical forces through the collar into the medial calcar [9].Several studies supported that collared stems might have better survival rates, less risk of stem migration, and comparable functional outcomes than collarless stems [11][12][13].
This systematic review aimed to ascertain any outcome discrepancies, including survival, functional and radiographic outcomes between collared and collarless hydroxyapatite coated (Corail Ò ) stems for primary cementless THA, evaluating only comparative studies between these two types of the same femoral stem.

Search strategy
The present systematic literature review followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines [14] and was in line with the protocol agreed by all authors.This comprehensive review protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under CRD42023435176.

Eligibility criteria
Randomized controlled trials (RCTs), case series, comparative, cohort, and observational clinical studies that assessed at least one comparative survival, radiological, or clinical outcome between collarless and collared Corail stem groups of adult patients undergoing primary THA were included in this metaanalysis.The cup outcomes were not evaluated.Papers published in English and French with a minimum one-year follow-up were considered eligible for inclusion in the analysis.
Studies reporting outcomes on cemented Corail stem or other than Corail uncemented stem (DePuy Synthes, Warsaw, Indiana Ò ) were excluded from the study.Additionally, case reports, narrative or systematic reviews, meta-analyses, letters to the editor, conference proceedings, and in vitro and cadaver studies were excluded.

Study selection
Two reviewers (V.G., E.K.) searched the literature independently.Initially, the articles were analyzed and selected by title and abstract based on the inclusion criteria.After excluding studies that did not meet the inclusion criteria, full texts of the selected papers were evaluated.Any disagreements between the two authors were resolved via discussion and consensus with a third author (E.T.).Excluded studies and reasons for exclusion are listed in Appendix 1.

Data extraction and analysis
The same authors (V.G., E.K.) reviewed the papers separately and extracted the data for each included study.9) postoperative complications: mechanical failures, aseptic loosening (AL), infections, dislocations, PPFs, revisions for any reason.When data were missing, attempts were made to email the authors with up to three tentatives.Any discrepancies were resolved through consensus with the senior author.

Methodological quality assessment
Quality assessment of the included studies was performed using The Cochrane Risk of Bias 2 tool (ROB 2) for RCTs [15] and the Newcastle-Ottawa Scale (NOS) [16] for cohort studies.The NOS evaluates study cohort selection, comparability, and exposure-outcome relationships using a "star rating" system of up to nine stars.The ROB 2 tool assesses the risk of bias in RCTs across five domains where bias may be introduced.Two authors conducted the quality assessment individually, resolving disagreements via consensus.
Finally, one study [18] found a substantial difference in the mean difference of subsidence by 0.9 mm at 12 months.The overall mean subsidence of the collared stems was 0.87 mm against 2.07 mm of the non-collared stems.The collarless stems demonstrated a significantly greater overall mean subsidence of 1.01 mm than the collared stems (95% CI, À1.77, -0.25; p = 0.009).The I 2 index of the studies that assess the subsidence was 86%. Figure 3 illustrates the forest plot of the reported stem subsidence of the included studies.

Functional outcomes
Two studies assessed the postoperative WOMAC score [12,18].One study [12] demonstrated significantly greater WOMAC and short form 12 mental and physical outcomes favoring collared stems, while the University of California/ Los Angeles activity score was comparable between groups [12].One study (8.3%) [11] investigated the timed-up-and-go test and the average daily step count with no significant differences between groups.Three studies evaluated the OHS with no available comparative data.

Methodological quality of the included studies
All ten retrospective cohort studies were graded as good/ high quality.According to the ROB2 tool, the two RCT studies were rated as "low risk" of bias.Details of the studies' quality assessment can be accessed in Appendix 2.

Discussion
The current systematic review is an initial effort to investigate overall differences in outcomes between collared and collarless Corail Ò stems in patients undergoing primary THA.The work demonstrated that overall revision rate, postoperative complications, and functional outcomes were similar between collared and collarless stems; however, collared stems showed significantly lower mean subsidence and PPFs' risk.The clinical significance of higher subsidence for collarless stems remains uncertain and further investigation is required to understand the association between having a collar and lowered risk of PPF.

Survivorship
This meta-analysis did not reveal significantly different stem revision rates between collarless and collared stems, supporting a high survival probability for both stem design types in a long-term follow-up [6,26,27].In this review, over half of the studies showed a 100% survival rate with no complications or revisions for both stem types [11,12,17,18,[20][21][22].However, the other included studies in this review did not estimate the revision rates.Only one national register study [8] in our review found the significantly better 10-year performance of standard collared stems over standard collarless in terms of any reason, AL and PPFs.According to the same  study [8], 88.4% of the standard collarless stems were free of stem revision at a 30-year follow-up.The largest national registry study reviewed in our analysis [22] conducted a subanalysis comparing the survival rate between collared and collarless stems and found no significant differences.Despite the higher risk of subsidence for collarless stems, our study found that both stems have similar overall survival rates.

Radiographic outcomes
The necessity of a collar in uncemented arthroplasty to prevent clinically meaningful subsidence remains a subject of ongoing debate.The meta-analysis of the included studies demonstrated a 1 mm significantly higher subsidence of the collarless stems.The reported subsidence was inconsistent over time, and different radiostereometric or radiographic analyses were used in the included studies.Dammerer et al. [17] used the EBRA-FCA system to demonstrate non-significant sub-sidence for collarless implants at 12 months post-op but a statistically significant difference between collared and collarless implants only at 18 months.Perelgut et al. [12] reported significantly higher subsidence for collarless than collared stems at 13 months follow-up compared to the first postoperative day; however, this difference was insignificant when subsidence was compared to the fifteenth postoperative day.Other reports have previously suggested no significant subsidence difference between collarless and collared stems [18,25,28].The current study indicated that the collar mitigates against subsidence but does not prevent it completely.Collared stem subsidence may be due to difficulty achieving optimal contact between the collar and the calcar.Also, orthopaedic surgeons who use collared stems to prevent subsidence may unconsciously opt for smaller stem sizes in osteoporotic femurs to avoid intraoperative fracture.However, an undersized stem may increase the risk of subsidence and revision due to AL in the long-term [29,30].Surgeons using collarless stems may opt for the largest stem size to avoid subsidence, increasing the PPF risk [13,22].It remains uncertain whether the higher subsidence for collarless stems has clinical significance, as the overall revision rate of collared and collarless stems is comparable.
Radiolucent lines (RLLs) and aseptic loosening (AL) Our meta-analysis found no significant difference in the incidence of RLLs between collared and collarless stems despite less than half of the studies reporting on them.Previous studies have reported RLLs in patients with the Corail stem over a longterm follow-up [31].Reports suggested that using collared stems could result in better outcomes [21,24].A ten-year follow-up study [20] found a significantly higher number of RLL in collared than non-collared stems, regardless of the bearing type.A non-comparative study of 636 Corail stems with a median 6-year follow-up found a significantly lower RLLs prevalence in zone 7 in the collared (2.6%) compared to the collarless stems (23.6%) [21].This study suggested that using the proper size Corail stem improves stability and promotes osseointegration, preventing the RLLs development in zone 7.
Cadaveric and in vitro studies supported that collared stems improve stability and promote osseointegration, especially in osteoporotic patients, by increasing resistance against various forces at the bone/implant interface [9,32,33].However, clinical data are limited.Most implant migration occurs from the day of surgery to 2 weeks after the operation; it stabilizes thereafter, suggesting adequate fixation and a low AL risk in both collared and collarless stems [11].During a 2-year follow-up, collarless stems showed an initial migration of 0.73 mm in the first 6 months, but no further measurable subsidence was reported in subsequent measurements using radio stereometric studies [34].Clinically significant subsidence may occur if the stem size is underestimated and full weight bearing is allowed in the initial postoperative period [35].Both collared and non-collared stems must be implanted at the appropriate size to ensure proper osseointegration.An undersized collared stem cannot rely on a collar's protective role to osseointegrate [30].
The impact of stress shielding around the calcar region on RLLs remains unclear.It is supported that a stem collar may cause stress shielding in the lesser trochanter cortical bone, increasing calcar resorption and stress levels over time [36,37].Finite element analysis showed that conical collars increase stress transfer and reduce micromotion compared to flat collars [38,39].The collar's optimal contact with the calcar is crucial to load the medial cortex properly, minimising bone loss from stress shielding [36]; however, this is a challenging task during surgery [36,40].Besides, a collar could generate an impingement within the calcar region during stem subsidence, triggering a cantilever-like motion that could ultimately lead to stem failure [9,41].The review did not find evidence of stress shielding, and more radio-stereometric studies may be beneficial.

Functional outcomes
The authors could not conduct a meta-analysis due to the various functional scores used in the included studies.However, data from recent studies indicate no significant functional differences.An RCT reported no significant physical activity or function differences between collared and collarless stem patients [12].Karayiannis et al. [20] found no clinical impact on the OHS at a 10-year follow-up despite the significantly higher RLLs presence in collarless than collared stems.Magill et al. reported favorable outcomes in unrevised Corail stems, regardless of the presence of RLLs or collar [21].Another study demonstrated favorable HHS and visual analogue scales for both stem types, even in patients over 75 years old [42].

Overall complications
This meta-analysis did not find significant differences in overall complication rates between stem types; however, collared stems showed a significantly lower PPF incidence.Limited reports have supported that a potential protective collar function can significantly reduce the AL incidence in a longterm follow-up [8,43].However, AL is a multifactorial process that may be influenced by various factors throughout the stem lifespan [24,[43][44][45]; therefore, it cannot be thoroughly evaluated in the long-term follow-up.Our study found no AL risk difference between collared and collarless Corail stems, indicating the need for further studies.
The current study demonstrated that collared Corail stems had a lesser PPF risk than collarless ones.The recent Norwegian registry study [8] reported significantly lower PPF rates for the collared than collarless Corail stems in a long-term follow-up.It is important to exercise caution when interpreting these findings.Biomechanical, cadaveric studies and metaanalyses have shown that collared stems provide increased implant stability due to reduced rotational and varus forces on the bone-implant interface [9,27,46].As a result, collared stems are primarily utilized in elderly patients with osteoporotic femurs, as they are thought to offer protection against subsidence and early failures [47,48].However, it is difficult to explain the reduced PPF risk.Most early postoperative PPFs occur intraoperatively, while late PPFs are mainly osteoporotic, depending on bone quality.When using an uncemented collarless stem, surgeons may opt for the largest possible stem size to achieve the best primary stability.However, this may increase the PPF risk, which could partly explain the difference in PPF rates between collared and non-collared stems.On the other hand, a surgeon using a collared, uncemented stem could choose a slightly smaller stem size and still rely on the collar to prevent subsidence.Additionally, this approach could potentially lower the PPF risk [30].Besides, the collared and collarless stems were traditionally used for different age populations, making direct comparisons between these two stem groups in registry results challenging.Collared stems were mainly used for the elderly population or patients with distinct anatomical characteristics according to Dorr's classification [49,50].Further clarification regarding the lower risk of PPFs in the collared stem group is needed.

Limitations
Our study has limitations.Firstly, the included papers were of low quality, primarily consisting of retrospective studies.Secondly, heterogeneity among studies and measurement bias 8 V. Giovanoulis et al.: SICOT-J 2024, 10, 8 may affect the meta-analysis interpretation due to variations in definitions and outcome reporting.Additionally, the analysis did not consider bearing surfaces, age/sex grading, or differences in follow-up periods, countries, and surgical approaches that could affect reoperation rates.However, the strength of the present work is that we analyzed a substantial number of stems following the appropriate methodology provided by the PRISMA guidelines.

Conclusions
In a long-term follow-up, both collared and collarless Corail stems showed excellent survivorship rates, similar radiographic and functional outcomes, and overall complication rates.Collared stems offer better protection against subsidence, but the clinical significance of this is still unclear and requires further evaluation.The lower risk of PPFs in collared stem cases may be due to increased rotational stability but should be further clarified.The decision to use a collared stem remains dependent on the surgeon's preferences, and the need for more studies in the future is evident.

Figure 1 .
Figure 1.Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flowchart.

Figure 2 .
Figure 2. Forest plot for comparison of (a) stems revisions and (b) overall complications between collared and collarless stems.

Figure 3 .
Figure 3. Forest plot for comparison of (a) subsidence, (b) RLLs, and (c) PFFs between collared and collarless stems in the included studies.

Table 2 .
Operative and implant characteristics of the included studies in the meta-analysis.

Table 1 .
Demographics and other study characteristics of studies included in the meta-analysis.

Table 3 .
Postoperative complications of the patients in the included studies.