Differences in mortality on the basis of laboratory parameters in an unselected population at the Emergency Department

Clin Chem Lab Med. 2005;43(5):536-41. doi: 10.1515/CCLM.2005.093.

Abstract

The objective of this study was to relate multiple laboratory results of initial blood gas analysis in patients visiting the emergency department to mortality. A total of 1806 adult patients visiting the emergency department for any reason, who underwent blood gas analysis, were included in this study. The results of nine laboratory parameters (ionised calcium, potassium, sodium, glucose, lactate, pH, pCO(2), pO(2) and saturation) were related to mortality. Comparison of the means of survivors' and non-survivors' age, pH, p(O)2 , glucose and lactate showed relevant differences (p < or = 0.001). When dividing the group of patients into quintiles, hypo- and hyperkalaemia (<3.6 or >4.3 mmol/L), hypocalcaemia (<1.12 mmol/L), hyperglycaemia (> 10.6 mmol/L), hyperlactataemia (>2.9 mmol/L) and acidosis (pH < 7.34) were all related to high mortality in comparison to the quintile with the lowest mortality (p < 0.0001). Interestingly, even within the reference range for lactate and ionised calcium, there was an increased risk of mortality, so these reference ranges should be interpreted with care. Glucose and lactate have additional clinical value, but are not always requested. Therefore, glucose and lactate should be routinely added to a blood gas request. Our study shows differences in mortality for laboratory parameters in patients visiting the emergency department, irrespective of the underlying pathology.

MeSH terms

  • Aged
  • Blood Chemical Analysis / statistics & numerical data
  • Blood Gas Analysis* / statistics & numerical data
  • Emergency Service, Hospital*
  • Female
  • Gases / blood*
  • Hospitals, University
  • Humans
  • Male
  • Middle Aged
  • Mortality*
  • Netherlands
  • Prospective Studies

Substances

  • Gases