Abstract & Commentary
Monitoring Hemoglobin by Pulse Oximetry in the ICU: Is it Accurate and Safe?
By David J. Pierson, MD, Editor, Professor Emeritus, Pulmonary and Critical Care Medicine, University of Washington, Seattle, is Editor for Critical Care Alert.
Synopsis: This study of ICU patients with acute gastrointestinal bleeding found that total hemoglobin as measured by pulse oximetry could not be determined in a substantial proportion of instances, and was inaccurate more than half the time when available, as compared to co-oximeter measurements on venous blood.
Source: Coquin J, et al. Precision of noninvasive hemoglobin-level measurement by pulse co-oximetry in patients admitted to intensive care units for severe gastrointestinal bleeds. Crit Care Med 2012; Jun 22. [Epub ahead of print.]
Coquin and colleagues evaluated the accuracy of noninvasive total hemoglobin measurement using a widely marketed pulse oximeter in patients admitted to the ICU with acute gastrointestinal (GI) hemorrhage. Non-moribund patients with GI bleeding and symptoms or signs of active hemorrhage within 12 hours of ICU admission were connected via fingertip sensor to a self-calibrating pulse oximeter (Masimo Rainbow SET, Irvine, CA) for continuous monitoring of total hemoglobin for the first 24 hours in the unit. Simultaneously every 8 hours during this period, hemoglobin concentrations were determined from the oximeter, from capillary blood using a portable photometer (HemoCue Hb 201, Meaux, France), and from venous blood using a laboratory co-oximeter (LH-780, Beckman Coulter, La Brea, CA). The authors used Bland-Altman analysis to assess the bias and precision of the pulse oximeter and capillary-blood hemoglobin measurements as compared to the co-oximeter readings. The primary study endpoint was the percentage of inaccurate measurements (defined as > 15% difference from the co-oximeter) by the other two devices.
During the 5-month study period, 75 patients were admitted to the ICU with acute GI bleeding. Of these, 31 were excluded for a priori study design reasons, 10 were admitted to attending physicians uninvolved in the study, and one refused, leaving 33 patients from whom data were reported. The patients received a median of four units of red blood cell transfusion (range, 0 to 36 units) during the 24-hour study period, and six of them (18%) received a norepinephrine infusion for hypotension. A reading for hemoglobin concentration could not be obtained from the pulse oximeter in 25 (19%) of the scheduled measurements. When readings could be obtained, they had a bias of 1.0 ± 1.9 g/dL, which was significantly greater than that for the capillary measurements (0.4 ± 1.0 g/dL, P < 0.05) in comparison to those in venous blood. Measurements with the pulse oximeter were inaccurate by the authors' criteria 56% of the time as compared to 15% for the capillary measurements (P < 0.05). Concordance was unaffected by norepinephrine infusion, although unavailability of readings by pulse oximetry was more frequent (42% vs 15%, P < 0.05).
Data from the pulse oximeter and the capillary-blood photometer were not used in managing the patients in this study. However, using the unit's transfusion thresholds based on hemoglobin concentration, the authors determined that erroneous decisions would have been made more often using the pulse oximetry readings, typically in the form of administering blood unnecessarily. Because of this possibility of unnecessary transfusion based on the pulse oximeter's readings, the authors conclude that the use of this device to guide transfusion in patients with acute GI bleeding may be hazardous.
Commentary
Over the last 20 or 30 years, there have been numerous instances where monitoring devices were introduced clinically into the ICU prior to rigorous study of their accuracy and clinical utility in that setting. Often such devices have been evaluated in normal volunteers, and then studied short-term in the controlled environment of the operating room, after which they have been promoted by their manufacturers or other advocates for use in critical care before clinical studies in critically ill patients have been done. The accuracy of the new pulse oximeters that measure total hemoglobin concentration has been evaluated in normal volunteers undergoing controlled hemodilution.1 However, patients with acute GI hemorrhage are typically vasoconstricted and their hemoglobin levels may change rapidly and unpredictably. The present study suggests that in this context these devices may be considerably less accurate.
Other studies should be done, both to confirm these findings and to assess the appropriateness of noninvasive hemoglobin monitoring in other types of critically ill patients. For now, the authors' conclusions seem reasonable: "Determination of pulse co-oximetry-based hemoglobin in patients presenting with severe gastrointestinal bleeds can be inaccurate, which renders its use to guide transfusions potentially hazardous. The unavailability of measurements, especially during vasopressor infusion, represents another serious limitation for hemorrhagic patients."
Reference
- Macknet MR, et al. The accuracy of noninvasive and continuous total hemoglobin measurement by pulse CO-Oximetry in human subjects undergoing hemodilution. Anesth Analg 2010;111:1424-1426.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content