Inflammation and aging

Inflammation is the immune system's response to infection and injury. In its acute form, immune cells gather at the affected site and release signalling proteins called cytokines, such as interleukin-6 (IL-6) and tumour necrosis factor alpha (TNF-alpha), that recruit further immune activity. The response clears microbes and damaged tissue and then subsides as healing begins. Acute inflammation is essential for survival.

Inflammaging is a different pattern: a chronic, low-grade rise in inflammatory activity that develops with age [1]. Franceschi and colleagues proposed in 2000 that it is driven by a lifetime of continuous exposure to antigens (substances that trigger an immune response) and other stressors. They argued that inflammatory responses which help to neutralise harmful agents early in life become harmful late in life, a period largely not shaped by evolution, consistent with the antagonistic pleiotropy theory of aging (the idea that genes helpful in youth can be harmful in old age) [3].

The same paper noted a paradox: healthy centenarians also show raised plasma levels of inflammatory cytokines, acute-phase proteins and clotting factors. The authors proposed that inflammation acts as a "first hit" that raises susceptibility to age-related disease, with overt disease such as atherosclerosis, Alzheimer's disease, osteoporosis or diabetes requiring a second, genetic hit [3].

Franceschi's 2000 paper introduced the term "inflamm-aging" as an extension of his network theory of aging, describing aging as a global reduction in the capacity to cope with stressors combined with a progressive increase in pro-inflammatory status [3]. A 2014 report from a US National Institute on Aging meeting, co-written by Franceschi and Judith Campisi, described inflammaging as a highly significant risk factor for illness and death in older people, noting that most if not all age-related diseases share an inflammatory origin, while its precise cause and causal role remained largely unknown [1].

In the 2013 hallmarks of aging framework, inflammation was not one of the nine named hallmarks [10]. The 2023 update listed chronic inflammation as a hallmark in its own right, one of twelve, alongside dysbiosis (disturbed gut microbial communities) and disabled macroautophagy (reduced recycling of cell components) [5].

A 2018 review in Nature Reviews Cardiology lists the proposed sources of inflammaging as genetic susceptibility, central obesity, increased gut permeability, changes in the gut microbiota, cellular senescence, activation of the NLRP3 inflammasome, oxidative stress from dysfunctional mitochondria, immune cell dysregulation and chronic infections [2]. Senescent cells, which accumulate with age partly as a result of telomere shortening, release inflammatory molecules into surrounding tissue; removing them in prematurely aging mice delayed several age-related disorders [11].

Several everyday factors are associated with inflammatory marker levels in people. Disturbed sleep is linked to higher CRP and IL-6 [12], and regular exercise has anti-inflammatory effects, thought to act partly by reducing visceral (abdominal) fat and partly through changes that follow each exercise session [13]. Most of these findings come from observational studies and do not show cause and effect.

Long-standing inflammatory diseases show the long-term effects of sustained inflammation. A meta-analysis of 11 studies of 51,819 people with early rheumatoid arthritis, an autoimmune disease of the joints, whose follow-up began between 1955 and 1995 found that mortality was 1.47 times that of the general population of the same age (95% CI 1.19 to 1.83) [17]. In a population study in British Columbia, people diagnosed with rheumatoid arthritis in 1996 to 2000 had higher five-year mortality than matched members of the general population (hazard ratio 1.40, 95% CI 1.30 to 1.51), while those diagnosed in 2001 to 2006 did not (0.97, 0.89 to 1.05); the study was designed to test whether improved treatment had reduced this excess [18].

Physical activity and diet change inflammatory markers in trials. A meta-analysis of 83 randomised and non-randomised controlled trials with 3,769 participants found that exercise training lowered CRP (effect size 0.26, 95% CI 0.18 to 0.34); the fall was larger when body mass index also fell, but remained significant without weight loss (effect size 0.19) [19]. In a substudy of the PREDIMED trial, 772 adults aged 55 to 80 at high cardiovascular risk were randomly assigned to one of two Mediterranean diets or a low-fat diet; after three months the Mediterranean diet with added virgin olive oil had lowered CRP by 0.54 mg/L more than the low-fat diet [20]. In a 17-week randomised study of healthy adults (18 per group), a diet high in fermented foods increased the diversity of gut microbes and lowered 19 of 93 inflammatory proteins measured in the blood, including IL-6, while a high-fibre diet did not change the main immune outcome [21].

Smoking, psychological stress and gum disease are other proposed sources. In the Multi-Ethnic Study of Atherosclerosis, 6,814 adults without cardiovascular disease were followed for a median 10.2 years; current smokers had 1.7 times the rate of cardiovascular disease of never smokers, and among current smokers those with hsCRP of 3 mg/L or more had 2.6 times the rate of coronary heart disease of those below 2 mg/L. Former smokers, a median of 22 years after quitting, had a risk similar to never smokers [22]. A review of psychological stress and cardiovascular disease describes pro-inflammatory and pro-clotting responses among the changes that allow stress to trigger events in people who already have extensive arterial plaque [23]. Periodontal disease, a chronic infection and inflammation of the gums, was associated with 1.24 to 1.34 times the risk of coronary heart disease in a meta-analysis of seven studies, independent of standard risk factors [24].

Full deep dive: inflammation.

Circulating inflammatory markers form a cascade. Interleukin-1 beta (IL-1β), produced when the NLRP3 inflammasome is activated, drives production of IL-6, which in turn stimulates the liver to make CRP. In atherosclerosis the NLRP3 inflammasome is activated by cholesterol crystals, neutrophil extracellular traps, disturbed blood flow and local tissue hypoxia. On this model high-sensitivity CRP (hsCRP) is a downstream surrogate for upstream IL-1β activity, and most mechanistic studies suggest CRP itself is unlikely to be a useful drug target [25].

Ferrucci and Fabbri propose that chronic inflammation contributes to cardiovascular disease, multimorbidity and frailty by inhibiting growth factors, increasing catabolism (the breakdown of tissue, including muscle) and interfering with homeostatic signalling. They note that this hypothesis is supported by mechanistic studies but still requires confirmation in humans [2]. Inflammaging is also a risk factor for chronic kidney disease, diabetes, cancer, depression, dementia and sarcopenia (age-related loss of muscle) [2].

Cellular senescence links inflammation to other aging processes. Senescent cells accumulate in tissues with age and are thought to disrupt tissue function through the molecules they secrete, a pattern known as the senescence-associated secretory phenotype; in the INK-ATTAC mouse model, lifelong removal of p16Ink4a-positive senescent cells delayed age-related changes in fat tissue, skeletal muscle and the eye, and late-life removal slowed disorders already established [11].

CRP and IL-6 are the most commonly measured markers of systemic inflammation. The largest analysis of CRP and long-term risk, by the Emerging Risk Factors Collaboration, combined individual records of 160,309 people without known vascular disease from 54 prospective studies, covering 1.31 million person-years and 27,769 fatal or non-fatal disease outcomes [4]. The analysis adjusted for within-person variation in repeated measurements.

CRP was associated roughly log-linearly with ischaemic vascular disease and non-vascular mortality. Per threefold higher CRP, after adjustment for conventional risk factors, the risk ratio (the risk in one group divided by the risk in another) was 1.37 for coronary heart disease, 1.27 for ischaemic stroke, 1.55 for vascular death and 1.54 for non-vascular death. Adjusting further for fibrinogen, another inflammatory protein, reduced these to between 1.23 and 1.34 [4]. The authors noted that CRP was associated with a wide range of disorders, including death from several cancers and lung disease, and that the relevance of CRP to such a range of conditions is unclear [4].

In cardiovascular prediction, hsCRP predicts risk with an effect estimate as large as that of total or HDL cholesterol [25]. CRP is non-specific, however: it also rises with acute infections and injuries, and a single raised reading does not by itself indicate chronic inflammation.

The CANTOS trial tested whether blocking inflammation alone could prevent cardiovascular events. It randomised 10,061 people with a previous heart attack and hsCRP of at least 2 mg/L to placebo or canakinumab, an antibody against IL-1β, at 50, 150 or 300 mg every three months [6]. Canakinumab lowered CRP without lowering cholesterol. Over a median 3.7 years the 150 mg dose reduced the combined rate of heart attack, stroke and cardiovascular death, with a hazard ratio (the event rate in the treated group relative to placebo; 0.85 means 15% lower) of 0.85 (95% CI 0.74 to 0.98) [6]. Fatal infections were more common with canakinumab, and all-cause mortality did not differ (hazard ratio 0.94, 95% CI 0.83 to 1.06) [6].

Two trials of low-dose colchicine, an older anti-inflammatory drug used for gout, supported the result. In COLCOT, 4,745 patients treated within 30 days of a heart attack took 0.5 mg daily or placebo for a median 22.6 months; the primary cardiovascular outcome occurred in 5.5% versus 7.1% (hazard ratio 0.77, 95% CI 0.61 to 0.96), and pneumonia was reported more often with colchicine (0.9% versus 0.4%) [8]. In LoDoCo2, 5,522 patients with chronic coronary disease were followed for a median 28.6 months; events occurred in 6.8% versus 9.6% (hazard ratio 0.69, 95% CI 0.57 to 0.83), while deaths from non-cardiovascular causes were numerically higher with colchicine (hazard ratio 1.51, 95% CI 0.99 to 2.31) [7].

Not every anti-inflammatory approach has worked. In the CIRT trial, 4,786 patients with coronary disease and type 2 diabetes or metabolic syndrome received low-dose methotrexate or placebo. The drug did not lower IL-1β, IL-6 or CRP, did not reduce cardiovascular events (hazard ratio 0.96, 95% CI 0.79 to 1.16), and was associated with raised liver enzymes and more non-basal-cell skin cancers; the trial was stopped after a median 2.3 years [9].

The trial evidence comes from people with established coronary disease and measures cardiovascular events, not aging as a whole. Ferrucci and Fabbri describe the link between inflammaging and cardiovascular disease as probably causal, but consider it controversial whether reducing inflammation benefits the course of non-cardiovascular problems such as kidney disease, diabetes, cancer, depression, dementia and sarcopenia [2]. Whether early treatment of inflammaging prevents or delays frailty has not been tested in clinical trials [2].

The drugs that lowered cardiovascular events carried infection signals: canakinumab increased fatal infections and did not lower all-cause mortality in CANTOS [6], and pneumonia was reported more often with colchicine in COLCOT [8]. The null CIRT result shows that the benefit depends on the pathway targeted rather than on anti-inflammatory action in general [9]. The origins of inflammaging and its causal role in age-related disease remain largely unknown [1], and in the hallmarks framework chronic inflammation is one of several interconnected processes rather than a single cause of aging [5].