From California Technology Assessment Forum

Metal on Metal Hip Resurfacing as an Alternative to Total Hip Arthroplasty: A Technology Assessment

Leah Karliner, MD, MAS

Authors and Disclosures

Introduction and Background

Introduction

The California Technology Assessment Forum (CTAF) is requested to review the scientific evidence for metal on metal (MoM) hip resurfacing as an alternative to total hip arthroplasty. This is an update to the CTAF assessment done on this same topic in 2007.^[1]

Background

Disease affecting the hip joint is usually caused by osteoarthritis (OA), the most common form of joint disease.^[2] OA is chiefly a disease of aging; 90% of all people have radiographic features of OA in weight-bearing joints by age 40.^[3] It is characterized by changes to the structure of the entire joint, particularly degeneration of cartilage and hypertrophy of bone at the articular margins.^[4] The presenting symptom of osteoarthritis of the hip is generally pain, which may be associated with a limited range of motion; though pain in the hip may be referred from other regions of the body, referred to other structures (such as the knee) or may be confused with other etiologies such as trochanteric bursitis.

Hereditary and mechanical factors may be involved in the pathogenesis of OA. Obesity is a risk factor for knee osteoarthritis and probably for the hip. Participation in competitive contact sports increases risk as do jobs requiring frequent bending and carrying; for example, farming carries a significantly increased relative risk for OA.^[5]

OA of the hip joint contributes to morbidity for the individual and costs to society. Overall, OA is the sixth leading contributor worldwide to total years lost to disability, or disability adjusted life years (DALYs).^[5] Individuals with hip OA may suffer from pain, stiffness and loss of function, adversely impacting their health related quality of life. The direct and indirect societal costs attributable to OA are enormous. For example, Individuals with OA are more likely to reduce work hours or take early retirement. Older adults with symptomatic arthritis report greater medical utilization and health care costs compared with people not reporting arthritis.^[6]

Rheumatoid arthritis (RA), an inflammatory arthropathy, may also lead to degeneration of the hip joint, but because it is a systemic condition is unlikely to affect the hip joint alone. Involvement of the hip joint in RA occurs in ten percent to 40% of individuals.^[2] Other conditions that can cause secondary OA are avascular necrosis, congenital dislocation, Paget’s disease, ankylosing spondylitis and traumatic arthritis.

Treatment for degenerative disease of the hip includes pharmacological and non-pharmacological measures, including lifestyle interventions. Analgesics such as acetaminophen and narcotics can treat the pain associated with the disease and improve function; non-steroidal anti-inflammatory medications, such as ibuprofen, can also be used and are more effective in more advanced disease but are less safe. Corticosteroid injections are a mainstay of treatment for OA and RA, though injections into the hip joint often need to be done under radiographic guidance.^[7] Lifestyle interventions used to prevent or ameliorate the progression of OA include weight loss, exercise and physical therapy. Surgical interventions include arthroscopy, MoM hip resurfacing arthroplasty (HRA) and total hip arthroplasty (THA). More than 168,000 total hip arthroplasties are performed annually in the United States.^[8]

In the 1950’s the concept of resurfacing the arthritic socket, as well as the femoral head, emerged, but the material used (first Teflon and then methacrylate cement) and design flaws led to numerous device failures and adverse events, such as avascular necrosis.^[9] The third generation of hip resurfacing emerged in England in the 1990s, and while there are at least ten different commercially available hip resurfacing systems, to date only the Birmingham Hip Resurfacing (BHR), the Cormet 2000, and the Conserve Plus systems are FDA approved. All three of these systems use a MoM Cobalt-Chromium articulation, but they use different combinations of materials for the backing of the acetabular component. While there are differences in materials and design, all hip resurfacing devices consist of two parts: a cup shaped acetabular component and a cemented femoral cap with a stem that inserts into the femur.

The potential advantage of MoM hip resurfacing over THA is that it allows for most of the femoral head to be preserved and only replaces the surface of the joint; this maintains the femoral canal and can make revision surgery, if necessary, less complex.^[10] Thus, MoM hip resurfacing has been promoted for younger patients with end-stage OA of the hip or RA, traumatic arthritis, hip dysplasia or avascular necrosis for whom conventional THA is not expected to last their lifetime. Other potential advantages of hip resurfacing over THA include earlier return to function and less restriction on function compared to THA. However, one expert points out that the posterior approach favored in resurfacing devascularizes the femoral head, possibly permanently, potentially leading to avascular necrosis over time.^[11] Hip resurfacing is considered by most to be a more challenging operation for the surgeon than THA and requires specialized training and a significant learning curve.^[12-14]

The 2007 CTAF assessment concluded that the peer-reviewed literature had not kept pace with changes in hip resurfacing technology; most of the published literature consisted of case series from single surgeons, and there were no randomized controlled trials of FDA approved hip resurfacing devices.^[1] We undertake the current assessment to review the peer-reviewed literature of the three FDA approved MoM hip resurfacing systems published since the 2007 assessment, with a particular focus on randomized controlled trials (RCT) and comparative study (level 1-4) evidence.

Technology Assessment (TA)

TA Criterion 1: The technology must have final approval from the appropriate government regulatory bodies.

The Birmingham Hip Resurfacing (BHR) system (Smith & Nephew Inc., Memphis, TN, USA) received FDA PMA approval in 2006.

The Cormet 2000 Hip Resurfacing System (Corin USA, Tampa, FL, USA) received FDA PMA approval in July 2007.

Wright Medical Technologies (Arlington, TX, USA) CONSERVE Plus Hip System received FDA PMA approval in November 2009.

There are several other hip resurfacing devices which have been developed but which have not received FDA clearance at this time.

TA Criterion 1 is met.

TA Criterion 2: The scientific evidence must permit conclusions concerning the effectiveness of the technology regarding health outcomes.

The Medline, Embase, and Cochrane clinical trials database, Cochrane reviews database and the Database of Abstracts of Reviews of Effects (DARE) were searched for relevant references through April 2010. While the majority of the search results were published case studies, there were four RCTs, two since the 2007 CTAF assessment and none of which studied an FDA approved device.^[15-18] One of these RCTs resulted in multiple publications.^[18-22] There was a final publication which presented the study protocol for a planned RCT comparing MoM hip resurfacing to THA, but didn’t state which MoM would be studied.^[23] In addition, there were nine publications reporting on six non-randomized comparison studies.^[24-32]

Level of Evidence: 3,4,5

TA Criterion 2 is met.

TA Criterion 3: The technology must improve net health outcomes.

Randomized Controlled Trials

As noted above, there are three published RCT’s of MoM hip resurfacing compared with total hip arthroplasty, none of which studied an FDA approved device. As noted in the 2007 CTAF assessment, the first, by Howie et al, was stopped early due to early failures in the MoM hip resurfacing group.^[17] The second, by Venditolli et al, found no difference in satisfaction or complication rates, but the MoM hip resurfacing group had better postoperative functional performance.^[18] The third, by Garbuz et al, compared a non-FDA approved device to large diameter MoM total hip arthroplasty and while they found no difference in quality of life (QoL) outcomes, they noted increased serum cobalt and chromium ion levels in both groups, but highest in MoM large head total hip arthroplasty, demonstrating that circulating metal ions are a concern for all MoM hip devices.^[16] The final trial, by Lavigne et al, compared a non-FDA approved device to large diameter total hip arthroplasty and found no difference in gait speed, postural balance, and clinical scores; both groups achieved similar function to a healthy control group by three months.^[15]

Non-Randomized Comparative Studies

Again, as noted above, our search turned up nine publications reporting on six non-randomized comparison studies. ( Table 1 .) All of these studies had small sample sizes and for many of them their comparator groups were either recruited selectively or were historical controls. Only one study – Brennan et al – reported on the BHR and they compared the weight of acetabular reamings for BHR patients to differentially selected un-cemented THA patients and found the weights to be similar; this is at best a surrogate marker for potential adverse outcome of a femoral neck fracture.^[24]

Of the six publications reporting on the Conserve Plus hip resurfacing system, three reported slightly different comparisons and/or outcomes for overlapping patient populations,^{[27, 28, 32]} and a fourth may well also be from an overlapping population given that the authors are also overlapping.^[29] These were retrospective analyses comparing MoM hip resurfacing patients to historical control patients with ‘conventional’ THA from a separate registry; taken together they found no difference in Harris Hip Score – a measure of pain, function, satisfaction, or range of motion. There was a difference in activity level; however, activity level was different in the two groups at baseline and no adjustments were made for this, possibly because of the small sample sizes. A fifth study had similar findings and similar limitations.^[25] The sixth study of Conserve Plus was a retrospective study in which patients acted as their own controls – they assessed range of motion in those patients in their registry who had MoM hip resurfacing on one side and THA on the other, some of which were revisions from prior resurfacing procedures.^[26] These authors found no difference by type of replacement, including no difference in the subset of patients with revisions and no difference when stratified by size of implant. Of note, the study was underpowered for these subanalyses.

Two papers reported on the Cormet 2000 hip resurfacing system in the same patient population.^{[30, 31]} The earlier report concludes that the resurfacing patients had equal overall clinical success with the ceramic-on-ceramic THA patients in terms of a composite score including Harris Hip Score, ‘radiographic evidence of success’, absence of device related complications and absence of revision; however, the MoM hip resurfacing group had substantially more revisions (7.1% vs. 1.9%) in a shorter follow-up time frame.^[31] Revisions were primarily due to femoral neck fractures and femoral component loosening. The THA group was the historical control. The later report assessed the Harris Hip Score at specific time-points over two-year follow-up in the same two groups of patients, concluding that while there were differences early on, with the THA group doing better at six weeks and the MoM hip resurfacing group doing better at six months, there was no difference at 12 or 24 months with >90% of both groups scoring in the excellent range.^[30]

Studies Focused on Adverse Outcomes or Side Effects

Our search found eight publications since the 2007 CTAF assessment which specifically focus on adverse events associated with FDA approved hip resurfacing device systems, four reporting on early failure^[33-36] and four reporting on blood metal ion levels.^[37-40] Of the four studies focused on early failure and revision rates, three were case series and one was a meta-analysis.^[36] The case series with the longest follow-up period (mean five years) included patients with varying MoM hip resurfacing devices in place, but most were BHR, and reported a 5.2% failure rate with pain being the predominant reason for revision and loosening of the femoral component the next most common cause.^[33] The other study of BHR had a much lower revision rate (2.8%) with a shorter mean follow-up time of 3.5 years, and pain was also the predominant cause, followed by femoral neck fracture.^[35] At time of revision, these authors found evidence of metallosis (immune reaction to metal) and necrosis. The meta-analysis included multiple devices for both MoM hip resurfacing and the comparator THA patients, although all of the THA had cementless femoral components, and all of the included studies had a young patient population (mean age <55).^[36] These authors found a higher overall failure rate for THA when failure was defined as all reasons, including after revision and asymptomatic radiographic failure. The THA patients had approximately double the follow-up time as the hip resurfacing patients (8.5 years vs. 3.9 years). When the authors defined failure more narrowly to focus on the greatest risk with MoM hip resurfacing as ‘femoral failure due to mechanical failure’, the hip resurfacing group had about double the rate of failure.

There has been increasing concern in the literature about blood metal ion levels as a result of wear from metal on metal hip replacements, in both hip resurfacing and THA. While there is no safe or toxic level of chromium and cobalt in the blood, the concern about increased levels ranges from local tissue toxicity, to chromosomal damage and malignant cellular transformation.^[41-44] The four publications that focused on this topic all studied patients with BHR. One found that both hip resurfacing and metal-on-metal THA patients with small diameter (28mm) heads had similar chromium and cobalt levels, both higher than healthy controls without metal implants.^[40]Another study compared unilateral MoM BHR, bilateral MoM BHR with metal-on-polyethylene THA and ceramic-on-ceramic THA.^[37] They report the highest chromium and cobalt ion levels with both types of MoM hip replacement, smaller elevation for the metal-on-polyethylene group, and negligible levels for the ceramic-on-ceramic group. A third study again reported increased ion levels for patients with MoM hip resurfacing implants; however, they found that as the femoral component size increased, ion levels decreased, implying that with less friction and wear there is less release of ions.^[39] The fourth study found enhanced lymphocyte activity for nickel, but not for chromium and cobalt; there was no difference for patients with and without pseudotumors (a cystic or solid mass relating to resurfaced hip), although the pseudotumor group was very small (n=10).^[38]

Table 2.

TA Criterion 3 is met.

TA Criterion 4: The technology must be as beneficial as any established alternatives.

The primary established alternative is total hip arthroplasty. While there are no RCTs of the FDA approved hip resurfacing devices comparing to current conventional THA devices, the comparison studies cited above all report results showing equivalent functional results after MoM hip resurfacing and THA. While there is concern about early failure requiring revision after MoM hip resurfacing, THA presents other concerns particularly for younger, more active patients who have risk of shattering their femur in the event of trauma and who will undoubtedly require repeat THA with less and less available bone over time. Concern about metal ions exists for both resurfacing and THA (metal-on-metal).

TA Criterion 4 is met.

TA Criterion 5: The improvement must be attainable outside of the investigational setting.

MoM hip resurfacing has been used broadly outside of investigational settings. Each device requires training and given a considerable learning curve – and adverse events including early failure associated with inexperienced surgeons – hip resurfacing should only be performed after appropriate training and early cases should be supervised by experienced surgeons.

Conclusion

While there are more published studies comparing hip resurfacing with THA than at the time of the CTAF 2007 assessment, all of which suggest equivalent results for resurfacing and THA, some of the same questions that were of concern at that time remain and we reiterate them here: What is the long term durability of the resurfaced hip compared with THA? What will be the short and long term results when this generation of younger patients who have undergone hip resurfacing are eventually converted to THA? Will there be unforeseen long term complications that will make this revision more problematic than anticipated? What are the long term health consequences of increased low levels of circulating metal ions produced by MoM hip resurfacing? It is incumbent upon the hip resurfacing community to continue to investigate these questions and assure that metal-on-metal hip resurfacing continues to be a viable and safe option for young active patients with degenerative hip disease.

Despite these remaining questions, after considerable discussion and expert testimony regarding the existing data on both benefits and harms, the forum supported approval of metal-on-metal hip resurfacing as an alternative to total hip arthroplasty.

Recommendation

It is recommended that metal on metal hip resurfacing by surgeons who have completed training using FDA-approved devices meets CTAF criteria 1 through 5 as an intervention for degenerative hip disease in appropriate patients younger or <65 who have an active life expectancy of more than 10 to 15 years.

June 2, 2010

This is the second assessment of this technology.

The CTAF panel voted in favor of the recommendation as noted above.

Recommendations of Others

Blue Cross Blue Shield Association (BCBSA)

A June 2007 assessment by the BCBSA Technology Evaluation Center found that “use of an FDA-approved metal-on-metal total hip resurfacing device as an alternative to THA in patients who are likely to outlive the 10 years or more functional lifespan of a traditional MoM prosthesis meets the TEC criteria.”

Centers for Medicare and Medicaid Services (CMS)

Neither a National Coverage Decision nor a Local Coverage Decision was found through a search of the CMS Coverage database.

California Orthopaedic Association (COA)

The COA was invited to have a representative attend the meeting and to provide an opinion on this technology.

Alliance for Orthopedic Solutions

The Alliance for Orthopedic Solutions provided an opinion and attended the meeting to provide testimony.

National Institute for Health and Clinical Excellence (NICE)

In 2002 NICE issued a guidance document which states (in part):
"MoM hip resurfacing is recommended as an option for people with advanced hip disease who would otherwise receive a conventional primary total hip replacement (THR) and are likely to live longer than the device is likely to last." This document is available at: http://guidance.nice.org.uk/TA44.

Canadian Coordinating Office for Health Technology Assessment (CCOHTA)

In March 2005 the CCOHTA published a document on Minimally Invasive Hip Resurfacing. This is the most current document as of March 2010. The CCOHTA notes: “Outcome-based research and long-term follow-up are necessary to assess the clinical and economic impact of a minimally invasive approach to hip resurfacing. There is also a need for defined criteria to determine which patients might benefit from this surgical approach”.

From BMC Musculoskeletal Disorders

Minimally Invasive and Computer-navigated Total Hip Arthroplasty: A Qualitative and Systematic Review of the Literature

Inge HF Reininga; Wiebren Zijlstra; Robert Wagenmakers; Alexander L Boerboom; Bregtje P Huijbers; Johan W Groothoff; Sjoerd K Bulstra; Martin Stevens

Authors and Disclosures

Abstract and Introduction

Abstract

Background: Both minimally invasive surgery (MIS) and computer-assisted surgery (CAS) for total hip arthroplasty (THA) have gained popularity in recent years. We conducted a qualitative and systematic review to assess the effectiveness of MIS, CAS and computer-assisted MIS for THA.
Methods: An extensive computerised literature search of PubMed, Medline, Embase and OVIDSP was conducted. Both randomised clinical trials and controlled clinical trials on the effectiveness of MIS, CAS and computer-assisted MIS for THA were included. Methodological quality was independently assessed by two reviewers. Effect estimates were calculated and a best-evidence synthesis was performed.
Results: Four high-quality and 14 medium-quality studies with MIS THA as study contrast, and three high-quality and four medium-quality studies with CAS THA as study contrast were included. No studies with computer-assisted MIS for THA as study contrast were identified. Strong evidence was found for a decrease in operative time and intraoperative blood loss for MIS THA, with no difference in complication rates and risk for acetabular outliers. Strong evidence exists that there is no difference in physical functioning, measured either by questionnaires or by gait analysis. Moderate evidence was found for a shorter length of hospital stay after MIS THA. Conflicting evidence was found for a positive effect of MIS THA on pain in the early postoperative period, but that effect diminished after three months postoperatively. Strong evidence was found for an increase in operative time for CAS THA, and limited evidence was found for a decrease in intraoperative blood loss. Furthermore, strong evidence was found for no difference in complication rates, as well as for a significantly lower risk for acetabular outliers.
Conclusions: The results indicate that MIS THA is a safe surgical procedure, without increases in operative time, blood loss, operative complication rates and component malposition rates. However, the beneficial effect of MIS THA on functional recovery has to be proven. The results also indicate that CAS THA, though resulting in an increase in operative time, may have a positive effect on operative blood loss and operative complication rates. More importantly, the use of CAS results in better positioning of acetabular component of the prosthesis.

Background

Total hip arthroplasty (THA) is considered to be one of the most successful orthopaedic interventions of the past 40 years, with 10-year survival rates exceeding 90%.^[1,2] In recent decades there has been considerable effort to improve the component designs and modes of fixation of total hip prostheses.^[3] The concept of minimally invasive surgery (MIS) was adopted recently in the orthopaedic society, leading to the development of minimally invasive techniques for THA. Computer-assisted surgery (CAS) has also gained popularity, since it has the potential to improve the accuracy of orthopaedic procedures.

Despite the increase in use of MIS THA, its risks and benefits are still an ongoing debate issue in the orthopaedic society. Proponents of MIS THA claim that it results in less soft-tissue trauma (smaller skin incision and less muscle damage), reduced blood loss and fewer blood transfusion requirements. Postoperative benefits include less pain, shorter hospital stay, quicker return to function and better cosmetic appearance.^[4,5] Opponents claim that MIS THA introduces additional risks due to limited visibility of anatomical landmarks and vital structures.^[6] Complications involve higher risks for thromboembolism, infection, neurovascular injury, femoral fracture and component malposition, which can result in increased prosthetic wear.^[7,8]

Proper positioning of the hip prosthesis is essential for improving the long-term success of THA. Higher rates of pelvic osteolysis, asymmetric polyethylene wear and component migration have been observed when the acetabular component is malpositioned.^[9] Lewinnek et al.^[10] determined a "safe zone" of 5° to 25° of anteversion and 30° to 50° of abduction. They found that the dislocation rate of hip prostheses, where the acetabular components were placed outside this safe range, was approximately four times higher. Most surgeons aim for this safe zone using mechanical alignment guides provided by the manufacturer of the hip prosthesis. However, these mechanical alignment guides have shown clear limitations in terms of accuracy and precision of proper orientation of the hip prosthesis.^[11]

As a result, the interest in computer navigation systems for orientation of the hip prosthesis is increasing, since it may be the solution for the aforementioned problems related to proper prosthetic positioning. Moreover, CAS is not only aimed at an improved alignment of the hip prosthesis, it also provides instant information and feedback to the surgeon, which may make the surgical technique easier to perform and may result in better clinical outcomes. The imaging systems that are used during CAS can be roughly divided into image-based and imageless systems. Image-based systems require the collection of morphological information by preoperative CT scans or MRI, or by means of intraoperative fluoroscopy. Imageless systems use a virtual anatomical model which is embedded in the software and is supplemented by intraoperative registration data of anatomical landmarks.^[12]

CAS in THA is not very common nowadays, due to the fact that current CAS systems may involve longer operation times and the introduction of new equipment in the operating room. Other factors that limit the broad application of CAS are costs and complexity of computer navigation systems.^[13] Several studies have shown however that inaccuracies in prosthetic placement through conventional THA techniques can be significantly reduced by using computer navigation, thereby reducing the risk of various complications such as dislocations.^[14–16]

The use of CAS may be the solution to the limited visibility of anatomical landmarks during MIS THA.^[17] Some even hypothesize that MIS in combination with CAS will result in better positioning of the prosthesis, compared to conventional THA techniques.^[18] Combining both techniques with claims of quicker recovery and less pain, together with accurate acetabular component positioning and a minimized risk of dislocation, may result in a more effective procedure for THA compared to the conventional technique. However, there is still controversy concerning the most effective technique for THA because of a lack of scientific evidence on the effectiveness of MIS, CAS and computer-assisted MIS for THA. Hence we performed a systematic review of published evidence on the effectiveness of MIS, CAS and computer-assisted MIS for THA.

Materials and Methods

Search Strategy

Following the recommendations of the Cochrane collaborations, an extensive computerised literature search of PubMed, Medline, Embase and OVIDSP was conducted on all studies published between 1995 and May 2009. We used database-appropriate terms, including hip arthroplasty(ies)/replacement(s), minimally invasive/MIS/mini-incision, and/or computer-assisted/navigation/CAS/CAOS. The search strategy was formulated by an experienced medical librarian. To find more studies, the reference lists of all relevant studies were reviewed for potential articles.

Inclusion Criteria and Procedure

A study was included in the review if 1) a randomized controlled trial or a clinical controlled trials was conducted; 2) the study was published in English, Dutch or German; 3) the study was a full-length published article or fully-written published report; 4) the study population comprised patients aged 18 years or older who were undergoing a THA; 5) the study group and control group were similar at baseline with respect to age, gender and BMI; 6) the study contrast was minimally invasive total hip arthroplasty, computer-assisted total hip arthroplasty or a combination of both; and 7) at least one of the following outcome measures was assessed: operative outcome including blood loss and operative time; length of hospital stay; adverse events including intraoperative and postoperative complications; radiographic outcomes including number of outliers of acetabular components outside the desired alignment range; and/or one of the Outcome Measures in Rheumatology Clinical Trials (OMERACT):^[19] pain, self-reported physical function and observed physical function, with a follow-up of at least 6 weeks up to one year postoperatively.

The procedure for inclusion of studies was based on the recommendations described by Van Tulder et al.^[20] The study selection was performed in two stages. The first selection, based on titles and abstracts and taking in consideration the inclusion criteria, was independently performed by two reviewers (IHFR and BPH). The next stage in the inclusion procedure was performed by the same two reviewers, who independently applied the selection criteria as stated above using the full reports. Disagreement was resolved by discussion. If agreement was not achieved at any stage, a third reviewer was consulted (WZ).

Assessment of Methodological Quality

The methodological quality of all articles was independently assessed by two reviewers (IHFR and BPH) using a criteria list.^[20] This list contains 11 criteria related to selection bias, performance bias, attrition bias and detection bias. The requirement of blinding patients or care providers (in this case orthopaedic surgeons) to the intervention (THA) was excluded because such blinding is not possible in this type of research. The question about acceptable compliance in all groups was also excluded, since the question was not applicable to this type of research. All criteria were scored as "yes", "no" or "unclear". Studies were considered to be of methodologically high quality when at least six items scored positively; a score of 3 to 5 was medium quality and a score below 3 was considered low quality. Table 1 shows the used criteria list. Disagreement was resolved by discussion and a third reviewer (WZ) was consulted if disagreement persisted.

Statistical Analysis

Analysis of the extracted data from the included articles was conducted in line with guidelines for systematic reviews from the Cochrane Collaboration Back Review Group.^[20] For continuous variables, the standardised mean difference (SMD) with corresponding 95% confidence intervals (95% CIs) was calculated whenever possible. These effect estimates were interpreted according to Cohen: an SMD of 0.2–0.4 was considered a small effect, 0.5–0.7 moderate and ≥0.8 large.^[21] For dichotomous outcomes such as postoperative complications and acetabular outliers the odds ratio (OR) and 95% CIs were calculated as the summary statistics. This ratio represents the odds of complications or acetabular outliers occurring in the study group compared with the control group. An odds ratio of less than 1 favours the study group and the point estimate of the odds ratio is considered to be statistically significant if the 95% CI does not include the value of 1. Analysis of the included articles was conducted using Review Manager 5 (version 5.0.18, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark).

Efforts to retrieve raw data or means and their standard deviations to compute effect sizes or odds ratios by contacting the authors of articles where these data were not reported, were unsuccessful. We therefore chose to summarise the results by means of a qualitative analysis using a rating system that consists of five levels of scientific evidence, taking into account the methodological quality and the outcome of the original studies (best-evidence synthesis) (Table 2).^[20]

Results

Selection of Studies

Since the search strategy for MIS, CAS and computer-assisted MIS for THA contained similar components, the results of these search strategies overlapped. After removing double citations, 1841 citations remained. A flow chart of the results of the selection procedure after selection based on title, abstract and full text is shown in Figure 1. The main reasons for exclusion of potentially relevant studies based on full-text articles are also presented in Figure 1.

(Enlarge Image)

Figure 1. Flow chart of inclusion procedure. * Multiple reasons for excluding were possible per study. RCT = randomized controlled trial; CCT = controlled clinical trial; MIS THA = minimally invasive total hip arthroplasty; CAS THA = computer-assisted total hip arthroplasty.

Eventually, 25 articles were included. In 18 of these articles the study contrast was minimally invasive THA.^{[4,17,22–37]} A computer navigation system was used during THA during the conventional as well as the MIS approach in two of these studies.^[17,26] Computer-assisted THA was the study contrast in seven articles.^{[15,16,18,38–41]} In two of these studies, a minimally invasive technique for THA was used in the freehand as well as the CAS group.^[18,39] In the study of Kalteis et al.,^[15] acetabular components were implanted either freehand or using a CT-based or an imageless computer navigation system. The results of the comparison of the two navigation systems are reported separately in this review. Najarian et al.^[39] report on the results of the first 49 cases of CAS THA and a second series of 47 cases of CAS THA. Since the first series were used to present data on the learning curve of CAS THA, the results of the second series are reported in this review.

None of the included articles had computer-assisted minimally invasive THA as study contrast. The characteristics of the included studies are presented in Additional file 1.

Methodological Quality

The results of the methodological quality assessment of the included articles are presented in Table 3. Overall, the methodological quality of the studies was found to be medium. Four of the studies with MIS THA as study contrast were of high methodological quality^[4,30,31,34] and 14 of medium methodological quality.^{[17,22–29,32,33,35–37]} Three of the studies with CAS THA as study contrast were of high methodological quality,^[15,16,40] the other four medium methodological quality.^{[18,38,39,41]}

Operative Time

Operative time was reported in 16 studies with MIS THA as study contrast (Table 4). Two high-quality studies and five medium-quality studies reported a moderate to large decrease in operative time for MIS THA.^{[23,27,30,32–34,37]}One other high-quality study and eight medium-quality studies reported no significant difference in operative time.^{[4,17,24–26,28,29,35,36]}

Operative time was reported in four studies with CAS THA as study contrast (Table 4). Except for the sub-study of Kalteis et al.^[15] on an imageless computer navigation system, all studies reported a moderate increase in operative time for THA when using computer navigation.^{[15,38,39,41]}

Intraoperative Blood Loss

Intraoperative blood loss was reported in 14 studies with MIS THA as study contrast (Table 4). Two high-quality studies^[4,34] and eight medium-quality studies^{[23,25–27,29,33,36,37]} reported a small-to-large decrease in intraoperative blood loss after MIS THA. One high-quality study and three medium-quality studies reported no significant difference.^{[24,30,32,35]}

Two studies with CAS THA as study contrast reported on intraoperative blood loss (Table 4). Sugano et al.^[41]reported no significant effect of the use of computer navigation during THA on intraoperative blood loss. However Najarian et al.^[39] reported a significant decrease in intraoperative blood loss.

Length of Stay

Ten studies reported on length of stay after MIS THA (Table 4). Three medium-quality studies reported a moderate-to-large decrease in length of hospital stay after MIS THA.^[23,26,32] Two high-quality studies^[4,34] and five medium-quality studies^{[17,24,25,27,36]} reported no significant differences in length of stay between the MIS THA group and the control group. None of the studies with CAS THA as study contrast reported data on length of stay.

Complications

Seventeen studies with MIS THA as study contrast reported on intraoperative and postoperative complications (Table 5). Two high-quality studies^[4,30] and two medium-quality studies^[35,37] reported higher complication rates after MIS THA, but these rates were statistically non-significant. The results of six medium-quality studies^{[23,25,26,29,32,33]} showed lower, though statistically non-significant, complication rates after MIS THA. Moreover, two high-quality studies^[31,34] (reporting on the same data) and five medium-quality studies^{[17,24,27,28,35]} reported no differences in complication rates between the study and control group.

Seven studies with CAS THA as study contrast reported on intraoperative and postoperative complications (Table 5). Both sub-studies of Kalteis et al.,^[15] which are high-quality studies, reported lower complication rates in the CAS group than in the control group. These results are also shown in a medium-quality study,^[41] yet in all these studies such differences in complication rates were statistically non-significant. One high-quality study^[40] and three medium-quality studies^[18,38,39] reported no significant difference either.

Acetabular Outliers

The number of acetabular components outside the desired alignment range (acetabular outliers) was reported in 13 studies with MIS THA as study contrast (Table 5). The high-quality study of Kim^[30] reported more acetabular outliers in the study group, but these rates were statistically non-significant. Fewer acetabular outliers were reported in one high-quality study^[34] and two medium-quality studies,^[33,36] though this difference was also non-significant. In addition, one high-quality study^[4] and eight medium-quality studies^[17,23–29] reported no differences in acetabular outliers.

All studies with CAS THA as study contrast reported on the number of acetabular outliers (Table 5). Five studies showed significant fewer acetabular outliers for CAS THA.^[15,38–41] The other two studies also reported fewer acetabular outliers for CAS THA, but this difference was statistically non-significant.^[16,18]

Physical Functioning

In order to evaluate physical functioning after THA, several physician-based and self-reported questionnaires are in use. Furthermore, objective assessment of physical function can be done by means of gait analysis. In total, thirteen studies with MIS THA as study contrast reported on physical functioning outcome measures. None of the studies with CAS as study contrast assessed physical functioning of patients after THA.

Physician-reported Physical Functioning Ten studies with MIS THA as study contrast reported on physician-based physical functioning outcome measures (Table 6). In these studies, two different outcome measures were used, namely the Harris Hip Score^{[17,24–27,34,36,37]} and the Merle d'Aubigné Hip Score.^[28,33] Six studies reported six weeks postoperatively follow-up data. One medium-quality study^[27] reported significant improvements in physician-reported physical functioning, and the other five studies (one high-quality and four medium-quality) reported no significant differences.^{[24–26,34,37]} Five studies reported three months postoperatively follow-up data. Three medium-quality studies^[17,28,37] reported significant improvement in physical functioning scores in favour of MIS THA, and two medium-quality studies^[27,36] showed no significant differences. Six medium-quality studies reported six months postoperatively follow-up data. Only one study^[17] reported significant improvement in physical functioning scores six months after MIS THA when compared to conventional THA; the other five studies^{[26–28,33,36]} showed no significant differences. Two medium-quality studies reported follow-up data at one year after THA.^[17,28] Neither study found significant differences in physical function.

Patient-reported Physical Functioning Five studies with MIS THA as study contrast reported on patient-reported physical functioning by means of two disease-specific outcome measures, namely the Western Ontario McMaster University Osteoarthritis Index (WOMAC)^{[29,34,36,37]} the Oxford Hip Score (OHS)^[25,34] (Table 6). Two of these studies also reported on the physical component of the MOS 36-item Short Form Health Survey (SF-36)^[29] and the Short Form-12 (SF-12),^[34] which are both generic questionnaires to assess health-related quality of life. Three studies reported six weeks postoperatively follow-up data. One high-quality study^[34] and one medium-quality study^[25] reported no to small but non-significant improvements on patient-reported physical function. However, one medium-quality study^[37] reported significant effects on the WOMAC in favour of MIS THA. Three medium-quality studies reported follow-up data of three months after MIS THA.^[29,36,37] Two of these studies reported significant effects on the WOMAC in favour of MIS THA^[29,37] and no significant difference on the physical component scale of the SF-12.^[29] Speranza et al.^[36] showed no difference on the WOMAC. One medium-quality study^[36] reported no significant differences on the WOMAC six months after MIS THA. Another medium-quality study^[29] however reported significant differences on the WOMAC one year postoperatively.

Gait Analysis Four studies with MIS THA as study contrast reported gait analysis data to evaluate physical function after THA (Table 6). All four studies reported six weeks postoperatively follow-up data. Two high-quality studies^[31,34] and two medium-quality studies^[22,26] reported no significant effect on gait function. Only one medium-quality study^[26] reported on three months postoperatively follow-up data. They reported no significant effect on gait function three months after MIS THA. Furthermore, none of the studies reported on follow-up data of six months and one year postoperatively.

Pain

Five studies with MIS THA as study contrast reported on pain (Table 7). One study was of high quality,^[30] the other four studies of medium quality.^{[17,27,28,33]} These studies used three different measures to assess pain: a Visual Analogue Scale (VAS)^[27,30] for pain, the subscale of the Merle d'Aubigné Hip score,^[28,33] and the pain subscale of the Harris Hip Score.^[17] Three studies reported six weeks postoperatively follow-up data, reporting a significant moderate decrease^[27] and no significant effect^[28,30] of MIS THA on pain. No significant differences in pain were reported at three months,^[17,27,30] six months^{[17,27,28,33]} and one year^[17,28,30] postoperatively. None of the studies with CAS as study contrast reported on pain after THA.

Best-evidence Synthesis

MIS THA Compared to conventional THA, strong evidence was found for a decrease in operative time and operative blood loss after MIS THA. The evidence for a shorter length of stay was moderate. Strong evidence was also found for no difference in complication rates and position of the acetabular component. Moderate to strong evidence was found for no difference in physical functioning six weeks and six months after surgery. The evidence of a positive effect of MIS THA on physical functioning three months postoperatively was conflicting, as was the evidence for less pain after MIS THA six weeks postoperatively. The evidence for no differences in pain levels three and six months after surgery was strong.

CAS THA Strong evidence was found for a positive effect of CAS THA on the position of the acetabular component. The evidence for a positive effect on operative blood loss was limited. Strong evidence was found for an increase in operative time and for no significant difference in complication rates after CAS THA.

Discussion

We have reviewed the current literature evaluating the effectiveness of MIS, CAS and computer-assisted MIS for THA. The extensive literature search resulted in 18 articles with MIS THA as study contrast, and seven with CAS THA as study contrast, yet no study with computer-assisted MIS for THA as study contrast was discovered. The results of this systematic review indicate that there were no significant differences in operative complications and acetabular component positioning between MIS THA and the conventional procedure. Furthermore, MIS THA resulted in a reduction in blood loss, operative time and reduced length of stay. The added value of MIS THA over the conventional procedure in terms of a faster functional recovery however remains to be proven. Computer-assisted THA results in better positioning of the acetabular component. It may also have a positive effect on operative blood loss and complications despite an increased operative time.

Contrary to what proponents of MIS THA stated, this review showed that MIS THA had no effect on physical functioning, as measured by questionnaires as well as gait analysis. Since the main purported benefit of MIS THA is a decrease in the amount of soft-tissue (muscle) damage, it can be postulated that a difference in improvement of physical functioning and pain will only be seen in the early postoperative period. Only eight studies reported data on physical functioning at six weeks postoperatively.^{[22,24–27,31,34,37]} Six of these studies assessed physical functioning by means of either surgeon-reported or patient-reported questionnaires.^{[24–27,34,37]} Although these are shown to be useful for detecting changes in physical functioning over time in patients with osteoarthritis of the hip and after THA,^[42,43] it is arguable whether these questionnaires are sensitive enough to detect subtle differences in improvement of physical functioning after conventional or MIS THA. A possible solution for this problem is to measure physical functioning objectively by means of quantitative gait analysis. However, only four studies assessed physical functioning using gait analysis.^{[22,26,31,34]} Quantitative gait analysis has been used for numerous applications and has provided insights into functional characteristics not identifiable by clinical exam or other methods. Several studies have compared two surgical techniques for THA, attempting to identify differences in functional outcome.^[44–46] The studies that used gait analysis^[44,46] revealed differences between the surgical approaches, while this result failed to be identified by means of a questionnaire.^[45]

The results for CAS THA demonstrate an increase in operative time and limited evidence for a decrease in operative blood loss, but CAS THA had no effect on operative complication rates. Additionally, the use of CAS during THA had a positive effect on the outliers of the acetabular component position outside the desired range. These results justify use of computer navigation during THA. With improved surgery patients should benefit from having lower morbidity rates, better functional outcome and greater longevity of implants.^[12] Wines and McNicol^[47] showed that during conventional THA it is technically difficult to achieve an accurate alignment of the acetabular component intraoperatively. As judged by postoperative CT scans, surgeons' intraoperative estimates of acetabular component positioning were inside the desired range in less than two-thirds of the cases. Since accurate component positioning benefits the longevity of the implanted prosthesis, CAS can help achieve this goal. However, broader application of computer navigation systems is still hindered by increased operative times, partly due to the complexity of the systems and the accompanying financial costs.

Despite efforts to get an ample overview of the available literature on MIS and CAS for THA, no articles with computer-assisted MIS for THA as study contrast were discovered. Some of the studies included compared computer-assisted MIS for THA with either MIS THA^[18,39] or CAS THA.^[17,26] Their results are in line with the other studies included in this review that compared MIS THA or CAS THA with a conventional approach. Still, an additive effect of the combination of MIS and CAS for THA needs to be established.

Some critical remarks can be made on the included studies. First, a wide variety of surgical approaches was used in them. We chose to analyse all surgical approaches together, since the aim of this systematic review was to assess the effectiveness of minimally invasive THA, but not of any specific minimally invasive THA approach. Second, the surgical approaches were too heterogeneous and often poorly described to perform subgroup analyses. Studies on image-based and imageless navigation systems were also analysed together, since research has shown that imageless navigation is as reliable as image-based navigation for positioning the acetabular component.^[15] Third, the studies included in this review use a variety of definitions of 'minimally invasive THA' or 'mini-incision THA'. The term 'minimally invasive' is clearly open to interpretation. There are patent differences between using an alternate surgical approach intended to gain access to the hip joint through less soft-tissue dissection and using intermuscular planes, and performing the conventional procedure through a smaller skin incision. In the literature, studies use the term 'mini-incision' while, according to the description of the surgical technique, it is a minimally invasive technique which has been used. Conversely, the term 'minimally invasive' is also used in the literature to indicate what appears to be a mini-incision technique. Fourth, the used definitions for the desired range of acetabular component angle varied enormously in the published results of MIS THA and CAS THA. The majority of the studies use the safe zone recommended by Lewinnek et al,^[10] including an abduction angle of 40 ± 10° and an anteversion angle of 15 ± 10°. Some studies reported slightly different operation goals, depending on the surgical approach used. The operation goal was nonetheless always the same in the study group and the control group. Finally, not all studies reported the experience of the surgeons with the specific surgical technique. The introduction of a new surgical technique is often accompanied by a learning curve, associated with a temporary increase of adverse events.^[48] To make an objective comparison between conventional technique and a minimally invasive or computer-assisted technique for THA, it is crucial to exclude the cases that are operated on during the time span of the learning curve for the new surgical technique.

Some limitations of this review and its conclusions need to be addressed. In this systematic review, a highly sensitive comprehensive search was conducted following the recommendations of the Cochrane collaboration in order to identify articles of interest. For practical reasons though, only studies published in English, Dutch or German were included in the final review, which might have led to selection bias. Additionally, in order to get a broad overview of all the literature on MIS, CAS and computer-assisted MIS for THA, we chose to include not only RCTs but also CCTs. Shrier et al.^[49] stated that including studies other than RCTs may provide important additional information, thereby improving inference of the results. Moreover, Poolman et al.^[50] suggested that readers should not assume that studies labelled as Level I are of a high reporting quality, or of a better reporting quality than Level II studies. This was also seen in the present review; some CCTs were of a higher methodological quality than several of the included RCTs. Of the studies included, only six were considered of high quality. None of the studies conducted their analyses following the intention-to-treat principle. Furthermore, several RCTs failed to report on the methods of randomisation and treatment allocation. Since several studies failed to report sufficient data to calculate SMDs, it was not possible to conduct a meta-analysis (quantitative statistics). We therefore used qualitative levels of evidence to summarize the results. Use of a best-evidence synthesis is a next best solution and is a transparent method commonly applied when statistical pooling is not feasible or clinically viable.^[20]

Conclusions

The results of this systematic review indicate that MIS THA is a safe surgical procedure, without increases in operative time, blood loss, operative complications and component positioning when compared to the conventional procedure. However, the surplus value of MIS THA over the conventional procedure in terms of a faster functional recovery remains to be proven. The results of this review also indicate that computer-assisted THA, despite an increased operative time, may have a positive effect on operative blood loss and complications. More importantly, the use of CAS during THA results in better positioning of the acetabular component of the prosthesis. Since minimally invasive THA and the use of computer navigation are becoming increasingly popular in orthopaedics, combining 'the best of both worlds' would be a sensible next step to take. With respect to future research, well-designed studies on MIS THA, CAS THA and especially computer-assisted MIS THA are needed, in which the used definitions, surgical technique, study population, outcome measures and study end-points are adequately described.

From Medscape Orthopaedics & Sports Medicine > Viewpoints

Minimally Invasive Total Knee Arthroplasty Compared With Conventional Surgery

Joseph K. Lee, MD

Authors and Disclosures

Posted: 08/10/2010

A Prospective Randomized Study of Minimally Invasive Total Knee Arthroplasty Compared With Conventional Surgery

Wülker N, Lambermont JP, Sacchetti L, Lazaró JG, Nardi J
J Bone Joint Surg Am. 2010;92:1584-1590

Summary

Minimally invasive total knee arthroplasty has grown in popularity over recent years, but long-term data to support its use over traditional knee replacement surgery are lacking.^[1-4] Wülker and colleagues performed a multicenter study evaluating 134 patients undergoing total knee arthroplasty either by a standardized traditional technique or by a minimally invasive technique on the basis of total range of motion, Knee Society total and function scores, and visual analog scores for pain and activities of daily living.

Follow-up documented at hospital discharge, at 4 to 6 weeks, and then at 1 year after surgery showed no statistically significant difference in any of the outcomes. Furthermore, increased operative times and equivalent estimated blood loss were seen with the minimally invasive group.

Viewpoint

Although this study suggests that no significant benefit was seen in using a minimally invasive surgical technique over a standard traditional technique for total knee arthroplasty, the authors note that these results and their potential implications should be given careful consideration due to several issues.

For instance, the authors noted their study was underpowered because a power analysis was not performed before the study began. They also did not focus on quality-of-life outcomes (eg, length of hospital stay, reliance on pain medications, and the need for inpatient rehabilitation after discharge), which the minimally invasive approach is purported to show an advantage.

Larger, randomized prospective trials that incorporate other relevant metrics may help to delineate the benefits of one type of surgical technique over another. What may be also relevant to see, especially in this changing healthcare environment, is a cost-benefit analysis of a minimally invasive surgical technique compared with a traditional surgical approach.

According to disclosure statements, this study was partially funded by Smith & Nephew, which provided reimbursement only for case report form documentation to the study centers. 

Timing of Anterior Cruciate Ligament Repair: Now, Later, or Never?

Thomas L. Schwenk, MD

Authors and Disclosures

Posted: 09/22/2010; Journal Watch © Massachusetts Medical Society

Abstract and Introduction

Abstract

Functional outcomes were similar for immediate reconstruction and optional later reconstruction.

Introduction

Anterior cruciate ligament (ACL) tears can lead to joint instability and impaired knee function. Standard treatment is reconstruction, usually using patellar or hamstring tendon autograft; 200,000 such operations are performed annually in the U.S., despite no evidence of benefit beyond that of rehabilitation therapy.

In a Swedish trial of immediate versus delayed reconstruction, 121 athletically active adults (mean age, 26) with total acute ACL tears received standardized structured rehabilitation. Participants were randomized to reconstruction within 10 weeks of injury or to delayed reconstruction, with the option of surgery in the presence of continued serious instability and dysfunction.

At 2 years, roughly a third of patients in the delayed group had undergone ACL reconstruction, usually after 6 months of initial rehabilitation. On standardized testing, knee stability was significantly better in the early reconstruction group than in the delayed group; however, no between-group differences were noted in several functional and quality-of-life assessments specific to knee disability. Notably, most patients in both groups also underwent meniscal repair at some point during the study.

Comment

Elite athletes with ACL tears who perform intense cutting and pivoting maneuvers usually will need early ACL reconstruction, but nonelite athletes might do well without reconstruction. An editorialist makes an additional point: "Ultimately, the most important predictor of the long-term outcome may not be instability due to ACL deficiency but rather the presence of articular-cartilage or meniscal lesions, which may lead to premature arthritis."