There is compelling evidence that Covid-19 is strongly linked with obesity or being overweight, but behind that headline lie many unanswered questions.
Estimating obesity in individuals admitted as emergency cases is challenging given that treatment is a far higher priority than weight and height assessment. This has led most studies to use surrogate data for this critically important end point.
In one large UK inter-hospital study, for example, obesity was based solely on the opinion of the nursing staff, in another no data was given on how obesity was measured and in yet another, surrogate obesity data was available on only 50 per cent of subjects included in the final study.
In UK Biobank studies – a large-scale biomedical database and research resource enabling new scientific discoveries to improve public health – weight and height were directly measured but this data was collected in the period 2006-2010 and it is highly questionable if these measurements were applicable a decade and a half later. Accepting all these limitations, the evidence for a link between obesity and Covid is substantial.
A second key question is whether obesity increases the initial risk of infection. The UK Biobank data shows among people who are tested positive for Covid-19, significantly more are overweight (albeit based on 16-year-old data on weight/height) compared with those with a negative test.
But that raises the question of why people would seek a test. It may be because of a close contact with an infected person or it might be because they are experiencing symptoms. But what about 80 per cent of infected people whom the WHO say suffer only mild symptoms or are asymptomatic?
To really gain a handle on the likelihood of initial infection and obesity, we would need to start off with a cohort who, at the outset, are confirmed as Covid-free and who are followed up to identify new cases. Recently, a consortium of scientists from Harvard, MIT, the Broad Institute and Boston's leading hospitals followed 4,500 employees of a space exploration company across several sites in the US.
Over a three-month period, 322 confirmed new cases were detected. Among these there was no evidence that obesity was related to the risk of catching Covid. The sample size is small, necessarily so with generally low infection rates and it is not representative of US demographic patterns. But it does raise the question of whether the immune system of the obese is such that they are more likely to be infected.
However, the study then showed that once infected, obesity dramatically heightened the severity of Covid complications. A similar finding is seen for many infectious diseases, including the seasonal flu, and the past H1N1 swine flu, where obesity greatly increases the severity of symptoms and risk of death.
Inhibited cells
Among the first line of immune defence mechanisms against viruses and tumours are two types of immune cells – natural killer (NK) and T cells – and their functionality is greatly diminished in obesity. Obesity is also associated with chronic inflammation, which means the immune system is always “activated” with elevated levels of cytokines, even in the absence of invading pathogens such as viruses or bacteria.
This apparent paradox of heightened cytokine-led inflammation and suppressed NK and T cell activity is seen in other conditions such as sepsis and ageing. This constant inflammation exhausts the immune system, so when a virus or bacteria comes along, immune cells are unable to respond quickly and effectively, allowing the virus to replicate vigorously in the lungs and other organs.
Covid has been associated with a “cytokine storm”, a sudden rapid inflammation where immune cells produce so many cytokines that they attack the body’s own cells, creating a form of acute auto-immune disease. Thus, while we know the immune system is critical for protecting us from Covid, and the immune system is dysfunctional in obesity, a key question that remains is: do patients with obesity have an underactive or overactive response to Covid?
Both scenarios could cause increased severity in symptoms, but treatment options would be very different depending on which scenario holds true. This is now an area funded by the Science Foundation of Ireland in the Lynch immunology lab at Trinity College Dublin.
The link between obesity and the risk of Covid complications is confounded by age and ethnicity, much as is the diabetes-obesity link. The risk of developing type-2 diabetes rises dramatically with increasing body weight but particularly so among younger age groups.
The same pattern is seen with the obesity-Covid link. The UK Biobank studies show there is only a minor rise in hospitalisation rates from normal weight to morbid obesity. However, among those who are hospitalised, the risk of death soars in those under 70. With morbid obesity (BMI>35), the risk of death was 5 times higher in those below 70 years of age compared with those above 70. Of course, age itself is a big risk factor for Covid complications, partly due to an altered immune function and partly to underlying medical conditions.
Irish risk
Ethnicity also muddies the water. Asians have the highest level of sensitivity to weight-induced diabetes, followed by those of African or Caribbean ethnicity, with white people well below both these groups. Again, UK Biobank studies show a similar pattern to this with weight-induced Covid complications.
Interestingly these UK Biobank studies break whites into two groups: “Irish whites” and “other whites”. And it appears there was a 40 per cent higher risk of developing Covid among Irish white people compared with other white people. Is it due to an older Irish generation who emigrated to the UK for manual work or is it the high representation of Irish healthcare staff in the UK NHS? The data simply cannot yet explain this.
The relationship between ethnicity and obesity-related Covid complications is confounded and complicated by socio-economic factors. Epidemiologists statistically control for many standard confounding factors such as income and educational attainment. But there are many other factors that might predispose to obesity and thus to Covid complications, which are rarely considered: emotional insecurity, major life events, financial and employment uncertainty and many other social factors.
And then there are the biological factors such as suboptimal maternal nutrition during the first 1,000 days covering conception to a child’s second birthday leading to heightened risk of several diseases including obesity and diabetes later in adult life.
There is a wealth of unanswered questions in this obesity-induced Covid conundrum, but the bottom line remains: obesity is a huge public health problem that in time will cause far more morbidity and mortality than Covid. But then, there are votes in Covid; not so in obesity.
Mike Gibney is emeritus professor of food and health at University College Dublin and Lydia Lynch is an associate professor in immunology at Trinity College Dublin