How to Prevent Aseptic Loosening in Cementless Arthroplasty: A Review

: Aseptic loosening is the main late postoperative complication of cementless total hip arthroplasties (THAs), leading to pain and functional impairment. This article aims to update the orthopedic surgeon on the various methods by which the aseptic loosening rate can be reduced. We performed a systematic review by searching the PubMed database for hip aseptic loosening. We included meta-analysis, randomized controlled trials, reviews, and systematic reviews in the last 10 years, which provided information on techniques that can prevent aseptic loosening in total hip arthroplasty. From a total of 3205 articles identiﬁed, 69 articles (2%) met the inclusion criteria, leading to a total of 36 recommendations. A lot of research has been conducted in terms of septic loosening in the last decade. Currently, we have various techniques by which we can reduce the rate of aseptic loosening. Nevertheless, further randomized clinical trials are needed to expand the recommendations for aseptic loosening prevention. the review of studies dealing with the factors affecting aseptic loosening in cementless total hip arthroplasty.


Introduction
Total hip arthroplasty (THA) is one of the most frequently performed orthopedic procedures worldwide, being used to treat hip osteoarthritis and hip fractures. The overall results of THAs are very good in the majority of cases, resulting in pain decrease, as well as improvement of mobility and hip function. There are two main types of THAs according to bone-implant fixation: cemented and cementless. There are still debates regarding the indications of the two methods of fixation. Generally, cemented THAs are chosen when poor bone quality is present, while cementless THAs are indicated in the case of good bone quality. Still, cementless fixation is gaining more indications in the last decade, thus becoming the most used method of fixation in the case of primary THAs. Nevertheless, there is still debate regarding the survivorship in cementless vs. cemented total hip replacements. A study on 170,413 THAs showed that cementless THAs had a higher risk of aseptic loosening compared to cemented THAs (RR = 1.5) [1].
Complications of cementless THAs vary from mild to life-threatening ones. The most frequent late complication is represented by aseptic loosening, meaning a failure at the implant-bone interface, without the presence of infection. This complication leads to pain and functional impairment of the involved hip, severely decreasing the patient's quality of life. A key factor in preventing aseptic loosening is a good connection between the bone and the implant, a process called osseointegration.
The treatment of aseptic loosening is generally represented by a revision total hip arthroplasty. This is a complex surgery performed in specialized centers, while the functional results are generally worse compared to a primary total hip replacement [2]. Moreover, revision surgery involves a huge cost to the healthcare system.
Researchers and clinicians are constantly trying to discover new ways to prevent this complication and therefore increase the survivorship of total hip arthroplasties. This review aims to update both clinicians and researchers on the advances in the field of aseptic loosening prevention in the last decade.

Pathophysiology of Aseptic Loosening
To prevent or treat the aseptic loosening of a cementless total hip replacement, we must first understand the pathophysiology of this process. Aseptic loosening can have different causes, that are (1) improper initial fixation, (2) mechanical loss of fixation over time, and (3) biological loss of fixation due to particle debris [3].
An improper initial fixation, or osseointegration, can have the following causes: implant properties, bone quality, surgical technique, host's healing process, and loading conditions [4]. The main cause of aseptic loosening is represented by wear debris from the implant's bearing surface, which further initiates an inflammatory response from the host's immune system [5]. This effect increases the bone resorption activity of osteoclasts and decreases the bone-forming process of osteoblasts [4]. The key mechanism which activates the osteoclast resorption process is the receptor activator of the nuclear factor-κB (RANK)/RANK ligand (RANKL) pathway [3]. This pathway is an important target for modern therapies.

Materials and Methods
A systematic review of the MEDLINE database was performed for articles published between 2011 and 2021 containing the terms "loosening hip". We included meta-analysis, randomized controlled trials, reviews, and systematic reviews. The authors reviewed the abstract of each article and only English-written articles were considered. All studies involving preoperative, intraoperative and postoperative factors that affect aseptic loosening in cementless total hip arthroplasty were included. More specific, implant properties (design, material, bearing surfaces), surgical factors, patient-related factors (comorbidities, demographics, genetics, habits), postoperative protocol (activities, weight-bearing), and use of perioperative drugs (antiosteoporotic drugs, opioids, statins, beta-blockers) were all included in our review. Studies on the aseptic loosening of both the acetabular component and the femoral stem were taken into account. Regarding the exclusion criteria, studies where only physiopathology, diagnosis or treatment of aseptic loosening were addressed were not included in our review. Moreover, only studies performed on cementless primary total hip replacements were included, while loosening of mega-prosthesis, hip revisions, hemiarthroplasties or hip resurfacing were not considered. The PRISMA flowchart is available in Figure 1.

Implant Properties
The cementless total hip replacements systems are made up of four elements: the acetabular cup, insert, femoral head, and femoral stem. The most common alloy used in acetabular cups and femoral stems is 90Ti-6Al-4V ELI. The acetabular insert can be made up of polyethylene or ceramic, while the femoral head can be made up of metal alloy or ceramic. The interaction of the femoral head and the insert is called the bearing surface. The bearing surfaces available today are ceramic-on-ceramic, ceramic-on-polyethylene and metal-on-polyethylene. There are three main characteristics of the implant which can influence the rate of aseptic loosening: implant design, alloy, and surface treatment. Implant design has a direct impact on the survivorship of an uncemented total hip replacement according to multiple studies. Different femoral stem designs, which can be classified according to the Mont group, have different contact areas between the implant and bone [6]. So far, there have been described three generations of acetabular cups and seven femoral stem types [6]. The contact area can affect the primary osseointegration process and even further, the rate of aseptic loosening.
The metal alloys used directly influence the implant's biocompatibility, Young's modulus of elasticity, strength, and corrosion resistance, which all have a great impact on implant survivorship, as previously shown [7]. Moreover, surface treatments such as hydroxyapatite coating, acid etching, alkali heat treatment, plasma treatment, ion implementation, sandblasting, and nanotechnology can prevent aseptic loosening by affecting the primary osseointegration process [7]. We will further describe the most recommended implant characteristics in terms of reducing the rate of aseptic loosening.

Implant Properties
The cementless total hip replacements systems are made up of four elements: the acetabular cup, insert, femoral head, and femoral stem. The most common alloy used in acetabular cups and femoral stems is 90Ti-6Al-4V ELI. The acetabular insert can be made up of polyethylene or ceramic, while the femoral head can be made up of metal alloy or ceramic. The interaction of the femoral head and the insert is called the bearing surface. The bearing surfaces available today are ceramic-on-ceramic, ceramic-on-polyethylene and metal-on-polyethylene. There are three main characteristics of the implant which can influence the rate of aseptic loosening: implant design, alloy, and surface treatment. Implant design has a direct impact on the survivorship of an uncemented total hip replacement according to multiple studies. Different femoral stem designs, which can be classified according to the Mont group, have different contact areas between the implant and bone [6]. So far, there have been described three generations of acetabular cups and seven femoral stem types [6]. The contact area can affect the primary osseointegration process and even further, the rate of aseptic loosening.
The metal alloys used directly influence the implant's biocompatibility, Young's modulus of elasticity, strength, and corrosion resistance, which all have a great impact on implant survivorship, as previously shown [7]. Moreover, surface treatments such as hydroxyapatite coating, acid etching, alkali heat treatment, plasma treatment, ion implementation, sandblasting, and nanotechnology can prevent aseptic loosening by affecting the primary osseointegration process [7]. We will further describe the most recommended implant characteristics in terms of reducing the rate of aseptic loosening.

Alloy
Trabecular metal acetabular cups have gained a lot of popularity in the last decade. This alloy showed significantly less radiolucency when compared to porous-coated titanium acetabular cups, according to a study performed by Wegrzyn et al. on 113 patients [8]. Moreover, it reduces the risk of aseptic loosening compared to non-trabecular metal cups, as shown in a study on 12,800 primary THAs at a five-year follow-up [9].

Surface Treatment
Hydroxyapatite-coated cups have a similar risk for aseptic loosening as the uncoated ones, as shown by a study on 28,605 THAs from the Nordic Arthroplasty Register Association [10]. On the other hand, a study on 20,993 primary THA with highly porous acetabular components has shown a lower incidence of aseptic loosening in long term [6]. On the contrary, Palomäki et al., after analyzing 6080 primary THAs, showed that the ultraporous-coated acetabular component has an increased risk of revision compared to conventional uncemented cups [11].

Design
Acetabular cups without screw holes have a lower rate of revision compared to ones with screw holes, as shown by Otten et al. in a study on 22,725 primary total hip replacements [12]. The authors showed that the rate of aseptic loosening was not statistically significant between the two, but a limitation of the study was a mean follow-up of only 3.4 years (0-18) [12]. On the other hand, randomized clinical trial on 100 patients showed no difference between solid cups and cluster hole cups at 10 years follow-up [13].

Alloy
Titanium alloy stems have almost half of the proximal bone loss compared to cobaltchromium (CoCr) stems, according to a systematic analysis performed by Knutsen et al. [15].

Surface Treatment
Hydroxyapatite coating did not influence the revision rate at 10 years when compared to porous coating or rough sand-blasted stems, as shown by a study that analyzed 152,410 primary total hip replacements from the Nordic Arthroplasty Register [16]. Nevertheless, a meta-analysis by Chen et al. showed that although the survival rate of hydroxyapatite coating is similar to porous coating, the femoral osteolysis is greater in the porous group [17]. Cementless plasma-sprayed porous titanium hydroxyapatite stems have a 98.5% survivorship in patients aged 70 or older for femoral neck fractures, on a study performed on 293 patients (301 hips) [18].

Design
The femoral stems are divided into seven types, according to geometry, as following: type 1 (straight stem, single wedge), type 2 (straight stem, double wedge), type 3 (straight stem, tapered), type 4 (straight stem, cylindrical, fully coated), type 5 straight stem, (modular), type 6 (curved stem, anatomic stem) and type 7 (short stem) [19]. Type 1, type 2, type 6 and type 7 have an exclusively metaphyseal fixation, while type 3 and type 5 have a metaphyseal and diaphyseal fixation [19]. Type 4 has a primarily diaphyseal fixation. A study in 2011 by Khanujah et al. showed that types 1, 2, 4 and 7 femoral stems have better survival, as follows: 98.8% at 17 years; 99.2% at 17 years, 97.9% at 15 years and 98.2% at 10 years, respectively [19]. Nevertheless, type 1, type 2 and type 4 are associated with a higher rate of stress shielding and thigh pain [19]. Some shortened and shoulderless double-taper press-fit stems have a high rate of aseptic loosening (1,4% at five years), as shown by Wagner et al., in a study on 517 Dorr type A patients [20]. Straight stems had less bone loss compared to anatomical, tapered, and press-fit designs, according to a systematic analysis performed by Knutsen et al. [15]. Regarding the femoral sizes, smaller Corail stem (sizes 8 and 9) have four times the rate of revision as compared to larger femoral sizes (10-20 stems) as found by a study on 42,265 primary THAs [21].

Bearing Surfaces
The micromotion between the acetabular cup and inlay produces wear debris which generate a biologic host response responsible for a periprosthetic bone loss that further leads to aseptic loosening [22].
Ceramic-on-highly cross-linked polyethylene (CoPx) has the best survivorship (0.54/ 100-component-years), followed by metal-on-polyethylene according to New Zealand Joint Registry study performed on 106,139 THAs [23]. This result is according to another study by Hu et al. performed on 974 primary THAs [23]. Metal-on-metal bearings had the lowest survivorship (1.43/100-component-years) [23]. This statement is supported by a meta-analysis performed by Lee et al. in which metal on metal (MOM) was associated with a higher risk of revision compared to a ceramic-on-ceramic (COC) bearing [24]. When comparing ceramic-on-polyethylene and ceramic-on-ceramic, a meta-analysis by Shang et al. on 2702 patients showed that the results are comparable in terms of revision rate at a long-term follow-up [25]. On the other hand, an analysis of the Dutch Registry on 33,454 patients pointed out a higher 2-year cup revision rate in the ceramic-on-ceramic group compared to ceramic-on-polyethylene (HR = 0.475) [26]. A study on 11,096 patients with primary THAs showed no difference at an 8.7-year follow-up in revision rates between ceramicon-ceramic compared to metal-on-polyethylene [6]. On the contrary, on a follow-up of 20 years, ceramic-on-ceramic bearing had a lower rate of aseptic revision rate compared to metal-on-polyethylene [27,28]. These studies suggest that the best bearing in terms of survivorship have the following order: ceramic-on-polyethylene, ceramic-on-ceramic, metal-on-polyethylene and metal-on-metal.
Another bearing which is gaining increasing interest in recent years is the dual mobility construct. A review which included 448 primary dual mobility cups at a mean follow-up of 7.6 years showed that aseptic loosening rate was less compared to the unipolar group (OR = 1.21, p = 81) [29,30]. Another study on 150 primary dual mobility THAs and 166 unipolar THAs at a mean follow-up of 31 months showed that dual mobility bearings had a lower rate of aseptic loosening (0.7% vs. 5%) [30]. An analysis of the Nordic Arthroplsaty Register Association on over 2000 matched patients concluded that there was no difference in overall risk of revision between dual mobility cups and metal-on-polyethylene or ceramic-on-polyethylene bearings [31]. Another review on mid-term follow-up of more than 500 dual mobility cups showed that there was no statistically significant difference in aseptic loosening rates compared to unipolar cups [32]. A systematic review of Darrith et al. on more than 10,000 primary dual mobility cups resulted in an aseptic loosening incidence of 1.3% at a mean follow-up of 8.5 years [33]. The only long-term study we found included 85 dual-mobility total hip replacements performed on obese patients and no aseptic loosening was found at a follow-up of 15 years [29,30]. More long-term follow-up studies are needed in order to properly assess the aseptic loosening rate.
Other UHMWPE have been recently introduced, with promising results, such as the X-linked UHMWPE, but further studies are needed in order to decide on the long-term effect on polyethylene wear and aseptic loosening.

Surgeon
In terms of surgeon, a study on 116 low revision rate surgeons (<3.84% revision rate at 5 years) and 433 other surgeons, showed that low revision rate surgeons use fewer types of implants compared to the higher revision rate surgeons [28,34]. Moreover, the same study pointed out that lower revision rate surgeons were more likely to conduct more than 100 total hip replacements per year [34]. In terms of experience, lower revision rate surgeons were related to a higher experience in terms of years and cases per year, but the differences were subtle [34].

Hospital
Regarding teaching-hospitals vs. non-teaching hospitals, it was proved that resident training does not influence the complication rate at a mean follow-up of 97 months [35].

Approach
One of the main factors affecting the rate of aseptic loosening is, surprisingly, the approach. A study on 48,716 primary cementless THAs showed that the anterior approach had the highest rate for femoral stem loosening [36]. The same result was found by a meta-analysis performed by Huang et al. in 2021 and by Hoskins et al. on a study of 122,345 patients, who showed a higher rate of early major revisions [37,38]. A meta-analysis by Docter et al. showed that the posterior approach has the lowest risk for loosening when compared to lateral and anterolateral approaches [39]. On the contrary, a study from the Norwegian Arthroplasty Register on 21,860 THAs showed that the risk of revision was not increased in the case of anterior minimally invasive surgery or anterolateral approach when compared to lateral or posterior approaches [40].

Acetabular Screws
Regarding the use of acetabular screws, they can lower the risk of revision surgery as shown by Natera et al. on 749 cases of primary THAs [41].

Demographics
A study on 63,158 patients showed that when performing the surgery at the age of under 70 years old, the rate of revision is up to 35%, compared to 5% in the case of patients aged 70 years or older when performing the THA [42]. A study showed that female patients have a significantly increased periarticular remodeling (Ot.Lac.Ar = osteocyte lacunae area) compared to male patients in the samples retrieved during total hip replacement revision surgery [43]. On the other hand, a systematic review showed that seven out of eight papers resulted in an increased risk for revision surgeries in male patients [44].

Obesity
Obesity increases the overall risk of early revision of primary THAs due to aseptic loosening 4.7-fold, as shown by Electricwala et al. on a study on 273 revision THAs [48]. Moreover, a study on 684 revision THAs showed that obesity increases the risk of early aseptic loosening in primary THAs to 30%, compared to 18% in nonobese patients, [49]. A meta-analysis by Haverkamp et al. including 15 studies, showed that obese patients have an odds ratio of 0.64 to develop aseptic loosening compared to nonobese patients [50].

Rheumatoid Arthritis
In the case of rheumatoid arthritis, acetabular cup migration is significantly higher compared to osteoarthritic patients [52]. Another study showed that higher inflammatory disease activity in patients with rheumatoid arthritis increases the risk for radiographic loosening [52,53].

Avascular Necrosis of the Femoral Head
When performing a DXA scan of the periprosthetic area in the case of avascular necrosis of the femoral head, Gruen zones 1 and 7 had a statistically significant lower BMD when compared to primary osteoarthritis [54]. This can lead to an impaired implant fixation.

Hip Dysplasia
When hip dysplasia is present, nonmodular conical stem design increases the risk of cup aseptic loosening (2.5%) [55].

Sickle Cell Disease and Hemophilia
A study on 881 patients with sickle cell disease showed that there is an increased risk of aseptic loosening at 2 years compared to the control group [56]. Hemophilic patients have a 10.2% risk for aseptic loosening at a mean follow-up of 11.5 years [57]. Parkinson's disease also increases the risk of aseptic loosening (6.1%) in the first year postoperatively [58]. Patients with chronic liver disease (CLD) have a 7% risk for aseptic loosening at a mean follow-up of 13.5 months, which is higher compared to the control group [59]. On the other hand, hemodialysis almost doubles the rate of aseptic loosening in total hip replacements, according to 25 studies involving 755 hips in 534 patients [60].

Spinal Fusion
Regarding previous surgeries, the presence of lumbar spinal fusion increases the risk for aseptic loosening (OR = 3.42 at 90 days and OR = 4.68 at 1 year) [61]. Another factor that can increase the risk for aseptic loosening is a previous hip arthroscopy (OR = 2.18) [62].

Smoking
It increases the risk for aseptic loosening by three times, as Lee et al. pointed out [24]. Bhaveen et al. showed that the overall implant survival of primary total hip arthroplasty at a mean follow-up of 51 months is 99% in the non-smoker group and 92% in the smoker group [63]. In addition, smoking cessation within two months before surgery does not add any benefits to the outcome of a THA according to a meta-analysis by Myers et al. [63]. Nevertheless, there was one randomized trial on 120 patients which showed that smoking cessation 6-8 weeks before surgery reduces the postoperative morbidity including woundrelated complications and the need of a revision surgery [63,64].

Postoperative Protocol
Early full weight-bearing has not been shown to affect the bone mineral density around the implant at 5 years postoperatively, as compared to partial weight-bearing [41]. Nevertheless, high-impact activities have an odds ratio of 4.24 to develop aseptic loosening, compared to a patient with medium or low activity levels [65].

Antiosteoporotic Drugs
Denosumab is a monoclonal antibody recently introduced in the treatment of osteoporosis. It protects against periprosthetic bone loss after THAs, but its effect diminishes after the administration is stopped [66]. Bisphosphonates decrease the rate of THA revision (2.8% vs. 5.3%), in a study performed on 56,043 patients in Korea [67]. Another study on 95,392 patients showed that oral bisphosphonates reduce the risk of revision surgery by 59% when the treatment is started after the primary arthroplasty surgery [68]. Muratore et al. showed in a study on 35 female patients a statistically significant higher BMD around the femoral implant when treated with ibandronate 3 mg iv. within five days after surgery, followed by oral ibandronate 150 mg/months, plus calcium carbonate (1 g) and vitamin D (880 IU) supplementation [69]. A meta-analysis performed by Zhao et al. showed that bisphosphonates decrease both early and late periprosthetic bone resorption in the main load-bearing zones [70].

Opioids
Preoperative use of opioids is significantly associated with an increased rate of aseptic loosening (OR = 1.3) [71]. Although the mechanism is not clearly understood, an explanation could be provided by the increase of the proinflammatory cytokines by opioids [71].

Statins
As shown by the Danish Hip Arthroplasty Registry, statins reduce the risk of revision arthroplasty [72]. The mechanism can be explained by reduced cytokine production by the administration of statins. Valles et al. showed that statins down-regulate the HMG-CoA/GGPP/RhoA/ROCK pathway, therefore reducing the proinflammatory state and decreasing the osteolysis around the implant [73]. Furthermore, a study on 735 primary THAs showed that statins reduce femoral osteolysis 5 years after the surgical procedure [74].

Beta-Blockers
These are associated with a lower risk of aseptic loosening in THAs (OR = 0.112) [75]. The effect can be due to the interaction between the adrenergic system and bone metabolism [74,75]. Previous studies have shown that beta-blockers decrease the risk of hip fracture and increase bone mineral density (BMD) [75]. All these effects are mostly due to inhibition effect of the the beta-blockers on osteoclastic activity [75].
Moreover, the inhibition of ferroptosis, a process generation aseptic loosening, with a specific inhibitor (Ferrostatin-1) reduces particle-induced ferroptosis in vitro, while overexpression of Nrf2 by siKeap1 or Nrf2 activator Oltipraz upregulates antioxidant response elements and suppressed ferroptosis in osteoblasts [85].

Experimental Studies on the Process of Osseointegration
When we compared the current data with a previous work of our team regarding the prevention of aseptic loosening, we found out that researchers and clinicians have added new recommendations in the last years, while the previous remained generally unchanged. A summary of the current recommendations is available in Table 1. Although the studies which will be further discussed in this chapter are not specifically performed on aseptic loosening in the case of total hip arthroplasty, they are focused on the osseointegration process or survival rates, which are closely related to a decrease in aseptic loosening.
As previous studies showed, the recommended pore size for a better bone ingrowth is 600 µm, when compared to 300 µm and 900 µm [87]. The impact of osteoporosis on the total hip arthroplasty aseptic loosening has not been decided, but studies show a delayed osseointegration process due to a decreased bone metabolism and estrogen deficiency [7]. Other comorbidities to impact hip replacement survival are cardiovascular disease and psychotic disorders [7].
Surgeon's experience is important in decreasing the revision rate of total hip replacements according to Fender et al. [83]. They showed that surgeons who perform 30 THAs per year are more likely to have a higher revision rate compared to those with over 60 THAs per year [88].
In terms of surgical technique, the main purposes are ensuring good stability and good coverage of the implant, while avoiding any excessive reaming and rasping [7]. Moreover, the use of worn reamers and rasps can alter the bone surrounding the implant, thus reducing the osseointegration process [7,89].
Although non-steroidal anti-inflammatory drugs (NSAIDs) have not been related to an increased rate of aseptic loosening, they have been linked to a decrease in the osseointegration process [7]. As a result, until more data regarding its safety are available, we recommend against the use of NSAIDs in the postoperative period of THAs. Other drugs in similar situations are selective serotonin reuptake inhibitors (SSRIs) and loop diuretics, which are recommended to be replaced when possible [7].

Conclusions
Aseptic loosening is a complication that generates a lot of interest in the research community due to its importance. There are currently various methods by which the aseptic loosening rate can be reduced, such as implant characteristics, surgical technique, as well as patient-related and postoperative factors. As there has been a lot of research in the field of osseointegration and aseptic loosening, we consider that an update is important for the clinician to be able to use all of the available techniques to prevent this serious complication. Moreover, the data in this review paper could help researchers find new ways of approaching the treatment of aseptic loosening.
A summary of the generally accepted recommendations include the use of trabecular metal acetabular cups, titanium alloy femoral stems, certain types of femoral stem design (types 1,2,4 and 7), and ceramic-on-polyethylene bearing surfaces.
In addition, surgeons performing more than 100 THAs per year and are consistent with their implant type tend to have a lower revision rate. Surgical factors which limit aseptic loosening include lateral and posterior approaches, good implant stability, at least 60% of acetabular cup bone coverage and an inclination of around 45 degrees. Postoperatively, immediate full weight bearing is accepted.
It is also important to inform on the higher risk of revision the patients with obesity, less than 70 years old, hyperglycemia, rheumatoid arthritis, avascular necrosis of the femoral head, hip dysplasia, sickle cell disease, Parkinson disease, chronic liver disease, lumbar spinal fusion, some genetic variations and a history of hip arthroscopy.