The Evidence from the Swedish Hip Register
Henrik Malchau, Göran Garellick, Peter Herberts
Summary
The Swedish Total Hip Arthroplasty (THA) Register was initiated in 1979. The primary reason was to document failures and the need for revision surgery to improve and redefine the primary indication, surgical technique and implant choice. The hypothesis is that feedback of data stimulates participating clinics to reflect and improve their health care accordingly. In addition to revision, which has been used as end-point definition to date, patient based outcome measures and radiographic results will be includ- ed in the future to improve sensitivity. The national average 7-year survival (revision as end-point), has improved from 93.5% (±0.15) to 95.8 (±0.15) between the two periods 1979–1991 and 1992–2003.The Swedish results are based on more than 90%, all cemented THA. National implant registers define the epidemiology of primary and revision surgery. In conjunction with individual, subjective, patient data and radiography they contribute to the development of evidence-based THA surgery.
Introduction
The rapid growth of new surgical techniques in con- junction with an accelerating development of new hip implant technology warrants a continuous and objective monitoring of the results paralleled with precise educa- tional efforts. For many years, the purpose of the Swedish National Hip Arthroplasty register was to monitor surgi- cal techniques and prophylactic measures to minimise complications by persisting continuous feed-back to all THA-performing units and to provide a warning system for rapid implant failures. A substantial part of the feed- back system (reporting), all publications, annual reports
and scientific exhibitions, are communicated via www.
jru.orthop.gu.se. All 81 orthopaedic units in Sweden, both public and private, participate voluntarily in the register.
The vast majority of the clinics are reporting data directly via the Internet. Ninety percent of THAs and 75% of re- operations are reported immediately online. There is a short delay in reporting for the remaining units. Copies of complete medical records from all re-operations/revisions are collected for further scientific studies.
The current end point of revision or re-revision is easy to define but leaves many questions regarding the true outcome. The low sensitivity of this end-point has prompted implementation of more sensitive alterna- tives such as individual health outcome (captured by the EQ-5D questionnaire) and a basic radiographic analysis.
These measures are now being implemented in a project that is continuously expanding to involve most parts of Sweden and eventually the entire country. This effort parallels the national health care providers’ (The Swedish Board of Health and Welfare) demand that individual patient outcome should be reported from all national quality registries.
This chapter present an extract of the latest report from the Swedish National Register and preliminary experiences with implementation of patient based out- come measures.
Materials and Methods Primary THA
The register contains information on primary hip ar- throplasties performed in Sweden since 1979. From 1979 until 1991 the number of primary operations and
the type of implant used were recorded from each unit.
The distribution of age, gender and index diagnoses was estimated through a corrective formula based on diag- nosis-specific incidence and prevalence figures given by Statistics Sweden (www.scb.se). From 1992 onwards, data has been collected on an individual basis regarding information on the primary procedure and any subse- quently open procedure by the use of the patient’s social security number (Swedish PNR number). The diagnosis is registered with the ICD-9 and ICD-10 codes. All im- plant parts are registered separately for e.g. cup/liner and stem/head as well as the method of fixation and type of cement. The register’s web application was introduced in January 1999. It uses article bar codes supplied by the manufacturer’s catalogues to ensure correct implant identification and cement brand. The type of incision is also registered per surgical procedure. 77 of the 81 hospitals (96%) report via the Internet and the remain- der within a week after surgery. We know now that the estimations made from 1979 until 1992 were valid [12], and at present 229,031 primary hip arthroplasties have been registered.
End-Points
The current failure end-point in the analyses is revision of either of the components. The revision burden, defined as the fraction of revisions in relation to all primary and revision procedures is used as a crude figure for national and international comparisons. Starting with 2003 annual
report (www jru.orthop.gu.se), all results are presented according to the Kaplan-Meier survival method using the date of death (provided by the Swedish Cause-of-Death Register). Separate survival analyses for cup and stem are presented (example see ⊡Fig. 11.1). In the survival analy- sis for the cup the definition of failure is the exchange of the cup or total revision. The analysis for the stem is performed in a similar way. When a total revision is performed, the register does not display information on which of the components failed. Based on a consensus meeting within the profession in 1996, implant survival for the individual units is public data.
Multivariate Cox´s regression and Poisson regres- sion are used for more complex risk models. However, whatever complex multivariate analyses we undertake, it is important to note that the register’s advantage and drawback is that its results depict the performance of »the average surgeon«.
Other open surgical procedures, apart from revi- sions, constitute only 10% of re-operations. Since 2000 we ceased registration of closed reduction of implant disloca- tion due to a suspected, systematically under-reporting of this procedure.
To increase the sensitivity of the register, patient- related outcome parameters and a radiographic analysis are now included. A standardised follow-up protocol was introduced as a pilot project in 2001 in the Western region in Sweden. All patients completed a question- naire containing 10 items including Charnley’s functional categories (A, B and C) [1], a pain visual analogue scale (VAS) (0–100, none to unbearable) and the EQ-5D [13]
11
⊡ Fig. 11.1. Survival for all diag- noses and all reasons for revi- sion. The survival rate and 95%
confidence limit are indicated for the Charnley stem (red line) and the Ogee all poly, cemented cup (green line). The blue line indicates the survival rate for both com- ponents
100
95
90
85
80
75
70
0 2 4 8 8 10 12
% not revised
Red line = stem revision Green line = cup revision
Blue line = total implantat survival
1992-2003, 12y = 94,2% (93,6-94,9), n = 23 054 1992-2003, 12y = 90,5% (89,6-91,3), n = 23 054
preoperatively and at 1 year, with the intention to repeat the measurement at 6 and 10 years. The functional class will allow for correction of comorbidity. The EQ-5D is a global health index, with a weighted total value for health, with the lowest value –0.594 and a maximum of 1.0. When the index is used for cost-utility analyses, all negative values are set to 0.0. As a complement to the follow-up instrument a satisfaction VAS (0–100, satisfied to dissatisfied) has been added. Data are preoperatively entered via an internet application either by a secretary or by the patient via a touch screen.
At the clinical follow-up (1, 6 and 10 years) the ques- tionnaire is mailed to the patient. Each unit can log in with a password and obtain the EQ-5D index, Charnley category and pain and satisfaction VAS results in real time and compare their results with the remaining users. The Internet site also includes a feature allowing the individual unit to download its own data at any time.
No results from the EQ-5D data will be presented in this chapter.
Results
Primary THA – Patient Demographics
The annual THA incidence in Sweden is approximate- ly 12,500 procedures (7.7–10.2 per 10,000 inhabitants).
More women than men are operated on and the women are at a slightly higher age. Generally, indications for THA have been remarkably stable over the past 20 years although with a different distribution in patients under 50 years of age (⊡Table 11.1).
Primary THA – Implants Used
There has been a marked concentration to fewer well- functioning prosthetic systems for all three fixation prin- ciples over the past 10 years. Among cemented systems, the Lubinus SP II dominates and has increased continu- ously during the last five years, followed by the Exeter and the Spectron prostheses. More than 90% of the total hip replacements performed in Sweden have been all cement- ed reflecting a rather conservative and evidence based at- titude among Swedish orthopaedic surgeons. Four unce- mented prosthetic systems, comprising 580 THAs in 2003 and all with well a documented survival in the medium- term, account for some 80%. Hybrid implants accounted for 512 hips in 2003. The Trilogy cup in combination with Spectron and Lubinus stems are currently dominant. The number of uncemented and hybrid arthroplasties have been increasing during the last two years (⊡Figs. 11.2 to 11.5). The specific age, fixation- and diagnosis-related distribution (1992–2003) are presented in ⊡Tables 11.1 and 11.2.
⊡ Table 11.1. Diagnoses at the index operation 1992–2003. Diagnoses are indicated for age groups: <50 years, 50–59 years, 60–75 years and older than 75 years
Diagnoses
<50 Years 50–59 Years 60–75 Years >75 Years Total
[n] [%] [n] [%] [n] [%] [n] [%] [n] [%]
Primary osteoarthritis
3,129 52.0 13,067 78.2 51,775 80,5 27,584 66.9 95,555 74,5
Fracture 207 3.4 686 4.1 5,106 7,9 8,658 21.0 14,657 11,4
Inflammatory disease 1,058 17.6 1,140 6.8 2,770 4,3 927 2.2 5,895 4,6
Avascular necrosis 374 6.2 456 2.7 1,298 2,0 1,569 3.8 3,697 2,9
Childhood disease 762 12.7 636 3.8 489 0,8 116 0.3 2,003 1,6
Secondary osteoarthritis
95 1.6 110 0.7 469 0,7 619 1.5 1,293 1,0
Tumour 71 1.2 127 0.8 234 0,4 125 0.3 557 0,4
Traumatic osteoarthritis
51 0.8 48 0.3 121 0,2 115 0.3 335 0,3
(Missing) 274 4.6 437 2.6 2,094 3,3 1,542 3.7 4,347 3,4
Total 6,021 100 16,707 100 64,356 100 41,255 100 128,339 100
11
⊡ Table 11.2. Type of fixation at the index operation 1992–2003. Fixation in four categories: cemented, hybrid (uncemented cup/ce- mented stem), uncemented and reverse hybrid is indicated for age groups: <50 years, 50–59 years, 60–75 years and older than 75 years
Type of Fixation
<50 Years 50–59 Years 60–75 Years >75 Years Total
[n] [%] [n] [%] [n] [%] [n] [%] [n] [%]
Cemented 2,807 46.6 11,403 68.3 60,971 94.7 40,679 98.6 115,860 90.3
Hybrid 1,354 22.5 2,812 16.8 2,247 3.5 287 0.7 6,700 5.2
Uncemented 1,532 25.4 1,894 11.3 665 1.0 15 0.0 4,106 3.2
Inverse hybrid 273 4.5 513 3.1 271 0.4 33 0.1 1,090 0.8
(Missing) 55 0.9 85 0.5 202 0.3 241 0.6 583 0.5
Total 6,021 100 16,707 100 64,356 100 41,255 100 128,339 100
700 600 500 400 300 200 100 0
79 81 83 85 87 89 91 93 95 97 99 01 03 Primary Revision RB, 1979-2003:
Total 9,4%
RB, 1992-2003 Total 10,8%
Men 10,4%
Women 11,3%
800
⊡ Fig. 11.5. Annual number of hybrid primary THR and the revisions generated from hybrid replacements in Sweden 1979–2003. The revi- sion burden (RB) is indicated for all observations and for 1992–2003 with separate RB for women and men
16000 14000 12000 10000 8000 6000 4000 2000 0
79 81 83 85 87 89 91 93 95 97 99 01 03 Primary
Revision
RB, 1979-2003:
Total 8,5%
RB, 1992-2003 Total 10,8%
Men 12,9%
Women 9,4%
⊡ Fig. 11.2. Annual number of primary and revision THR in Sweden 1979–2003. The revision burden (RB) is indicated for all observations and for 1992–2003 with separate RB for women and men
14000 12000 10000 8000 6000 4000
2000 0
79 81 83 85 87 89 91 93 95 97 99 01 03 Primary
Revision
RB, 1979-2003:
Total 7,8%
RB, 1992-2003 Total 9,9%
Men 12,2%
Women 8,4%
⊡ Fig. 11.3. Annual number of cemented primary THR and the revi- sions generated from cemented replacements in Sweden 1979–
2003. The revision burden (RB) is indicated for all observations and for 1992–2003 with separate RB for women and men
700 600 500 400 300 200 100 0
79 81 83 85 87 89 91 93 95 97 99 01 03 Primary Revision RB, 1979-2003:
Total 20,2%
RB, 1992-2003 Total 28,1%
Men 25,8%
Women 30,4%
800 900
⊡ Fig. 11.4. Annual number of uncemented primary THR and the revisions generated from uncemented replacements in Sweden 1979–
2003. The revision burden (RB) is indicated for all observations and for 1992–2003 with separate RB for women and men
Primary THA – Regions and Units
The number of primary procedures is increasing in rural hospitals, which might reflect the political ambition to concentrate prosthetic surgery to fewer elective units.
Since 2001, these units have performed more THAs than the county hospitals. There is a reciprocal decrease in the regional and university hospitals.
⊡Figures 11.6 and 11.7 illustrate the changes over time in crude prosthetic survival for each participating unit.
The average national 10-year survival has improved from 89.4% (±0.15) to 92.5% (±0.15) between the observation periods 1979–1991 and 1992–2003. During the period 1979–1991, 27% of the units did not differ significantly from the national average; 19% performed worse, and 44% performed better. For the period 1992–2003, 53% did not statistically differ from the national average, only 13%
performed worse and 34% were above average.
Revision Burden
The revision burden for the period 1992–2003 was 9.9%
for cemented implants, 28.1% for uncemented implants and 10.8% for hybrid implants (see ⊡Figs. 11.2 to 11.5).
For 1979–2003 the total revision burden for cemented implants was 7.8%. During the last ten-year period, the revision burden has been higher for men than that for women with the exception of uncemented and hybrid fixations. The revision burden for the uncemented and hybrid implants had a decelerating increase.
Re-Operations and Revision
Revision was the dominant subsequent procedure, ac- counting for 86% of all re-operations. Among the revi- sions aseptic loosening (73.9%), deep infection (7.9%), dislocation (7.5%) and periprosthetic fractures (5.7%) are the primary causes (⊡Table 11.3). A small, but continu- ous reduction of the total number of revisions has been observed over the past 5 years, which might indicate a trend of improved national quality, since the num- ber of patients at risk is constantly increasing. Patients with index diagnoses of rheumatoid joint disease and sequel to childhood hip disease are overrepresented in the group of multiple revisions as are those revised due to deep infection, periprosthetic fractures and dislocation (⊡Tables 11.3 and 11.4).
The reasons for revision have been relatively station- ary during recent years, except for revisions due to dis- location and/or technical reasons which have increased.
For patients with 5 years’ follow-up the cumulative revi- sion rate is 5–6 times higher for the group operated on in 1998 compared to those operated on in 1984 (⊡Fig. 11.8).
During the entire period the quality has improved in terms of fewer revisions because of aseptic loosening (⊡Fig. 11.9)
For cemented implants the results for the stem are generally better than the cup, with the flanged Charnley cup (see ⊡Fig. 11.1) as the sole exception. In uncemented and hybrid implants, the stem results are generally good, whereas the cups show poorer result.
100 98 96 94 92 90 88 86 84 82 80
⊡ Fig. 11.6. Survival rate with 95% confidence limits 1979–1991 for the individual units/hospitals in Sweden. All diagnoses and all reasons for revision are included. Each tick mark on the x axis indicates one hospital unit. The national average 89.4% is shown with horizontal red line (95% confidence limits indicated). 44% of the units had a result significant above the average and 19% below
100 98 96 94 92 90 88 86 84 82 80
⊡ Fig. 11.7. Survival rate with 95% confidence limits 1992–2003 for the individual units/hospitals in Sweden. All diagnoses and all reasons for revision are included. Each tick mark on the x axis indicates one hospital unit. The national average 92.5%is shown with horizontal red line (95% confidence limits indicated). 34% of the units had a result significant above the average and 13% below
11
⊡ Table 11.3. Diagnoses at index operation 1979–2003 in Sweden. The diagnoses at the index operation are indicated for first revisions and for 1, 2 and >2 repeated revisions
Diagnose at Index Operation
0 1 2 >2 Total
[n] [%] [n] [%] [n] [%] [n] [%] [n] [%]
Primary osteoarthritis 13,252 74.1 2,034 71.6 361 69.4 82 65.1 15,729 73.6
Fracture 1,684 9.4 233 8.2 36 6.9 6 4.8 1,959 9.2
Inflammatory disease 1,451 8.1 272 9.6 61 11.7 15 11.9 1,799 8.4
Childhood disease 843 4.7 182 6.4 38 7.3 15 11.9 1,078 5.0
Avascular necrosis 280 1.6 46 1.6 9 1.7 2 1.6 337 1.6
Traumatic osteoarthritis 150 0.8 45 1.6 9 1.7 6 4.8 210 1.0
Secondary osteoarthritis 49 0.3 6 0.2 1 0.2 0 0.0 56 0.3
Tumour 23 0.1 5 0.2 2 0.4 0 0.0 30 0.1
(Missing) 149 0.8 17 0.6 3 0.6 0 0.0 169 0.8
Total 17,881 100 2,840 100 520 100 126 100 21,367 100
⊡ Table 11.4. Reason for revision 1979 – 2003 in Sweden. The reason for revisions is indicated for first revisions and for 1, 2 and
>2 repeated revisions.
Reason for Revision
0 1 2 >2 Total
[n] [%] [n] [%] [n] [%] [n] [%] [n] [%]
Aseptic loosening 13,581 76.0 1,829 64.4 319 61.3 59 46.8 15,788 73.9
Infection 1,292 7.2 316 11.1 64 12.3 26 20.6 1,698 7.9
Dislocation 1,176 6.6 325 11.4 69 13.3 27 21.4 1,597 7.5
Periprosthetic fracture 966 5.4 221 7.8 38 7.3 2 1.6 1,227 5.7
Technical reason 447 2.5 71 2.5 17 3.3 2 1.6 537 2.5
Implant fracture 276 1.5 45 1.6 7 1.3 3 2.4 331 1.5
Miscellaneous 86 0.5 24 0.8 5 1.0 6 4.8 121 0.6
Pain only 57 0.3 9 0.3 1 0.2 1 0.8 68 0.3
Total 17,881 100 2,840 100 520 100 126 100 21,367 100
1 3 5 7 9 11 13 15 17 19 21 23 25
1993 1989 1986 1983
% revised 30 25 20 15 10 5 0
⊡ Fig. 11.8. Cumulative revision rate for all diagnoses and all reasons for revision for 1979, 1983, 1986, 1989 and 1993
1 3 5 7 9 11 13 15 17 19 21 23 25
1998 1993 1990 1987
% revised 0.8
0.6
0.4
0 0.2
1984
⊡ Fig. 11.9. Cumulative revision rate for recurrent dislocation for 1979, 1983, 1986, 1989 and 1993
Regions
The unadjusted procedure frequency per 100,000 in- habitants for patients aged 50 years and older and with the diagnosis of primary osteoarthritis is shown for the period 1992–2003. The national average is given for comparisons between the different regions (⊡Fig. 11.10).
The variation in procedure frequency (77–102/100 000 inhabitants) can be explained by a real difference in incidence of osteoarthritis requiring treatment but more likely also reflects limited resources, as for example in the Western region.
Discussion
The overall aim for the national register is to improve the quality of THA. The register can generate hypotheses suitable for either a specific study based on register data [5] or a carefully planned prospective clinical study. This is in accordance with the experience from the Norwegian Register [3]. The failure end-point definition used in registries has traditionally been revision. Revision is a well-defined event, but can potentially bias the result as use of both patient satisfaction and radiographic changes probably would give an inferior outcome [4]. In the 2003 report from the Swedish Total Hip Arthroplasty Register (www.jru.orthop.gu.se) the results are based on data for each primary procedure, as this has been captured since 1992, and adjustment for death is made on-line, which is a major improvement compared with to previous reports where part of the statistics were based on assumptions and estimates [11]. The improved failure definitions and accuracy in the epidemiological data will also facilitate comparisons and benchmarking among different national registries. Many countries have used registries for several years (Finland, Norway, Denmark, New Zealand, Hun- gary, Australia, Canada, and Romania) and others have started recently or are in the planning phase (Czech Re- public, Turkey, Slovakia, Moldova, Austria, England and Wales, France, Germany and the USA). Revision burden is one of the possible key figures that will enable crude comparison between different countries. It is important to clearly define and internationally agree on which key features that should be presented from national registers in order to make comparisons unbiased. An international Register Society could facilitate this process and there are ongoing efforts to initiate this.
A new and important development in the Swedish National THA Register during the past 3–4 years is the integration of individual patient related outcome data as well as the effort to register basic radiographic data. The Swedish health authorities encourage the registers current in function to give high priority for registration and inclu- sion of patient reported outcome [8].
General Swedish Trends in THA
The number of surgical procedures performed is probably too low to meet the future demands in an ageing popula- tion [9]. The regional variation in surgical procedures performed can be explained by local differences in patient demographics, incidence of disease, indications and eco- nomical restrictions.
For patients operated on in 1993, only about 5% are revised after 10 years. The proportion revised for the most common complication (aseptic loosening) has decreased to one third. In contrast, we see an increase in revision due to dislocation. This worrisome development may be related to an increase in primary THA for displaced cervical femoral fractures in the elderly, which is in contrast to a long tradi- tion of using percutaneous techniques with screws or pins as the primary intervention [10]. Another explanation is the equally long tradition in Sweden with use of small head sizes (22 or 28 mm). Besides these possible reasons surgical education, choice of surgical technique and implant design are factors that need a more detailed analysis in future studies. It is of particular importance to establish whether there has been deterioration in the teaching of surgical techniques. At present, only some of these factors can be evaluated from the register data, but this is an example of a hypothesis generated from register data where a more detailed analysis could provide answers.
The openness in the register provides a basis for fur- ther discussions locally, as each orthopaedic unit receives (in the confidential report) revision data, no matter where in Sweden the patients have been revised. Since the reg- ister started, it has been anticipated that a continuous improvement will follow. Although we do not exactly know the true impact of the register, the rate of implant survival has improved from 89.4% to 92.5% between the two periods 1979–1991 and 1992–2003. It is very satisfying that the proportion of orthopaedic units lying
120 100 80 60 40 20
0Stockholm Southeast South West Central Northern
⊡ Fig. 11.10. The unadjusted procedure frequency per 100,000 inha- bitants. Six regions and patients aged 50 years and older with the dia- gnosis primary osteoarthritis are shown for the period 1992–2003. The national average 92 THR/100,000 is indicated with a horizontal line
significantly below the national average has decreased.
However, the individual units’ patient profiles influence their results and in the analyses case-mix differences are not taken into account. Still, these crude revision rates only give a rough idea.
There is a potential conflict of interest between hip implant manufacturers and results reported from the register. Several manufactures have expressed concern that the register inhibits evolution, market introduction and clinical use for better functioning implants and surgi- cal techniques. Therefore the register, for more than 10 years, has invited all Swedish hip-implant manufacturers to their annual meetings. At these meetings, the current findings in the register are presented and discussed as well as any future projects. The companies can obtain on-line information about the results for their products. We have found this cooperation very rewarding.
The incidence of THA varies considerably between different regions as well as between different hospital types. There are systematic variations in choice of fixa- tion mode because university and regional hospitals tend to use more unproven devices and mainly in prospective evaluations. Research implies closer clinical and radio- graphic follow-up. Such monitoring might result in ear- lier re-operations due to polyethylene wear and osteolysis without any apparent symptoms or gross loosening.
The revision burden (RB) in the different regions var- ies between 8.9% and 11.5%. The lowest RB is noted in the Northern regions. One explanation might be that mostly well-documented implants have been used, but cultural differences, patients’ expectations and so-called patient’s delay might play important roles as well. An important bias lays in the fact that even if general practitioners find patients with failed implants, the patients may still hesi- tate to be referred for additional surgery. Furthermore, there is a consistently higher RB for men compared to women. This difference is accentuated inter-regionally, with a variation for men between 14.3% and 10.2%. This finding may be related to greater body weight and higher activity level followed by increased implant wear.
The average age at primary THA is generally higher for women than for men except when the indication is sequel to childhood disease. Whether this reflects dif- ferent access to surgery for women or that women sys- tematically tend to seek orthopaedic consultancy later is not clear, but it is important to analyse this phenomenon further. However, sequel to childhood disease and rheu- matoid joint disease, mostly in younger patients, are both over-represented in the multiple revision groups, indicat- ing that such patients should prompt extra care both in the primary and revision situation.
In a public health economy subjected to financial restrictions computation of cost-utility is important in order to allocate resources as optimally as possible. THA has advantageous cost-effectiveness when compared with
other medical treatments. League tables of different health care interventions, calculated by cost utility analyses, have been reported in the literature [6, 7, 14]. If a cost-utility table is going to be used as an instrument for allocating health-care resources, the studies should preferably be done at the same time, with use of identical outcome and cost evaluation methods. The current reported 10 years QALY cost of USD 3,000 (SEK 22,000) is very low com- pared to the Swedish threshold value USD 71,000 [2].
Summary and Conclusion
During recent years our ambition has been to improve the value of the register by analysing the patient’s own opin- ion of the results of THA. The preoperative routine has shown and confirmed that the unoperated patient with osteoarthritis is suffering from severe pain and has low general health related quality of life. The prospective 1- year results show extremely good pain relief and very high patient satisfaction, and a self-rated quality of life (EQ5D) that is equal to that of an age-matched normal popula- tion. At present we can show a very good cost–utility result and assert THA in comparisons with other medical interventions when priorities and resource allocation are discussed. Increased sensitivity of the register analysis and creation of routines that can reduce the number of follow- up visits after hip-replacement surgery and late revision cases with severe bone stock deficiencies is desirable. Fur- ther validation of our follow-up instruments combined with development of new instruments for adequate eco- nomic evaluations of THA surgery may provide powerful tools to further optimise the results of THA.
In the Swedish population, cemented THA has given an excellent result with decreasing revision frequency despite an increasing number of patients at risk. This is probably a combined result of both adequate and contem- porary surgical and cementing techniques as well as use of well-documented implants. The cemented THA is still a primary choice for the majority of Swedish patients in need of a THA.
Take Home Messages
I I
▬ The purpose of the Swedish National Hip Arthro- plasty Register is to monitor surgical techniques and prophylactic measures to minimise complica- tions by persisting continuous feed-back to all THA-performing units and to provide a warning system for rapid implant failures.
▬ All 81 orthopaedic units in Sweden, both public and private, participate voluntarily in the register.
The vast majority of the clinics are reporting data directly via the Internet.
11
▼
▬ All participating units have agreed to display their results open and public. However, the individual units’ patient profiles influence their results and in the analyses case-mix differences are not taken into account.
▬ Although we do not exactly know the true impact of the register, the rate of overall implant survival (including poor designs) has improved from 89.4%
to 92.5% between the two periods 1979–1991 and 1992–2003.
▬ The revision burden for cemented implants is con- siderable lower than for uncemented implant and even recent hybrid combinations. This is probably related to inferior liner fixation and polyethylene quality of the contemporary uncemented implants used in Sweden.
▬ Cemented THA constitutes more than 90% of the implant surgeries in Sweden and remains a pri- mary choice for the majority of Swedish patients.
▬ The surgical technique is of outmost importance and must be implemented through extensive educational efforts.
References
1. Charnley J. The low friction arthroplasty of the hip. Springer, New York, 1979
2. Eichler H, Kong SX, Gerth WC, Mavros P, Jönsson B: Use of cost- effectiveness analysis in health-care resource allocation deci- sion-making: how are cost-effectiveness thresholds expected to emerge? Value in Health 2004, 7(5):518–28
3. Furnes OH, Havelin LI, Espehaugh B, Engesæter LB, Lie SA, Vollset SE: Det norske leddproteseregistret – 15 nyttige år for pasienterne og helsevesenet. Tidskr Nor Lægeforen, 2003, 123:1367–69 4. Garellick G, Malchau H, HerbertsP: Survival of hip replacements.
A comparison of a randomized trial and a registry. Clin Orthop 2000(375):157–67
5. Havelin LI, Espehaug B, Engesæter LB: The performance of two hydroxyapatitecoated acetabular cups compared with Charnley cups. From the Norwegian Arthroplasty Register. J Bone Joint Surg Br 2002, 84B:839–45.
6. Laupacis A, Bourne R, Rorabeck C, Feeny D, Wong C, Tugwell P, Leslie K, Bullas R: Costs of elective total hip arthroplasty during the first year. J Arthroplasty, 1994. 9: p. 481–92.
7. Maynard A: Developing the health care market. Econ J 1991, 101:1277
8. National Health Care. Quality Registries in Sweden 1999. 2000, Stockholm: Information Department, The Federation of Swedish County Councils
9. Ostendorf M, Johnell O, Malchau H, Dhert WJ, Schrijvers AJ, Ver- bout AJ. The epidemiology of total hip replacement in The Nether- lands and Sweden: present status and future needs. Acta Orthop Scand, 2002, 73(3):282–6
10. Rogmark C, Carlsson A, Johnell O, Sernbo I: A prospective ran- domized trial of internal fixation versus arthroplasty for displaced fractures of the neck of the femur. Functional outcome for 450 patients at two years. j Bone Joint Surg Br, 2002, 84-B(2):183–88
11. Soderman P: On the validity of the results from the Swedish National Total Hip Arthroplasty register, in Dept. of Orthopaedic Surgery. Institute of Surgical Sciences. 2000, University of Gothen- burg: Gothenburg
12. Soderman P, Malchau H, Herberts P, Johnell O. Are the findings in the Swedish National Total Hip Arthroplasty Register valid? A comparison between the Swedish National Total Hip Arthroplasty Register, the National Discharge Register, and the National Death Register. J Arthroplasty, 2000, 15(7):884–9
13. »The_EuroQol_Group«, EuroQol – a new facility for the mea- surement of health-related quality of life. Health Policy, 1990, 16(3):199–208
14. Williams A: Economics of coronary artery bypass grafting. Br Med J, 1985, 291:326–34
Part V Perioperative Management, Complications and Prevention
Chapter 12 We Need a Good Anaesthetist for Cemented THA – 302
A. DowChapter 13 Perioperative Management – Rapid Recovery Protocol – 307
A.V. Lombardi, K.R. Berend, T.H. Mallory
