Objective To evaluate whether measurement of the hemoglobin (Hb) concentration with a blood gas analyzer approximates that determined by a conventional coulter counter in critically ill adults.

      Design
      Prospective patient series.


      Setting
      A 32-bed cardiovascular, neurosurgical, trauma and medical-surgical intensive care unit in a single Canadian center.


      Patients
      We consecutively recruited 202 critically ill adults, the majority of whom had recent cardiac or vascular surgery, neurosurgery or trauma.


      Measurements
      The nurse obtained a single arterial blood sample within a few hours of the patient's admission to the intensive care unit. The Hb concentration was determined from each blood sample in a masked fashion, using both a blood gas analyzer and a conventional laboratory coulter counter.


      Main results
      A total of 202 consecutive paired analyses were conducted. There was a highly significant correlation between the coulter counter and blood gas analyzer methods of Hb measurement (r2 = 0.98, 95% confidence interval [CI] = 0.97–0.99; P < 0.0001). Using the method of Bland and Altman, the overall mean difference in Hb concentration between the coulter counter and the blood gas analyzer was -4.3 g/l (95% CI = -11.0 to 2.4). Of the 11 (5.4%) Hb measurements that extended beyond the upper and lower 95% CI, 10 (5.0%) were within ± 3 g/l of these confidence limits.


      Conclusions
      An arterial blood gas analyzer may provide a valid alternative method to the traditional coulter counter for the rapid assessment of Hb concentration among critically ill adults. Since issues related to its safety, quality control, data entry and cost savings have yet to be addressed, however, use of such point of care testing should be viewed as a supplement to conventional laboratory testing.