Aortic Valve Replacement in Patients 80 Years of Age and Older
Survival and Cause of Death Based on 1100 Cases: Collective Results From the UK Heart Valve Registry
Background Aging of the population and advances in preoperative and postoperative care are reflected in an increasing number of patients ≥80 years of age undergoing aortic valve replacement (AVR) in the United Kingdom. The present study presents data on postoperative 30-day mortality, actuarial survival, and cause of death based on a large collective patient population.
Methods and Results Data were extracted from the UK Heart Valve Registry. From January 1986 to December 1995, 1100 patients ≥80 years of age underwent AVR and were reported to the registry. Six hundred eleven patients (55.5%) were women. The mean follow-up time was 38.9 months. The 30-day mortality was 6.6%. Of the 73 early deaths, 42 were due to cardiac reasons. The actuarial survival was 89%, 79.3%, 68.7%, and 45.8% at 1, 3, 5, and 8 years, respectively. After the first 30 postoperative days, 144 of the 205 deaths were due to noncardiac reasons. Malignancy, stroke, and pneumonia were the most common causes of late death. Bioprosthetic valves were implanted in 969 patients (88%) and mechanical valves in 131 (12%) patients. There was no difference in early mortality and actuarial survival between the two groups (P>.05).
Conclusions The above results suggest that under the selection criteria for AVR currently applied in the United Kingdom, patients ≥80 years of age show a satisfactory early postoperative outcome and moderate medium-term survival benefit.
Life expectancy has increased in all age groups of the UK population over the last 3 decades. In 1994, an 80-year-old female could expect to live another 8.4 years and a male 6.5 years compared with 6.3 and 5.2 years, respectively, in 1960.1 Hence, by the year 2000, in England alone, 2 million people (4.2% of the population) will be ≥80 years of age.2 The increase in the elderly population and advantages in surgical technique and postoperative care have correlated with a recent rise in the number of people >80 years of age who undergo cardiac surgery. A number of studies, mostly from the United States, have evaluated the outcome of AVR in patients of this age group.3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Nevertheless, the total number of patients included in these reports remains relatively small.
The aim of the present study was to evaluate the survival and cause of death after AVR in patients ≥80 years of age by analyzing data extracted from the UKHVR.
The UKHVR is a computerized database that is based at the Hammersmith Hospital (London) and collects prospective data of patients undergoing heart valve replacement in ≥30 UK cardiac centers. It is funded by the Department of Health (Medical Devices Agency) and includes data from January 1, 1986, comprising data of ≈50 000 patients by the end of 1995. A report on the UKHVR is published annually.18 Data collected includes patient identification as well as comprehensive data relating to the operation date, valve position, and type of implanted prosthesis. These data are provided by the hospital at which the valve replacement is performed. To keep the UKHVR data complete and achieve maximum compliance by the participating cardiac centers, information regarding preoperative status and combined CABG is not included. Date and cause of death are tracked through the ONS, formerly the Office of Population Censuses and Surveys. The ONS receives death certificates for all UK citizens and provides complete data on the exact date, place, and certified cause of death. The UKHVR is notified when any flagged patient subsequently dies, and within 6 weeks of the patient’s death, a copy of the patient’s death certificate, which includes the date, place, and certified cause of death, is sent to the registry office where the details are entered into the database.
For this study, we extracted information on patients who underwent AVR between January 1986 and December 1995 and were ≥80 years of age at the time of the operation. Analyzed data included patient age and sex, date of operation, valve position, type of valve implant, follow-up time, and time and cause of death. Status was assessed at December 31, 1996. Early death was defined as death occurring within the first 30 days after surgery.
Results were analyzed under consideration of the revised “Guidelines for Reporting Morbidity and Mortality After Cardiac Valvular Operations” issued by the Ad Hoc Liaison Committee for Standardizing Definitions of Prosthetic Heart Valve Morbidity.19
Results are expressed as the mean±SD. The χ2 test was used to compare categorical variables. Actuarial survival curves were calculated by the Kaplan-Meier method and compared by means of the log rank test. The predictors of survival were analyzed by the Cox proportional hazards model. A value of P<.05 was considered significant.
During a 10-year period, from January 1986 to December 1995, 29 360 patients underwent first-time AVR and were registered in the UKHVR. Of these patients, 1100 (3.7%) were ≥80 years of age at the time of the operation. Of these octogenarians, 2 underwent a second AVR during the study period. Since 1986, the number of AVRs performed annually in patients >80 years of age has nearly doubled every 2 years (Fig 1⇓).
Complete follow-up was available in 983 (95.7%) of the 1027 patients who survived the first 30 postoperative days (mean, 38.9 months; range 1 to 128 months). In total, 278 patients died during the follow-up period.
The age of the 1100 patients included in the study ranged from 80 to 93 years with a mean of 82.2±2 years. Six hundred eleven (55.5%) patients were female.
Of the 1100 patients, 73 (6.6%) died during the first 30 postoperative days (Table 1⇓). There was no significant difference in the 30-day mortality between men (6.3%) and women (6.8%). The total group of 29 360 patients, including all ages, who underwent first AVR and were registered in the UKHVR displayed a 30-day mortality of 4.3%, which is significantly lower compared with the octogenarians (P<.001).
The overall 1-, 3-, 5-, and 8-year actuarial survival rates were 89%, 79.3%, 68.7% ,and 45.8%, respectively (95% CIs: 87.12 to 90.92%, 77.23 to 81.47%, and 72.42 to 65.02% at 1, 3, and 5 years respectively; Table 1⇑). Patients who died within the first 30 postoperative days were also included in the survival analysis. By multivariate analysis, based on the Cox proportional hazards model, male sex was associated with reduced overall survival (P=.026; hazard ratio male/female, 1.3/1). In contrast, age at the time of operation (P=.954) and valve type (P=.326) were no predictors of survival (Table 2⇓).
Actuarial survival rates in the overall group of patients based on 29 360 first AVRs were 91.6%, 86.4%, 81.2%, and 71.1% (95% CIs: 91.30% to 92.03%, 85.94% to 86.90%, 80.68% to 81.87%, and 70.19% to 72.04%) at 1, 3, 5, and 8 years, respectively. As expected, the overall group displayed significantly better long-term survival compared with the patients >80 years of age (P<.05). Fig 2⇓ shows the actuarial survival curves in patients >80 years of age and in the overall group undergoing AVR.
Cause of Death
The causes of early (30-day) and late death are listed in Table 3⇓. Of the 278 deaths, 73 (26.3%) occurred within the first 30 postoperative days. Forty-two early deaths (15.1%) were related to cardiac causes. Of the 205 (73.7%) late deaths, only 61 (21.9%) were cardiac in nature. When a death occurred within the first 30 postoperative days, it was significantly more likely to be of cardiac causes (P<.001). In total, 103 of all 278 deaths (37%) were cardiac. In addition to the causes of death shown in Table 3⇓, several patients died because of a variety of other reasons such as gastrointestinal obstruction, respiratory failure, accidental trauma, and suicide.
Type of Valve Implant
Twenty-eight different types of aortic valve implant were used in the 1100 patients (Table 4⇓). Of these patients, 969 (88%) received a bioprosthetic valve implant and 131 patients (12%) had a mechanical valve implant. There was no significant difference in the 30-day mortality and actuarial survival between the two groups (Tables 2⇑ and 5⇓ and Fig 3⇓), although the small number of patients in the mechanical valve group remaining at risk after the third postoperative year precludes reliable comparison of the survival curves. Nine of the 82 female patients (10.9%) versus only 1 of the 49 male patients (2%) who received a mechanical valve implant died within the first 30 postoperative days. However, this difference failed to reach statistical significance (P=.055).
Heart valve replacement operations are being performed with increasing frequency in patients >80 years of age. Although mitral valve replacement remains a rather uncommon procedure in this age group,20 AVR has been increasingly performed over the last decade.3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 When the decision to operate is made, the general condition of the patient is often above average for his or her age group. Previous studies have suggested that AVR in octogenarians carries an increased but acceptable risk compared with that in younger age groups. Table 6⇓ summarizes the publications reporting operative outcome after AVR in patients >80 years of age. They give a collective total number of ≈800 patients with a wide variation in the definition of early mortality.
The significance of the present study is that it comprises the largest reported series of AVR in the >80-year-old population and reflects experience from >30 UK cardiac surgical centers. The UKHVR is provided with reliable data regarding date and cause of death, allowing accurate calculation of 30-day mortality and actuarial survival rates. Fig 1⇑ shows that the number of AVR operations performed per year in patients >80 years of age has significantly increased in the past 10 years in the United Kingdom. Twenty-three AVRs were performed in 1986 and 144 AVRs were performed in 1992, increasing to 239 AVRs in 1995. In 1986, only 0.81% of the total number of AVRs was performed in octogenarians (23 of 2810). The ratio was 5.88% in 1995 (239 of 4064). This change parallels the general increase in the number of cardiac operations performed on octogenarians during the past decade and the rising mean age of patients undergoing heart valve replacement in the United Kingdom. The mean patient age was 58.5 years in 1986, 61.0 years in 1990, and 63.4 years in 1994, including all ages and all types of valve replacement. The percentage of patients >70 years of age undergoing valve replacement in the United Kingdom increased from 12.3% in 1986 to 28.9% in 1994.18
Thirty-Day Mortality and Actuarial Survival
The scope for AVR in the octogenarian with symptomatic aortic stenosis was summarized in a recent review article.21 In previous studies, early mortality rates in patients over 80 vary from 0% to 11.1% after isolated AVR and from 0% to 25% after combined AVR and CABG (Tables 5⇑ and 6⇑). Most of these studies include relatively small patient groups, although the wide variation in early mortality appears to be unrelated to the number of patients studied. In our report, the 30-day mortality was 6.6% in patients >80 years of age, whereas the total all-ages group of patients undergoing AVR had a 30-day mortality of 4.3% (1263 of 29 360).18 Despite the statistically significant difference (P<.001), these results suggest that the octogenarians performed well within the total group.
It is worth noting that male sex in octogenarians appear to associated with lower overall survival (P=.026). This is probably expected in view of the shorter life expectancy of the male sex >80 years of age, although the 95% CI includes an area of nonsignificance (Table 2⇑).
With regard to actuarial survival and for a mean follow-up time of 38.8 months, patients >80 years of age displayed significantly lower survival rates compared with the total group. Previous literature reports demonstrated actuarial survival rates in patients >80 years of age that are similar to those of our study.8 9 10 13 Survival rates of 84% to 93%, 79% to 82%, and 57% to 82% were reported after isolated AVR at 1, 3, and 5 years, respectively. After combined AVR and CABG, the survival rates were 76% to 92%, 65% to 79%, and 47% to 78% at 1, 3, and 5 years, respectively.
Cause of Death
Similar to previous reports5 9 15 , most early deaths were due to cardiac reasons (Table 3⇑). Non–valve-related cardiac failure was the most common certified cause of death from cardiac-related causes. Myocardial infarction caused early death in only 4 patients (5.4% of early deaths). Most late deaths were caused by noncardiac reasons, with malignancy being reported as the most common cause of late death. Forty-two patients died of cancer (20.4% of late deaths) after a mean time of 31 months (range, 3 to 93 months). It is likely that the malignancy became detectable after the patients’ AVRs had been performed. However, it is worth noting that 15 patients died of cancer within 18 months after cardiac surgery. This fact suggest that there is probably scope for improvement of the preoperative cancer screening of elderly patients.
Non–valve-related cardiac failure, valve-related stroke, and pneumonia were other common causes of late postoperative death. The relatively high incidence of these events is likely to be related to the advanced age of the population studied.
Bioprosthetic Versus Mechanical Valve Implant
Bioprosthetic valve implants have been recommended and used extensively for heart valve replacements in the elderly. The potential avoidance of lifelong anticoagulation and the associated relatively slower rate of structural valve deterioration in this age group are perceived as the main advantages. Previous investigators have identified the influence of age on structural failure of bioprosthetic valves, showing 100% freedom from structural valve deterioration at 13 years for AVR with the Carpentier-Edwards porcine bioprosthesis for patients >80 years of age.22 The implantation of porcine bioprostheses for AVR in elderly patients has been recommended in recent articles.23
In our study, 88% of the >80-year-old patients undergoing AVR received a bioprosthesis. There was no significant difference in survival between the bioprosthetic and mechanical valve groups across the study period. Longer follow-up, particularly for patients operated on in the 1990s, may be of interest.
In the elderly, preoperative risk factors such as emergency surgery, myocardial infarction, and New York Heart Association class IV disease have been shown to be associated with the likelihood of early death after cardiac surgery.5 Multivariate analysis in a recent study of 322 patients >80 years of age who underwent AVR showed that female sex, renal impairment, CABG, ejection fraction <35%, and chronic obstructive airway disease are independent predictors of operative mortality.15 Simultaneous coronary revascularization in octogenarians undergoing valve replacement is known to be related to an increased early mortality rate (Table 5⇑), probably as a result of the obviously higher incidence of ischemic heart disease in this group. As stated earlier, the number of patients registered in the UKHVR who had simultaneous CABG is not known. A further limitation of our study is the lack of information on the preoperative clinical status of the patient, which precludes risk assessment of the different clinical groups.
Despite these limitations, this study presents data based on a significantly large collective population of octogenarians undergoing AVR (n=1100) and suggests that the operative selection criteria currently applied to this age group in the United Kingdom produce a satisfactory early postoperative outcome (6.6% 30-day mortality). The mid-term survival benefit is, as expected, moderate because of the advanced age of the population studied.
Selected Abbreviations and Acronyms
|AVR||=||aortic valve replacement|
|CABG||=||coronary artery bypass graft surgery|
|ONS||=||Office for National Statistics|
|UKHVR||=||UK Heart Valve Registry|
We are grateful to Deborah Ridout for help with statistical analysis. We would also like to thank the Medical Devices Agency (Department of Health) for funding the UKHVR and the following centers of cardiac surgery for providing us with data: Aberdeen Royal Infirmary, United Bristol Health Care Trust, Broadgreen Cardiothoracic Centre, Brompton Hospital, Brook General Hospital, Castle Hill Hospital, Edinburgh Royal Infirmary, Freeman Group of Hospitals, Glasgow Royal Infirmary, Glenfield General Hospital, Guy’s Hospital, Hammersmith Hospital, Harefield Hospital, John Radcliffe Hospital, Killinbeck Hospital, King’s College Hospital, Leeds General Infirmary, The London Hospital, London Chest Hospital, Manchester Royal Infirmary, Middlesex Hospital, The National Heart Hospital, Northern General Hospital, North Staffordshire Royal Infirmary, Nottingham City Hospital, Papworth Hospital, Queen Elizabeth Hospital, Royal Victoria Hospital, South Cleveland Hospital, Southampton General Hospital, St Bartholomew’s Hospital, St George’s Hospital, St Mary’s Hospital, St Thomas’ Hospital, University Hospital of Wales, Victoria Hospital, Walsgrave Hospital, Western Infirmary, and Wythenshawe Hospital.
- Received April 28, 1997.
- Revision received June 18, 1997.
- Accepted June 26, 1997.
- Copyright © 1997 by American Heart Association
Population Trends. J Office Natl Stat. 1986;86:50.
Office of Population Censuses and Surveys. Subnational Population Projections for England. 1995. Series 9:3.
Levinson JR, Akins CW, Buckley MJ, Newell JB, Palacios IF, Block PC, Fifer MA. Octogenarians with aortic stenosis: outcome after aortic valve replacement. Circulation. 1989;80(suppl I):I-49-I-56.
Elayda MAA, Hall RJ, Reul RM, Alonzo DM, Gillette N, Reul GJ Jr, Alonzo DM, Gillette E, Reul GJ Jr, Cooley DA. Aortic valve replacement in patients 80 years and older: operative risks and long-term results. Circulation. 1993;88(pt 2):11-16.
Pasic M, Carrel T, Laske A, Bauer E, Turina J, Jenni R, von Segesser L, Turina M. Valve replacement in octogenarians: increased early mortality but good long-term result. Eur Heart J. 1992;13:508-510.
Logeais Y, Langanay T, Roussin R, Leguerrier A, Rioux C, Chaperon J, de Place C, Mabo P, Pony JC, Daubert JC, Laurent M, Almagne C. Surgery for aortic stenosis in elderly patients: a study of surgical risk and predictive factors. Circulation. 1994;90:2891-2898.
Medical Devices Agency. The United Kingdom Heart Valve Registry. Report 1994.
Edmunds LH Jr, Clark RE, Cohn RE, Grunkemeier GL, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Eur J Cardiothorac Surg. 1996;10:812-816.
Asimakopoulos G, Edwards M-B, Brannan J, Taylor KM. Survival and cause of death after mitral valve replacement in patients aged 80 years and over. Eur J Cardiothorac Surg. In press.
Springings DC, Forfar JC. How should we manage symptomatic aortic stenosis in the patient who is 80 or older? Br Heart J. 1995;74:481-484.
Burr LH, Jamieson WRE, Munro AI, Miyagishima RT, Janusz MT, Ling H, Hayden RI, Tutassaura H, Fradet G, Gudas VM, Tyers GFO. Structural valve deterioration in elderly patient populations with the Carpentier-Edwards standard and supra-annular porcine bioprostheses: a comparative study. J Heart Valve Dis. 1992;1:87-91.
Burr LH, Jamieson WRE, Munro AI, Miyagishima RT, Germann E. Porcine bioprostheses in the elderly: clinical performance by age groups and valve positions. Ann Thorac Surg. 1995;60:S264-S269.