Body mass index as a biomarker for the evaluation of the "Obesity Paradox" among inpatients.

Background: Overweight and obesity are, on the one hand, recognized as risk factors for many health-related disorders, and, on the other, as favorable prognostic factors in various patients treated for several different conditions; what is called the "obesity paradox". Until now, the existence of this phenomenon among a general population of consecutive inpatients has not been evaluated. We decided, therefore, to perform an evaluation.

Patients And Methods: Historical prospective analysis of the medical documentation of 23 603 hospitalizations during two consecutive years in one center was performed. The outcomes measured were as follows: length of stay, in-hospital all-cause mortality, and non-scheduled readmission in the 14-day, 30-day and one-year periods following discharge.

Results: Overweight and obese patients had a lower or similar prevalence of the measured outcomes than malnourished patients and those of normal weight. Adjustment of the standard WHO BMI ranges for patients aged ≥65 y (normal weight BMI range 23-33 kg/m) made these differences more apparent. In logistic regression, the ratio of fat to fat-free body mass was a stronger and unfavorable risk factor compared with BMI for the measured outcomes.

Conclusions: The greatest risk of all-cause in-hospital death and readmission concerned malnourished inpatients. Compared to patients with a normal BMI range, overweight and obesity had a lower or similar (but not greater) risk of the outcomes measured. However, due to several BMI limitations, our observations should be interpreted as suggesting a "BMI paradox", rather than an "obesity paradox".