Exp Clin Endocrinol Diabetes 2006; 114 - P37
DOI: 10.1055/s-2006-954730

Comparison of salivary cortisol concentrations using two different saliva collection methods

E Poll 1, E Yagmur 2, Y Langejuergen 1, JM Gilsbach 1, I Kreitschmann-Andermahr 1
  • 1Department of Neurosurgery, University Hospital, RWTH Aachen University, Germany
  • 2Institute of Clinical Chemistry and Pathobiochemistry, University Hospital, RWTH Aachen University, Germany

Cortisol measurement in saliva is an important method in stress research and has some advantages over serum cortisol measurement. Often-used methods for saliva sampling are the Salivette® sampling device (Sarstedt) and plastic tubes. Recently it has been suggested, that the use of a cotton roll such in the Salivette® reduces the amount of measured cortisol. In order to establish the reliability of the two different sampling methods we compared salivary cortisol sampled by Salivettes® and plastic tubes (polypropylene) to serum cortisol levels. Ten healthy volunteers provided saliva samples taken by both methods as well as blood samples at 8, 11, 14, 17, 20 and 23h on two consecutive days. After calculating mean cortisol profiles for each person and sampling device, regression analyses were performed separately with both methods as predictors for serum cortisol at each of the six time points. The determination coefficients for the regression equations including cortisol in Salivettes® were higher than for the ones with cortisol in plastic tubes for each point of time (ranging from R2=.27 to .56 and R2=.01 to .46, respectively). Spearman rank correlation coefficients ranged from .40 to .65 for serum and Salivette, reaching statistical significance in three of six cases, and from .07 to .60 for serum and plastic tube (n.s.). Despite the advantage of cortisol in Salivettes® over cortisol in plastic tubes for predicting serum cortisol, the total amount of salivary cortisol sampled with Salivettes® was lower than in plastic tubes at all points of time (n.s.). We conclude that different forms of saliva sampling deliver different cortisol values with cortisol in Salivettes® being slightly lower than in plastic tubes, but providing a better prognosis of serum cortisol. Because the two sampling methods seem to be slightly different, devices should not be switched during a study period and because of the specific pros and cons the device should be chosen with regard to study characteristics.