Environmental Factors in Northern Italy and Sickle Cell Disease Acute Complications: A Multicentric Study

Abstract

Background: Environmental factors seem to influence clinical manifestations of sickle cell disease (SCD), but few studies have shown consistent findings. We conducted a retrospective multicentric observational study to investigate the influence of environmental parameters on hospitalization for vaso-occlusive crises (VOC) or acute chest syndrome (ACS) in children with SCD.

Methods: Hospital admissions were correlated with daily meteorological and air-quality data obtained from Environmental Regional Agencies in the period 2011-2015. The effect of different parameters was assessed on the day preceding the crisis up to ten days before. Statistical analysis was performed using a quasi-likelihood Poisson regression in a generalized linear model.

Results: The risk of hospitalization was increased for low maximum temperature, low minimum relative humidity, and low atmospheric pressure and weakly for mean wind speed. The diurnal temperature range and temperature difference between two consecutive days were determined to be important causes of hospitalization. For air quality parameters, we found a correlation only for high levels of ozone and for low values at the tail corresponding to the lowest concentration of this pollutant.

Conclusions: Temperature, atmospheric pressure, humidity and ozone levels influence acute complications of SCD. Patients' education and the knowledge of the modes of actions of these factors could reduce hospitalizations.

Keywords: acute chest syndrome; air quality; environmental factors; meteorological parameters; sickle cell disease; vase-occlusive crises.

Figure 1

Relative risk of hospital admission in the lowest (“RR5vs10”, in blue) and highest tail of the variable distribution (“RR95vs90”, in red) for the temperature difference with respect to the previous day (D2D).

Figure 2

Relative risk of hospital admission in the lowest (“RR5vs10”, in blue) and highest tail of the variable distribution (“RR95vs90”, in red) for average atmospheric pressure.