Qualitative real-time analysis by nurses of sublingual microcirculation in intensive care unit: the MICRONURSE study
Sébastien Tanaka1, Anatole Harrois2,3, Camille Nicolaï4
1AP-HP, Service d'Anesthésie-Réanimation, Hôpitaux Universitaires Paris-Sud, Université Paris-Sud, Hôpital de Bicêtre, Le Kremlin-Bicêtre, France. sebastientanaka@hotmail.fr.
Introduction:
We aimed to determine i) the feasibility of nurses taking bedside measurements of microcirculatory parameters in real time in intensive care patients; and ii) whether such measurements would be comparable to those obtained by the classical delayed semi quantitative analysis made by a physician.
Methods:
This prospective observational study was conducted in a university hospital and was approved by our local Institutional Review Board (IRB 00006477). After ICU admission and study inclusion, a set of measurements of macrocirculatory and microcirculatory parameters was taken by the nurse in charge of the patient every 4 h within the first 12 h after admission and before and after every hemodynamic therapeutic intervention. Seventy-four sublingual microvascular measurements were performed with incident dark field illumination (IDF) microscopy in 20 mechanically ventilated patients hospitalized in the ICU.
Results:
There were no significant differences between the microvascular flow index (MFI) taken in real time by the nurses and the delayed evaluation by the physician. In fact, the nurses' real-time measurement of MFI demonstrated good agreement with the physician's delayed measurement. The mean difference between the two MFIs was -0.15, SD = 0.28. The nurses' real-time MFI assessment showed 97 % sensitivity (95 % CI: 84-99 %) and 95 % specificity (95 % CI: 84-99 %) at detecting a MFI <2.5 obtained by a physician upon delayed semiquantitative measurement. Concerning the density, 81 % of the paramedical qualitative density measurements corresponded with the automatized total vessel density (TVD) measurements. The nurses' real-time TVD assessment showed 77 % sensitivity (95 % CI: 46-95 %) and 100 % specificity (95 % CI: 89-100 %) at detecting a TVD <8 mm/mm(2).
Conclusion:
A real-time qualitative bedside evaluation of MFI by nurses showed good agreement with the conventional delayed analysis by physicians. The bedside evaluations of MFI and TVD were highly sensitive and specific for detecting impaired microvascular flow and low capillary density. These results suggest that this real-time technique could become part of ICU nurse routine surveillance and be implemented in algorithms for hemodynamic resuscitation in future clinical trials and regular practice. These results are an essential step to demonstrate whether these real-time measurements have a clinical impact in the management of ICU patients.
