Obesity prevalence and associated outcomes in cardiothoracic patients: a single-centre experience

The objective of this study was to investigate the prevalence of obesity and its relationship with adverse outcomes in ICU cardiothoracic patients. We performed a retrospective analysis of cardiothoracic patients admitted to The Royal Melbourne Hospital ICU between 2002 and 2014. Eight thousand and sixty-four patients who underwent coronary artery bypass, valve replacement/repair, or both, were divided into six categories of body mass index using World Health Organization criteria. Prevalence of obesity over time in the ICU was measured and compared to prevalence of obesity in the adult Australian population. The association between obesity and adverse postoperative outcomes was then analysed. Obesity is currently 1.2 times more prevalent in the Royal Melbourne Hospital ICU cardiothoracic patients than in the adult Australian population, with 33.5% of patients having a body mass index =30 kg/m(2). Over time, this was relatively constant, but an increasing proportion were morbidly obese. Obesity, but not morbid obesity, was associated with reduced 30-day mortality (odds ratio [OR] 0.41). Both obese and morbidly obese patients had reduced odds of return to theatre for bleeding (OR 0.49 and OR 0.19, respectively), but increased odds of new-onset renal failure (OR 1.62 and OR 3.17, respectively). Morbidly obese patients had double the odds of an ICU stay longer than 14 days (OR 2.05). In summary, a growing proportion of our obese ICU patients are morbidly obese, with a dramatically increased length of ICU stay. This has major implications for resource allocation in the ICU, and may inform modelling of future bed utilisation. Obesity, but not morbid obesity, conferred a mortality benefit.