, , , , , , ,

Australian Orthopaedic Association National Joint Replacement Registry – Annual Report 2010

This year’s report confirms that hip and knee replacement in Australia continues to increase. In 2009, the number of hip and knee replacements reported to the Registry increased by 3.2% compared to 2008 (3.4% for hips and 3.0% for knees). Most procedures were undertaken in the private sector (58.6% for hips and 68.8% for knees in 2009).

The changing use of different classes of partial hip replacement previously reported has continued in 2009. The use of unipolar monoblock prostheses continues to decline, particularly the Austin Moore type prosthesis. Since 2003, use of this prosthesis has decreased by 52.2%, and when used the patients selected are generally older. The proportion of patients aged 85 years or older receiving an Austin Moore has increased from 49.0% in 2004 to 57.2% in 2009. The use of bipolar prostheses has also decreased, reducing by 43.2% since 2003. The use of unipolar modular prostheses has increased by 318.2% during the same period.

Fractured neck of femur is the most common reason for performing primary partial hip replacement. The Registry has previously identified that revision rates of primary partial hip replacement for fractured neck of femur are affected by a number of factors. These include the age at time of surgery, class of prosthesis, method of fixation and the type of prosthesis used.

Unipolar modular and bipolar replacement continue to have a lower cumulative percent revision compared to unipolar monoblock prostheses. The ETS, a recently introduced monoblock prosthesis, has the same outcome as cemented Thompson prostheses.

Bipolar prostheses are revised less frequently than unipolar modular prostheses when individuals are less than 75 years of age. The use of cement fixation reduces the risk of revision by approximately half regardless of the class of partial hip replacement.

The two main classes of primary total hip replacement are total resurfacing and total conventional hip replacement.

Primary total conventional hip is undertaken more often than primary total resurfacing hip replacement (95.0% and 4.9% respectively of all primary total hip replacement in 2009). The use of primary total conventional hip replacement continues to increase, not only in absolute numbers but also as a proportion of all primary total hip replacement (91.6% in 2003).

There are many factors known to influence the outcome of total hip replacement. Previously the Registry has reported the effect of different patient and prosthesis characteristics on the risk of revision. As well as updating this analysis, the Registry is reporting on the use of prostheses with exchangeable femoral necks and a more detailed analysis of bearing surface.

Femoral stems with exchangeable femoral necks have twice the risk of revision compared to all other femoral stems. This is due to increased rates of dislocation and loosening. All but three femoral stems with exchangeable necks have an increased risk of revision. The three exceptions have a short follow up period.

It is now clear that primary total conventional hip replacement using metal on metal bearing surface and head sizes over 28mm have a higher risk of revision
compared to all other bearing surfaces. The impact of head size is more apparent in head sizes greater than 32mm.

The increased risk of revision of metal on metal bearing surface is due to higher rates of loosening and metal sensitivity. It is not age related. There is however an interaction between age and head size. The risk of revision for head sizes larger than 32mm is higher regardless of age and this risk is greater the younger the patient.

There is gender variation in outcome, with females having a higher risk of revision when metal on metal bearing surfaces are used. This gender difference is only evident when head size is greater than 32mm.

The higher risk of revision with metal on metal bearing surfaces is not isolated to a small number of prostheses. Of those with head sizes greater than 32mm and over 200 procedures recorded, almost all have a cumulative percent revision that is higher than the entire group of primary total conventional hip replacement.

Initial outcomes for modified polyethylene show a lower risk of revision compared to non‐modified polyethylene in the first nine years. This lower risk is becoming more evident with time. There is no difference in outcome related to head size when using modified polyethylene with head sizes greater than 32mm having the same outcome as those 32mm or less.

New prostheses have continued to come onto the market in 2009. The number of new femoral and acetabular prostheses combinations used in primary total conventional hip replacement increased, with a further 154 combinations recorded.

The use of primary total resurfacing hip replacement has declined for the fourth consecutive year. There was a 17.6% reduction in primary total resurfacing procedures compared to 2008. Analysis on a variety of factors affecting outcome have been presented. These include primary diagnosis, gender, age, head
size and type of prosthesis. Patients having a total resurfacing for osteoarthritis are revised less frequently than patients with developmental dysplasia. Females have a significantly higher risk of revision compared to males. Males have an age related risk of revision, which is significantly higher from 65 years or older.

As reported last year, the difference in outcome related to gender is largely due to the size of the femoral component. There is an inverse relationship between risk of revision and size of the femoral head component.

The five classes of primary partial knee replacement are partial resurfacing, unispacer, bicompartmental, patella/trochlear and unicompartmental knee replacement. Two of these (partial resurfacing and bicompartmental) are relatively recent technologies introduced to the Australian market and reported for the first time two years ago. Both of these single product procedures continue to have higher rates of revision compared to other knee replacement procedures with the exception of the unispacer.

Patella/trochlear procedures continue to be undertaken in small numbers (226 reported in 2009). The cumulative percent revision at nine years for patella/trochlear replacement is 25.1%. Age and gender are significant risk factors.

The use of unicompartmental knee replacement continues to decline. There were 26.2% less unicompartmental knee replacements undertaken in 2009 compared to 2003. Age at the time of surgery is a major factor affecting the outcome, the younger the patient the greater the risk of revision.

Primary total knee replacements account for 80.0% of all knee replacements and has increased in use by 55.9% since 2003.

Patient and prostheses characteristics are important factors affecting the outcome of primary total knee replacement. Primary diagnosis, age and gender as well as prosthesis stability, bearing mobility, patella resurfacing and the type of prosthesis used all impact on the risk of revision.

Rheumatoid arthritis has the lowest risk of revision of any primary diagnosis. The risk of revision increases with decreasing age. After three and a half years, those aged less than 55 years have over four and a half times the risk of revision compared to those aged 75 or older. Males have a higher incidence of revision. This is in part due to a higher rate of reported infection.

Tibial prostheses with fixed bearings have a lower risk of revision compared to mobile bearing tibial prostheses. The Registry identifies three types of fixed bearings with all‐polyethylene tibial prostheses having the highest risk of revision of the fixed bearing tibial prostheses.

Posterior stabilised primary total knees are revised more than minimally stabilised knees. The risk of revision in the first nine years is increased if the patella is not resurfaced and this risk is highest if a posterior stabilised prosthesis is used.

The Registry has had sufficient data to present preliminary information on the outcome of revision hip and knee replacement for a number of years. Although this analysis is complex, it has become increasingly evident that regardless of the class of primary procedure revised and the type of revision, there is a high risk of subsequent re‐revision.

The risk of re‐revision of a revised primary total resurfacing hip is similar to the risk of re‐revision of a revised primary total conventional hip replacement. The risk of re‐revision of a primary unicompartmental knee revised to a total knee is the same as the rerevision risk of a revised primary total knee replacement. For primary total knee replacement insert only revision has the highest risk of re‐revision compared to all other types of revision. Patella resurfacing undertaken after a primary total knee has over three times the cumulative percent revision at seven years compared to a primary total knee replacement.

A unique and important function of registries is the ability to provide population based data on the comparative outcome of individual prostheses in a community. In this report, the Registry has presented the outcome of individual prostheses within each class of prosthesis. There is significant variation depending on the type of prosthesis used.

The Registry specifically highlights prostheses or prostheses combinations identified as having a higher than anticipated rate of revision. These have been reported in the section on ‘Prostheses with Higher than Anticipated Rates of Revision’.

This year the Registry has identified 65 prostheses or prostheses combinations. For the first time detailed analysis for each of these prostheses and prostheses combinations is available as a supplementary report on the Registry website.

Full Report – To Read on-line or Download