Vascular disease (VD) is the main cause of morbidity and mortality worldwide.1,2 Its prevalence increases with age, and age in turn is one of the most determinant factors of the disease.3,4

The progressive rise in life expectancy has caused our population to become increasingly older. Demographic projections by Eurostat estimate that the percentage of people ≥80 years of age in the European Union will increase from 5.8% to 14.6% between 2019 and 2100,5 and that the number of people >90 years of age will increase from 4.25 million in 2015 to 13.66 million in 2050.6 In Spain, life expectancy is currently 81.2 years, but will practically reach 90 by the year 2100.3

It is estimated that 25% of all nonagenarians have established VD.4 Elderly patients also experience a greater number of events and have a poorer prognosis. Frailty, comorbidity, multiple disease conditions and cognitive impairment cause this age group to be underrepresented in most clinical trials, with no available evidence-based clinical practice guides; recommendations are thus extrapolated from younger populations, and the individualization of decisions is advised. The aforementioned situation evidences the need for real-life studies to obtain evidence on the best therapeutic strategies, and this in turn requires knowledge of the characteristics and approach to VD in elderly patients.

For these reasons, the decision was made to conduct the NONAVASC-2 registry, with the primary objective of determining the clinical characteristics as well as the suitability of pharmacological management in nonagenarian patients with established VD admitted to Spanish Departments of Internal Medicine due to any reason.

The NONAVASC-2 registry is a prospective, multicenter observational study conducted in the Departments of Internal Medicine of 37 hospitals from all Spanish regions (Autonomous Communities) (with the exception of Murcia, the Balearic Islands, Ceuta and Melilla). A total of 38 investigating physicians participated in the study. Each of them was requested to consecutively include at least 10 patients ≥90 years of age with a documented history of VD and/or who had been admitted due to a vascular complication (coronary and/or cerebrovascular and/or peripheral arterial disease), and who gave written or telephone (due to the COVID-19 pandemic) informed consent. In those cases where the clinical or cognitive condition of the patient did not allow for signing the informed consent, the latter was obtained from the primary caregiver. The patients were enrolled between May 2017 and March 2021. Data collection was performed through an online anonymized database including sociodemographic, clinical, laboratory, therapeutic and evolutive parameters.

Coronary disease was defined as stable angina, acute coronary syndrome, or coronary revascularization procedure. Cerebrovascular disease was classified as transient ischemic attack or stroke. Peripheral arterial disease was defined as intermittent claudication, revascularization, or amputation. The classification of paroxysmal, permanent or persistent atrial fibrillation (AF) was based on the 2020 European Society of Cardiology (ESC) guidelines for the diagnosis of AF.7 Thrombotic and bleeding risk was assessed using the CHA2DS2–VASc and HAS-BLED scales. A HAS-BLED score ≥3 points was considered to be indicative of a high risk of bleeding.8

Social, functional and cognitive performance was assessed upon admission by a face-to-face interview with the patient or primary caregiver. Comorbidity was assessed using the Charlson index,9 functionality was explored with the Barthel index,10 and cognitive impairment was assessed using the Pfeiffer scale.11 High comorbidity was defined as a Charlson score ≥3 points; severe or total dependence as a Barthel score <35, and advanced cognitive impairment was considered when the sum of errors on the Pfeiffer scale was ≥8 points. Treatments at admission and discharge were recorded, grouped by therapeutic categories. The study was approved by the Medicinal Product Research Ethics Committee (mpREC) of Hospital Universitario de La Princesa in 2017. The project was sponsored by the Spanish Foundation of Internal Medicine (Fundación Española de Medicina Interna) and the Spanish Society of Internal Medicine (Sociedad Española de Medicina Interna).

A total of 1049 patients were included in the study, with a predominance of females (57.8%), and with a mean age of 93.14 (±2.71) years. The clinical characteristics are described in Table 1. The prevalence of vascular risk factors was high: 84.9% presented arterial hypertension and 50.9% dyslipidemia. A total of 51.3% of the patients had a history of cerebrovascular disease, 42.2% ischemic heart disease and 16% multiple arteriosclerotic involvement (≥2 vascular territories).

In turn, a total of 41.6% had a prior diagnosis of AF on admission, mostly permanent AF (75.8%). A total of 134 patients (13%) had experienced previous hemorrhage. Gastrointestinal bleeding was the most common location of hemorrhage, and 56.7% of the patients required hospitalization.

Regular treatment on admission

At admission, the mean number of drugs prescribed was 8.25 (±3.57), with an average of 8.92 (±4.51) daily tablets. There were changes in prescription at discharge in 46.3% of the cases, with no changes in the mean number of tablets 8.95 (±4.14).

Of the total patient cohort, and despite the fact that all were in secondary prevention, only 82.9% were receiving some antithrombotic treatment. The antithrombotic therapeutic strategy at admission is shown in Fig. 1. A total of 47.2% received acetylsalicylic acid, 14.9% other antiplatelet agents, 15.5% anti-vitamin K, 11.5% direct-acting anticoagulants and 2.2% low molecular weight heparins.

Figure 1.

Antithrombotic therapy at admission in the patients included in the NONAVASC-2 registry.

(0.24MB).

Among the patients with AF (n = 430), only 56.9% received anticoagulation, 28.2% received antiplatelet medication, and 8.8% received no antithrombotic treatment.

The comparative analysis of treatments at admission and VD by vascular territories is shown in Table 3.

Table 3.

Correlation between the indicated treatments and vascular disease by territories, based on the chi-square test.

Treatments	General cohort	Ischemic heart disease	Cerebrovascular disease	Peripheral arterial disease	Multivascular	p
ASA	47.2%	56.8%	44%	30.4%	53.6%	0.00
Other antiplatelet drugs	14.9%	16.1%	13.5%	7.2%	28.3%	0.00
ACEIs	27.6%	28.4%	27.2%	24.6%	25.9%	0.89
ARA II	22.1%	19.5%	23.1%	21.7%	20.5%	0.72
Beta-blockers	35%	52.1%	22.8%	27.5%	42.8%	0.00
Calcium antagonists	25.9%	28.1%	23.6%	27.5%	31.3%	0.256
Statins	38.2%	52.4%	29.3%	29%	48.8%	0.00

Multivascular (grouped variable, involvement of ≥2 territories).

ASA: acetylsalicylic acid; ACEIs: angiotensin converting enzyme inhibitors; ARA II: angiotensin II receptor antagonists.

Among the general cohort, 35% received beta-blockers — this percentage increasing to 52.1% in patients with ischemic heart disease and to 42.8% in those with multivascular involvement. As regards lipid-lowering treatment, 38.2% of the sample received statins, with variation of the distribution depending on the affected vascular territory (52.4% in ischemic heart disease versus 29% in peripheral artery disease). The mean LDL-cholesterol concentration at admission was 92.8 mg/dl (±34.04). Only 21.2% of the patients presented LDL-cholesterol <55 mg/dl, while 38.9% had LDL-cholesterol <70 mg/dl.

Diuretic treatment was more often prescribed in the subgroup of patients with heart failure versus the general population: loop diuretics (79.7% vs 55.6%) and potassium-sparing agents (19.4% vs 10.5%).

Most diabetic patients received oral antidiabetic drugs (64.4%), and 29.4% received insulin. Metabolic control was good, with a mean glycosylated hemoglobin level of 7.0% (±1.2).

Benzodiazepine use was high (39.3%), in the same way as the use of antipsychotics (35.6%) and antidepressants (25.4%).

The NONAVASC-2 study has evaluated the characteristics of patients with established VD > 90 years of age admitted to Departments of Internal Medicine due to any reason.

Our data reveal a distribution similar to that of the national registry of Mostaza et al., which included 59,423 nonagenarians, of whom 14,302 had established VD.12 The gender distribution was as expected, since women have a longer life expectancy.13 In our registry, cerebrovascular disease was the most common VD, exceeding ischemic heart disease. This distribution differs from that seen in the general population and may be explained by the fact that it is the most prevalent condition in women and increases in frequency with advancing age.14

With regard to the baseline characteristics of our patients, it should be noted that most lived at home (80%), as observed in other series of nonagenarians admitted to Departments of Internal Medicine.12,15–17 Thirty-seven percent of the patients had some degree of cognitive impairment. This high prevalence is within the expected range, as reported in the literature,18 but is also greater than in other series.12,16,17 More than 50% of our patients had moderate or severe dependence, and three out of four had important comorbidity. These data are consistent with those of similar registries12,15,19,20 and corroborate the high morbidity associated with VD.

The prevalence of polypharmacy has increased in recent years among elderly people, being more common in those >80 years of age.21 Polypharmacy adversely affects elderly patients, leading to increased frailty, hospitalizations and even death.22,23 In our cohort, the nonagenarians presented polypharmacy, with a number of daily tablets of close to 9 on average. After hospital discharge, virtually no deprescribing was recorded, despite the fact that this strategy forms part of good clinical practice and is recommended in elderly patients, including those with VD.24

According to the clinical practice guides, all nonagenarians included in this registry should at least receive antiplatelet therapy, considering the secondary prevention context involved.25 However, 17.1% were not receiving any antithrombotic treatment. This proportion of patients is lower than or similar to that reported in other cohorts.12,26 High bleeding risk, falls and a history of previous bleeding are more common in elderly patients and constitute some of the limitations in prescribing antithrombotic treatment.27 In our sample, 69.4% had a high risk of bleeding according to the HAS-BLED scale, and 13% presented a history of bleeding events, which could explain why part of our patients were not receiving antithrombotic treatment.

The prevalence of AF upon admission was very significant (41.6%). Due to the intrinsic characteristics of our population, the thrombotic burden was very high, with a mean CHA2DS2–VASc score of 5.72. Within this subgroup, 8.8% of the patients had no antithrombotic treatment, and 28.8% were receiving antiplatelet medication. These data are similar to those of the ESFINGE registry,28 though they differ slightly from the results obtained in previous cohorts such as NONAVASC,29 where 13.8% were not receiving therapy and 17.8% were under antiplatelet treatment. In any case, these results show that antiplatelet therapy as an antithrombotic strategy in AF remains a common practice, probably conditioned by the false perception of being associated with a lower bleeding risk as compared to anticoagulants - despite the current scientific evidence refuting this idea.30

It should be noted that in the group of patients with AF not receiving oral anticoagulation or receiving antiplatelet medication alone, the percentage of severe dependence was significantly greater than among the anticoagulated patients. These data suggest that severe dependence could be one of the main determinants of the underprescription of oral anticoagulation in these patients.

Statins were part of treatment in only 38% of the patients, this being lower than reported in the literature.12,26 In our cohort, the group of patients without statins had significantly greater severe dependence and also greater severe cognitive impairment. Greater frailty among elderly patients and the fear of side effects are some of the factors that lead to not prescribing this therapeutic class.12 However, the use of statins in secondary prevention in the elderly has been shown to be safe and effective.31,32

In our cohort, patients with a history of ischemic heart disease were more likely to receive treatment consistent with the clinical practice guides.25 This is probably due to the greater perception of cardiac risk on the part of the professionals, as well as to increased evidence regarding treatments in elderly patients with vascular disorders of this kind. Paradoxically, the patients with more advanced VD, such as peripheral arterial disease, were the worst treated. This highlights the need for more studies and registries in the elderly population to improve the level of evidence on the efficacy and safety of the treatments we offer in this age group.

In-hospital mortality in our series was 19%, a high figure, but similar to that reported in other national and international series of nonagenarian patients.16,17,26,33 The main reported causes of mortality in nonagenarians are infectious diseases followed by VD, particularly terminal heart failure,16,17,34–36 in concordance with the findings in our own cohort.

The present study has a number of limitations. The findings described can only be compared with hospitalized elderly populations. The NONAVASC-2 registry is a cross-sectional study, which prevents us from establishing causal relationships; we therefore can only generate associative hypotheses. In addition, the recorded information was obtained from clinical interviews with the patients or relatives, along with data recorded in the case histories. Some degree of information bias therefore cannot be ruled out. On the other hand, it should be noted that part of the inclusion work took place during the COVID-19 pandemic, which caused difficulties in collecting certain variables related to dementia, frailty or nutritional status, impeding their inclusion in the analyses. Nevertheless, we believe that this study has important strengths, since it is a national multicenter project with a significant sample size, and has allowed us to respond to the proposed objectives.

Nonagenarians admitted to Internal Medicine with established VD are characterized by important comorbidity, dependence and mortality. Most of the patients in secondary prevention received some type of antithrombotic treatment. However, less than half received treatment with other drugs that have shown their vascular benefits (statins, RAAS inhibitors or beta-blockers). The underprescription of these treatments appears to be conditioned by the degree of severe dependence, and probably by other social and cognitive aspects, and may condition the prognosis of these patients. These results underscore the need for more studies to obtain better evidence on the advisable clinical management of an increasingly aging population.

The present study has no received specific funding from public, commercial or non-profit sectors.

There have been no conflicts of interest in relation to conduction of this study or writing of the manuscript.