Indapamide is beneficial in reducing the incidence of adverse clinical outcomes in patients over the age of 80 years, and with systolic blood pressure between and mm Hg, with a good safety profile. After initial screening, patients with hypertension were randomized to either indapamide SR 1. A total of 3, patients were randomized: 1, to the indapamide arm and 1, to the placebo arm. The mean age at baseline was
|Published (Last):||11 September 2005|
|PDF File Size:||15.81 Mb|
|ePub File Size:||8.69 Mb|
|Price:||Free* [*Free Regsitration Required]|
Metrics details. Treatment for hypertension with antihypertensive medication has been shown to reduce stroke, cardiovascular events, and mortality in older adults, but there is concern that such treatment may not be appropriate in frailer older adults.
To investigate whether there is an interaction between effect of treatment for hypertension and frailty in older adults, we calculated the frailty index FI for all available participants from the HYpertension in the Very Elderly Trial HYVET study, a double-blind, placebo-controlled study of antihypertensives in people with hypertension aged 80 and over, and obtained frailty adjusted estimates of the effect of treatment with antihypertensive medication on risk of stroke, cardiovascular events, and mortality.
Data relating to blood pressure, comorbidities, cognitive function, depression, and quality of life were collected at entry into the study and at subsequent follow-up visits. The FI was calculated at entry, based on 60 potential deficits. Cox regression was used to assess the impact of FI at entry to the study on subsequent risk of stroke, total mortality, and cardiovascular events.
Models were stratified by region of recruitment and adjusted for sex and age at entry. Extending these models to include a term for a possible interaction between treatment for hypertension and FI provided a formula for the treatment effect as a function of FI.
For all three models, the point estimates of the hazard ratios for the treatment effect decreased as FI increased, although to varying degrees and with varying certainty. We found no evidence of an interaction between effect of treatment for hypertension and frailty as measured by the FI.
Both the frailer and the fitter older adults with hypertension appeared to gain from treatment. Further work to examine whether antihypertensive treatment modifies frailty as measured by the FI should be explored. Peer Review reports. The global population is ageing and with ageing comes an increased prevalence of both frailty and hypertension [ 1 - 3 ].
Treatment for hypertension with antihypertensive medication has been shown to reduce stroke, cardiovascular events, and mortality in older adults [ 4 , 5 ]. Nevertheless, the potential benefit associated with providing treatment for hypertension must be weighed against the potential risk of overtreatment excessive blood pressure lowering , polypharmacy, and the impact of side effects.
There is concern that the treatment may not be beneficial in all older adults, particularly the frailest [ 6 ]. The FI theoretically ranges from 0 to 1. FI may be calculated from any data set with sufficient information related to participant deficits the recommended minimum is 30 deficits that are not saturated within the data set and in a number of global population datasets has shown consistent relationships between ageing and mortality, with higher frailty scores at older ages and higher frailty associated with an increased risk of death [ 11 - 16 ].
The deficits included in the FI do not need to be independent of each other. Population studies of very old adults report a mean FI of around 0. The FI is applicable at any age, but of particular use in very elderly people, in whom frailty levels are more widely distributed than in the general population, i. The HYpertension in the Very Elderly Trial HYVET was a double-blind, placebo-controlled study of antihypertensives in people with hypertension aged 80 and over, which found that treatment with antihypertensives would lead to a reduction in risk of stroke, cardiovascular events, and total mortality [ 4 , 5 ].
However, participants recruited to HYVET, in common with many clinical trials and other studies, are likely to have been healthier than the general very elderly hypertensive population [ 18 ]. In consequence, the applicability of the results to the wider elderly population has been questioned, so that uncertainty remains as to whether treatment benefits also extend to the frailer elderly people [ 6 ]. To investigate further, we calculated the FI for all available HYVET study participants and obtained frailty-adjusted estimates of the effect of antihypertensive treatment in very elderly people.
Participants gave written informed consent and were assessed at baseline and at 3-monthly intervals for the first year and 6-monthly intervals thereafter until they either died, withdrew from the study, or the trial ended. Data relating to blood pressure, comorbidities, cognitive function, depression, and quality of life were collected at baseline and at subsequent follow-up visits.
Key relevant endpoints were selected prior to the start of the trial and these included incident stroke the primary endpoint , total mortality, and incident cardiovascular events. Data relating to trial endpoints were collected as these occurred and the endpoints validated by an independent blinded committee. All appropriate ethical and regulatory permissions were obtained. Trial registration number is NCT The FI was calculated at entry to the study and based on 60 deficits as detailed in Additional file 1.
Each deficit was coded as either present or absent in accordance with previous published methodology [ 19 ]. The number of deficits present was counted for each study participant and divided by 60, the maximum number of deficits possible in our data, to give the FI. Information on all 60 deficits was not available for all participants, partly because some had opted out of the quality of life sub-study and partly because of the usual missing data issues. Where missing data meant that the FI calculation for a particular participant would be based on fewer than the intended 60 possible deficits, the FI was calculated as the number of deficits divided by the number available for that participant.
Where missing data meant that the FI calculation for a particular participant would be based on fewer than the recommended minimum of 30 possible deficits, the FI was not calculated but set to missing and the subject excluded from the analysis.
Cox regression [ 20 ] was used to assess the impact of FI at entry to the study on subsequent risk of fatal and non-fatal stroke, total mortality, and cardiovascular events fatal and non-fatal stroke, myocardial infarction and heart failure.
Models were stratified by region of recruitment Western Europe, Eastern Europe, China to allow each to have a different baseline hazard function calibrate FI and adjusted for sex and age at entry to the study. There was no adjustment for baseline cardiovascular disease as this forms part of the FI.
We fitted Cox regression models stratified by region with terms for baseline age, sex, FI, treatment, and an interaction between treatment and FI to obtain a formula for the treatment effect as a function of FI [ 22 ] to illustrate the impact of FI upon the estimate of treatment effect obtained from our models. The relationship between baseline FI and subsequent drop out was investigated to establish whether participants for whom the FI could not be calculated, owing to incomplete data, differed substantially from those for whom the FI could be calculated.
The HYVET trial randomised 3, participants, of whom only the 2, who consented to complete an additional quality of life questionnaire provided sufficient data to allow the calculation of the FI. There did not appear to be any imbalance between the treatment groups.
We also compared those for whom the FI was calculable with those for whom it was not. There was no difference between them in terms of age, sex, previous cardiovascular disease, or baseline sitting systolic blood pressure but there was a difference in cognitive test score; the median Mini-Mental State Exam MMSE score at entry to the study in those in whom FI was calculable was The distribution of the constituents used to calculate the FI was similar in the two treatment groups Additional file 2 : Table S1.
The distribution of FI at entry to the study was skewed with mean and median values 0. On average, FI scores increased slowly with age 0. Greater FI at entry to the study was associated with an increased risk of death HR, 1. Adjustment for treatment group, age, and sex, and stratification by region of recruitment did not alter these findings.
Median follow-up was similar in the two treatment groups. The proportionality assumption was not violated P values for the global Grambsch and Therneau tests of the proportional hazards assumption for the models for time to death, time to cardiovascular events, and time to stroke were 0. The adjusted hazard ratios HR for all three endpoints were similar to those seen in the main trial results which were HR, 0. Repeating these analyses with previous cardiovascular disease excluded from the calculation of FI did not materially affect the results.
There was no significant interaction between treatment effect and frailty for any of the three endpoints P values for the interaction term in the interaction models for stroke, cardiovascular events, and total mortality were 0.
For all three models the point estimate of the HR for the treatment effect decreases as FI increases, although to varying degrees and with varying certainty. There were patients for whom baseline FI was calculable but who withdrew from the study and were therefore censored in the analysis at the date of withdrawal. The median FI was 0. Frailty as measured by the FI was a strong predictor of stroke, total mortality, and cardiovascular events in the HYVET trial, which is in agreement with multiple analyses from observational datasets [ 8 - 13 ].
We found no evidence of an interaction between baseline FI and treatment with antihypertensives on risk of stroke, death from all causes, or cardiovascular events in very elderly people. Furthermore, the burden of frailty amongst HYVET participants at baseline was similar to that seen in population studies [ 13 - 17 ].
Overall this suggests both that the HYVET population is more representative in terms of frailty than may have been supposed, and that benefits associated with blood pressure lowering treatment are accrued in both frailer and fitter older adults. In general, as the degree of frailty increases, so does the chance of functional impairment [ 28 , 29 ] or mobility impairment [ 30 ].
Furthermore, as the associations between risk factors and adverse outcomes may differ at extreme age and in frailty subgroups, and be dependent on risk factor change over time, a more nuanced approach to the interpretation of results may be required.
However, our results must be interpreted with caution. Nevertheless, it is reassuring that those for whom an FI was not calculable differed chiefly only in relation to cognition, scoring on average one point higher on the MMSE at entry, statistically but not clinically different. It is also possible that treatment with an anti-hypertensive may have affected participant withdrawals differentially, leading to bias in our estimate of the treatment effect.
However, we found no difference in withdrawal rate between the treatment groups Although only 2, participants were available from the HYVET trial for these analyses these data still represent a significant number of older adults. Although the estimate of the log HR decreases as FI increases for all three endpoints, the interaction term in the respective models were not significant overall the relationship between frailty and treatment effect was not strong.
This analysis has some strengths. In particular, the use of data from a double-blind, placebo-controlled clinical trial also allows exploration of this by randomised group providing a more robust finding that would be possible from observational data alone. The similarity of the FI at baseline with large observational datasets supports the potential applicability of the results. As far as the authors know, this is the first time the FI has been used in analysing the results of a clinical trial, particularly a hypertensive very elderly group.
Our analyses show that in the HYVET study participants there was no evidence of an interaction between treatment effect and frailty. Further work in is needed to confirm these findings in other similar datasets and to explore whether antihypertensive treatment modifies frailty as measured by the FI. Additionally, examining this more holistically, looking at impact of treatment over time, would be of additional benefit and is motivating further research by our group.
This included approval in all participating countries by national drug or healthcare agencies, national ethics committees, and where required local ethical committees. The documentation is therefore large and complex. Full copies are archived by Imperial College London. World Population Ageing — Health Survey for England , Trend tables: adult trend tables. Accumulation of deficits as a proxy measure of aging. Sci World J. N Engl J Med. Immediate and late benefits of treating very elderly people with hypertension: results from the active treatment extension to Hypertension in the very elderly randomised controlled trial.
Abramov D, Cheng H. Controversy in treating the oldest old with hypertension: is the hypertension in the very elderly trial the final answer? J Am Geriatr Soc. Frailty consensus: A call to action. J Am Med Dir Assoc. Outcome instruments to measure frailty: a systematic review. Ageing Res Rev. Frailty in elderly people: an evolving concept. A comparison of four frailty models.
Frailty and mortality among Chinese at advanced ages. Analysis of frailty and survival from late middle age in the Beijing longitudinal study of aging.
The Hypertension in the Very Elderly Trial (HYVET)
Metrics details. Treatment for hypertension with antihypertensive medication has been shown to reduce stroke, cardiovascular events, and mortality in older adults, but there is concern that such treatment may not be appropriate in frailer older adults. To investigate whether there is an interaction between effect of treatment for hypertension and frailty in older adults, we calculated the frailty index FI for all available participants from the HYpertension in the Very Elderly Trial HYVET study, a double-blind, placebo-controlled study of antihypertensives in people with hypertension aged 80 and over, and obtained frailty adjusted estimates of the effect of treatment with antihypertensive medication on risk of stroke, cardiovascular events, and mortality. Data relating to blood pressure, comorbidities, cognitive function, depression, and quality of life were collected at entry into the study and at subsequent follow-up visits. The FI was calculated at entry, based on 60 potential deficits. Cox regression was used to assess the impact of FI at entry to the study on subsequent risk of stroke, total mortality, and cardiovascular events.
Treatment of Hypertension in Patients 80 Years of Age or Older
Among very elderly patients with hypertension, does active treatment with antihypertensives reduce the rate of fatal or nonfatal stroke when compared with placebo? A meta-analysis of RCT data found that treatment of hypertension in this age group was associated with a statistically significant reduction in major CV events and HF but no reduction no CV mortaltiy and an increased risk of all-cause mortality. After a 2-month placebo run-in phase, 3, patients were randomized to active treatment or placebo with stratification according to age and sex. Stepwise treatment consisted of a diuretic indapamide sustained release 1. Additional non-protocol—specified antihypertensives were allowed for up to three months, after which patients were given the option of coming off study or entering open follow-up. The primary outcome was the rate of fatal or nonfatal stroke excluding TIA.