Pekka Paavolainen et al.
Causes
of death after total hip arthroplasty
A nationwide cohort study with 24,638 patients
J Arthroplasty April 2002 • Volume 17 • Number 3 pages 274 - 81
(Adapted and abbreviated)
Total hip arthroplasty
(THA) improves all health-related
quality-of-life functions significantly including walking ability [1,2].
More than 5,000 primary THAs are performed in
Finland (population 5.0 million) annually.
There are few or only small-scale reports concerning other causes of death among these patients [23]. Although all the previous reports support the conclusion that there is a better overall survival rate among these patients, we were interested in ascertaining whether the mortality resulting from certain diseases differs from that of the general population. We investigated the causes of death for all the patients with primary THA for primary arthritis through the National Register of Arthroplasty in Finland from 1980 to 1995. The risk level in the ICD-10 categories was compared with that of the general population matched for age and sex.
|
Material and methods |
All THA patients who were operated on between 1980 and 1995 and for whom primary arthritis had been the indication for the operation were identified from the files of the National Register of Arthroplasties kept by the National Agency for Medicine in Finland [3,4]. Using the unique personal identification numbers assigned to all residents of Finland from January 1, 1967, onward, the Population Register Centre of Finland was searched for all THA patients, and the dates of death were obtained from the same source. The first surgical intervention (date of THA) was taken as the starting point of the follow-up period. The causes of death for this THA cohort were obtained from the Central Statistical Office of Finland through the statistics covering the period to December 31, 1996. All of the patients we identified were available for follow-up evaluation.
It has been reported that patients with rheumatoid arthritis and other autoimmune diseases have a higher mortality rate than the general population [24], an increased risk of non-Hodgkin's lymphomas and leukemia [25–27], and a less favorable survival rate after a THA or total knee arthroplasty [16,28]. The increased use of antirheumatic drugs also may affect the risk of cancer [29]. Patients suffering from rheumatoid arthritis were excluded.
From 1980 to 1995, a total of 33,694 primary THAs for primary arthritis were performed on 24,638 patients (62% were women). Greater than 40% of the patients who underwent surgery were of working age. Seventy-one percent of the femoral components and 65% of the acetabular components were cemented. The articulation of all these prostheses was of a polyethylene-on-metal design; no metal-on-metal constructions have been used since 1980.
The cohort for follow-up included 9,479 men and 15,159 women. The numbers of person-years were 56,599 and 96,811 (Table 1).
|
Table 1. Numbers of male and female patients followed for causes of death by age at operation and number of person years to December 31, 1996
|
||||
|
|
Males |
Females |
||
|
Age at Operation (y) |
n |
Person-Years |
n |
Person-Years |
|
15–29 |
10 |
33 |
11 |
36 |
|
30–44 |
156 |
500 |
168 |
652 |
|
45–59 |
2,060 |
7,385 |
2,216 |
8,198 |
|
60–74 |
5,751 |
31,658 |
9,214 |
49,854 |
|
|
1,502 |
17,023 |
3,550 |
38,071 |
|
Total |
9,479 |
56,599 |
15,159 |
96,811 |
75The mean length of follow-up per person was 6.2 years. Of the person-years, 64% were in the follow-up category of <5 years. The number of perioperative deaths among these patients whose operations had been reported to the National Register was 70; this figure equals 0.16% of all operations.
The total SMR was 0.69 (Table 2), and there was no difference between male and female patients.
The risk increased slightly during the follow-up period, although it was still lower than unity (SMR, 0.84; 95% CI, 0.81–0.87) from the 5th year onward (Table 3).
|
l
|
There were significantly low SMRs for cancers (0.54), accidents (0.74), cardiovascular diseases (0.70), and respiratory diseases (0.46). The decreased risk of death resulting from lung cancer (SMR, 0.53; 95% CI, 0.44–0.62) was mainly attributable to men, especially during the first 5 years after THA (SMR, 0.33; 95% CI, 0.24–0.44). The total mortality resulting from cancers of the lymphoid and hematopoietic tissues was slightly lower than that of the population matched for age and sex (SMR 0.87) but increased during the follow-up period, being 1.31 after the 5th postoperative year.
During the 16-year follow-up period, 2,059 male patients died. The expected number was 3,002, and the SMR was 0.69 (95% CI, 0.66–0.71). The risk was especially low during the first 5 years after THA (SMR, 0.57; 95% CI, 0.54–0.60), but it increased to unity thereafter. The SMR for prostate cancer was 0.80 (95% CI, 0.64–0.99).
Among the women, there were 2,567 observed deaths versus 3,743 expected (SMR, 0.69; 95% CI, 0.66 to 0.71). Similar to the men, the risk increased with the duration of follow-up: from 0.55 during the first 4 years to 0.84 from the 5th year onward. The mortality for cancers of the female genitals did not differ from that expected. The risk of colon and rectum cancer was low in women only (SMRs, 0.49 and 0.43).
Among the diseases of the nervous system and sensory organs, there was a
constant and significant decline in risk for THA
patients in the category of dementia and Alzheimer's disease for men (SMR,
0.50) and women (SMR, 0.52); total
SMR was 0.51 (95% CI, 0.44–0.58). Total
SMRs for diabetes were low as well: 0.40 for the
men and 0.34 for the women. In the category of the circulatory system, there
were low SMR values for most of the diseases and
for both genders: myocardial infarction (0.73), hypertension (0.68), other
ischemic diseases (0.70), other heart diseases (0.57), and
cerebrovascular diseases (0.70) Tables 2
and 3).
|
Discussion |
A particular feature of mortality in Finland has been the large difference between that of men and women [34,40]. During the follow-up period, 21.7% of the men and 16.9% of the women died. The relative reduction in mortality for all causes was identical (SMR, 0.69). In terms of absolute mortality, THA benefits men more than women. This benefit is attributable mainly to the male patients' reduced risk of death as a result of cardiovascular and respiratory diseases. The systematically low SMRs of alcohol-related and tobacco-related diseases also indicate that the living habits of the cohort members are likely to be healthier than those of the average Finn [41].
The findings for this cohort indicate that THA patients are not at greater risk of dying of a disease in any specific category. In the case of the causes of death with lowered SMRs, the explanation is attributable to factors other than the THA per se. The greatest length of our follow-up (16 years) is still fairly short, and further follow-up of the cohort is needed. Whether the lowered mortality of the THA patients will disappear over a longer period or whether it is a permanent phenomenon remains obscure for the time being.