Coronavirus and Obesity: Could Insulin Resistance Mediate the Severity of Covid-19 Infection?
Francis Finucane, Colin Davenport
Abstract
The rapidly evolving global epidemic of coronavirus disease 2019 (COVID-19) (1), caused by thenovel severe acute respiratory syndrome corona virus 2 (SARS-CoV-2), presents with variableclinical severity, being fatal in some and asymptomatic in others (2). Preliminary indications fromthe UK (3), China (4), and the USA (5) suggest that patients with obesity, at least in hospital,have a worse prognosis. This is consistent with long-established observations that patients withacute respiratory distress syndrome (ARDS) who have obesity do worse, for several reasons (6).Obesity causes atelectasis, particularly in the posterior dependant lung zones (7) and this, alongwith collapse of alveolar capillary units because of raised pleural pressures (8) leads to diminishedrecruitability of lung tissue. Also, parenchymal heterogeneity leads to high lung shear forces, evenwhen applied ventilatory pressures are low and without well-established lung injury (9). This isconsistent with population-based observations that obesity adversely affects lung function (10).However, it is becoming increasingly apparent that there may bemore specific mechanismsby which obesity worsens the outcome of COVID-19, arising from metabolic and inflammatoryderangements rather than merely the mechanical effects of increased fat tissue. For example,categorizing overweight and obesity in the recent audit of intensive care patients with Covid-19(who had a mean age of 60 years) (3), their prevalence was very similar to that in the backgroundpopulation of older British adults (11). In a case series of 112 Chinese adults with prevalentcardiovascular disease and Covid-19, 88% of those who died vs. 19% of those who survived werecategorized as overweight or obese, but only 16 patients in this small study received intensive careand the mean body mass index (BMI) was 22 vs. 25.5 kgm−2for normal weight and overweightpatients, respectively (4). Thus, there may have been substantial differences in the proportions ofpatients who died in the normal weight vs. overweight categories, without there being a strong orclinically relevant association between BMI and mortality,especially when relevant confoundersare not taken into consideration. A larger French study showed that COVID-19 patients withsevere obesity in intensive care were seven times more likelyto require invasive mechanicalventilation compared to Covid-19 intensive care patients with a normal BMI, but this trend wasnot statistically significant for those with BMI<35 kg m−2(12). This is similar to observationsfrom large prospective cohort studies of hospitalized patientsduring the 2009 Influenza (H1N1)pandemic, which showed that severe obesity was associated with a much higher risk of intensivecare admission and mortality, but overweight and obesity were not (13). Despite an exhaustiveliterature search, we were unable to identify any studies that found BMI (as a continuous variable)to be a good predictor of outcomes in patients with SARS or COVID-19. \nConversely, several of the features of the so-called metabolic syndrome seem to be associatedwith a worse prognosis in patients with COVID-19. Males seem tobe worse affected than females,with a male preponderance in several studies (3,14–16). Older Covid-19 patients were foundto have a worse prognosis in China (2,16), Italy (17) and the UK (3). As was seen with SARSin 2003 (18), diabetes and dysglycaemia have been found to be highly prevalent in Covid-19 patients (16,19). They often have a transaminitis (16,19), usuallyattributed to shock but which might indicate underlying fattyliver disease. Hypertension is associated with worse outcomes(17,20). These early clinical observations that patients withsevere Covid-19 tend to be older, male, hypertensive, withelevated blood glucose levels and abnormal liver blood tests raisethe prospect that insulin resistance could play an important rolein mediating disease severity. Is there a plausible mechanistictheory for such an association?