Temperature measurement during cardiac surgery To the Editor: We read with interest the paper by Robinson et al. that addresses the best site for temperature monitoring during cardiac surgery) Although the authors succeeded in determining how various temperature sites compare with the still debatable 'gold standard' pulmonary artery (PA) site and concluded that oesophageal temperature is most accurate, this does not necessarily mean that it is the most appropriate temperature monitoring site during cardiac surgery. The true best site is determined by the balance of its precision and accuracy with its clinical utility. That is, does it accurately reflect the temperature in the organ system(s) of interest. For example, if one is interested in whole-body cooling and heating, then PA or oesophageal sites may well be appropriate. However, if one is interested in the cerebral thermal environment, particularly during rapid temperature fluxes (as during CPB), then an alternative site (tympanic or nasopharyngeal) may be better. Also, the authors mention their lack ofnasopharyngeal monitoring due to the risk ofepistaxis during systemic heparinization. We accept their concern but have used (and advocate) this site with no reported complications in many thousands of cases. Finally, how one uses the temperature information to guide clinical management is equally, arguably more, important than the measurement itself. For example, the patient outlined in their Figure 1 had temperatures > 40~ during rewarming. Strategies such as slower rewarming, limiting inflow temperature, and accepting lower maximum temperatures seem prudent considerhag the indisputable data implicating hyperthermia in propagating cerebral ischemic injury. 2