Background
In December 2019, cases of pneumonia with unknown origin pathogen, now known as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), emerged in Wuhan, the capital city of Hubei Province in China [
1]. The clinical manifestations of this novel disease resembled viral pneumonia [
2]. The disease has rapidly spread from Wuhan to other areas in China and the World Health Organization has declared the outbreak of COVID-19 a pandemic on 11 March 2020 [
3,
4]. As of 12 May 2020, over 4.0 million of confirmed COVID-19 cases and over 270 thousand deaths have been reported in over 200 countries [
5].
Data from China have indicated that old age and underlying illnesses are two strong risk factors for illness and death related to COVID-19 [
6‐
8]. Although the majority of reported COVID-19 cases were mild and the overall case-fatality rate was only 2.3%, over 80% of deaths occurred among adults aged ≥ 60 years, and the case-fatality rate increased dramatically from 3.6% among persons aged 60–69 years to 14.8% among those aged ≥ 80 years [
9]. COVID-19 has also posed a disproportionately high threat to older adults in other parts of the world, including the USA and Europe [
10,
11]. The COVID-19 pandemic has placed an unprecedented burden on health systems and forced health care professionals to make difficult decisions about how to allocate the increasingly scarce resources efficiently [
12,
13]. In some places, decisions are being made about who should be prioritized for medical resources, such as ventilators and ICU beds, based on chronological age [
14,
15]. However, chronological age may not truly reflect the differences underlying the biological ageing process and, therefore, is not an ideal basis for efficient resource allocation and establishing care plans for older COVID-19 patients. Recognizing frailty could help in early warning of older patients at high-risk with severe COVID-19 pneumonia.
Frailty is an age-related clinical syndrome of decreased reserve to stressors and is strongly associated with a wide range of adverse outcomes including death, disability, and hospitalization [
16,
17]. Frailty affects over 10% of older adults in the world and its prevalence is higher at advanced age [
18]. The concept of frailty and its assessment has been gradually integrated into clinical practice for evaluating prognosis and establishing goals of treatment [
19]. In this sense, provision of frailty screening to older COVID-19 patients may help identify high-risk group for poor prognosis and formulate patient-tailored treatment goals. Thus, we conducted a prospective cohort study to examine whether frailty status at baseline, assessed by a clinically friendly frailty assessment, would increase the risk of development of severe disease among COVID-19 patients aged ≥ 60 years.
Discussion
We conducted a prospective cohort study of 114 older COVID-19 patients to examine the association between frailty and severe disease. We found that frailty, assessed by a self-reported frailty screening tool, was an independent risk factor for severe disease among older COVID-19 patients. These findings suggest that the FRAIL scale can be easily applied in a busy clinic setting to identify potentially severe pneumonia patients early and allows us to optimize treatment strategy in advance for older COVID-19 patients.
Since the epidemic of COVID-19, many studies have showed that older age was associated with ARDS after being infected [
6,
7,
24]. According to a recent meta-analysis of community-dwelling older adults in Europe, the prevalence of physical frailty is around 15% for adults aged 65 years and over [
25], and it increases to over one quarter among those over 85 years [
18]. It would appear that some elderly patients still have a good prognosis, so it seems to be inappropriate to assess the prognosis and make a medical decision simply based on age. Some recent studies suggest that frailty provides better risk stratification for post-operative complications than chronological age [
26,
27]. Therefore, it has been suggested that screening of frailty may be suitable for detecting older persons at increased risk of adverse outcomes [
28‐
30]. De Silva et al. reported a positive association between frailty and mortality among nursing home residents [
31]. The largest study of frailty in critical illness showed that frail adults were twice as likely to die in the hospital and within 1 year than the non-frail [
32,
33]. Frailty is not only independently associated with hospitalizations and mortality in adults with pneumonia but also appears to have a synergistic effect on respiratory function along with lung disease [
34,
35] Previous studies have found that age, cardiovascular disease, and cerebrovascular disease were predictive of fatal outcomes [
24]. In our study, comorbidities were not associated with higher risk of severe disease in either univariate or multivariate analyses including frailty. These results suggest that frailty may better predict poor prognosis of the older patients with COVID-19 than comorbidities. Thus, we should conduct early frailty assessment, strict medical supervision, and optimal treatment for frail older patients with COVID-19, so as to improve their prognosis. Besides, it is essential to note that there is much potential for frailty to be reversed, particularly in its early stages, such as pre-frailty [
36‐
38]. Therefore, even among the pre-frail or early frail patients with COVID-19, it is possible to improve their prognosis with early and timely management. For this reason, frailty screening should be used to detect old persons with COVID-19 for risk stratification and management guidance.
The mechanisms underlying frailty are multiple and reflect the complexity of the ageing process [
35]. Theou et al. suggested that every additional year of age was associated with a 3.5 and 2.8% higher mean frailty index in lower- and higher-income countries, respectively [
39]. A low-grade, persistent chronic inflammation, which is so-called inflamm-ageing, is one of the primary reason and pathobiological changes for both ageing and age-related diseases, such as frailty [
35,
40]. Pro-inflammatory cytokines, including IL-6, IL-1, tumour necrosis factor (TNF)-a, C-reactive protein, and fibrinogen levels, may directly accelerate frailty by promoting muscle protein degradation [
41], and the pro-inflammatory state may lead to overall suppression of the inflammatory response that is needed to fight an acute respiratory infection [
42], including acute virus pneumonia such as COVID-19. However, frailty is a process with persistent chronic inflammation, whereas severe cases of COVID-19 are characterized by an inflammatory storm. Therefore, the immunologic and inflammatory mechanisms involved in the course of severe COVID-19 cases, which might be different from those involved in frailty, deserve further study. Besides, inflamm-ageing suggests a failure of the cell clearance mechanisms which could aid in the resolution of inflammation after tissue injury or/and pathogen infiltration by indirectly affecting important metabolic signalling pathways [
43], which could result in the development and progression of ARDS in COVID-19 [
44]. We also found that the CD8+ counts in frail COVID-19 patients were significantly lower than that in non-frail COVID-19 patients. This is consistent with the report by Liu et al., showing that the counts of lymphocyte subset (CD4+ and CD8+ T cell) are proportionally associated with disease severity [
45]. Cell-mediated immune responses play an important role in virus clearance. CD4+ and CD8+ T cells are required for virus clearance during primary infection in the mucosal tissues [
46]. CD8+ T cells are cytotoxic and can kill virally infected cells. Immunosenescence has been suggested to contribute to frailty and characterized by the progressive decline in both the innate and adaptive immune systems, and vice versa [
42]. Viral infections usually lead to abnormal changes in the levels of lymphocyte subsets which further impaired immune system functionality [
47,
48]. Therefore, frailty-related decline in immune function may explain the association between ageing and increased risk of adverse outcomes.
Our analyses were restricted to older patients with COVID-19 pneumonia, which may introduce collider bias that would undermine the validity of the observed association between frailty and progression to severe pneumonia. Selecting a sample of older patients with COVID-19 pneumonia could induce collider bias in the estimated effect of frailty on progression to severe pneumonia in the absence of adequate control for the unmeasured confounders [
49]. We adjusted for several confounders including health characteristics that were associated with progression to severe pneumonia to minimize this bias. Also, simulation studies consistently showed the collider bias to only to have minimal impact on the exposure-outcome association unless the associations between some unmeasured confounders and infection of COVID-19 pneumonia were extremely large [
50,
51]. Moreover, collider bias is generally thought to lead to unexpected associations, such as the paradoxical relationship between higher body mass index and better outcomes among heart failure patients [
52]. We found that frail patients were more likely to experience severe disease than the non-frail; these results were consistent with the associations between frailty and poor prognostic outcomes among other populations. Therefore, the influence of collider bias on the estimate of the exposure-outcome association in the present study is not likely to be large.
Strengths of our study include that we focus on the population with COVID-19 over 60 years of age—a population that carries a disproportionally high disease burden than younger populations. Our results showed that frailty might be a risk factor for the development of severe disease in older COVID-19 patients. The FRAIL scale can be easily performed in a busy clinic setting and it will not increase the risk of the exposure of the healthcare provider to severe infectious diseases.
Our study has some limitations. First, this study was conducted at a single-centre hospital with limited sample size. A larger, multicentre cohort study of older patients with COVID-19 pneumonia would help further identify the association between frailty and prognosis. Second, all clinical and laboratory characteristics were measured once on admission. It is, therefore, challenging to distinguish confounders from mediators, which may lead to over-adjustment. Future research with repeated measures could advance our understanding about mechanisms underlying the association between frailty and poor prognosis among older COVID-19 patients. Third, the association between frailty and death was not examined in the present study. There were only eight deaths in our sample, making it difficult to study death as an outcome in a rigorous manner. Because deaths met the diagnostic criteria for then severe illness of COVID-19 and we considered death cases as “severe illness” in analyses. In the future, studies with larger sample sizes may help elucidate the relationship between frailty and mortality among COVID-19 patients. Lastly, we did not collect data on treatment options because treatment varied widely from patient to patient due to comorbidities except for antiviral regimen.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.