Near-infrared spectroscopy for assessing tissue oxygenation and microvascular reactivity in critically ill patients: a prospective observational study

Abstract

Background: Impaired microcirculatory perfusion and tissue oxygenation during critical illness are associated with adverse outcome. The aim of this study was to detect alterations in tissue oxygenation or microvascular reactivity and their ability to predict outcome in critically ill patients using thenar near-infrared spectroscopy (NIRS) with a vascular occlusion test (VOT).

Methods: Prospective observational study in critically ill adults admitted to a 12-bed intensive care unit (ICU) of a University Hospital. NIRS with a VOT (using a 40 % tissue oxygen saturation (StO2) target) was applied daily until discharge from the ICU or death. A group of healthy volunteers were evaluated in a single session. During occlusion, StO2 downslope was measured separately for the first (downslope 1) and last part (downslope 2) of the desaturation curve. The difference between downslope 2 and 1 was calculated (delta-downslope). The upslope and area of the hyperaemic phase (receive operating characteristic (ROC) area under the curve (AUC) of StO2) were calculated, reflecting microvascular reactivity. Outcomes were ICU and 90-day mortality.

Results: Patients (n = 89) had altered downslopes and upslopes compared to healthy volunteers (n = 27). Mean delta-downslope was higher in ICU non-survivors (2.8 (0.4, 3.8) %/minute versus 0.4 (-0.8, 1.8) in survivors, p = 0.004) and discriminated 90-day mortality (ROC AUC 0.72 (95 % confidence interval 0.59, 0.84)). ICU non-survivors had lower mean upslope (141 (75, 193) %/minute versus 185 (143, 217) in survivors, p = 0.016) and AUC StO2 (7.9 (4.3, 12.6) versus 14.5 (11.2, 21.3), p = 0.001). Upslope and AUC StO2 on admission were significant although weak predictors of 90-day mortality (ROC AUC = 0.68 (0.54, 0.82) and 0.70 (0.58, 0.82), respectively). AUC StO2 ≤ 6.65 (1st quartile) on admission was independently associated with higher 90-day mortality (hazard ratio 7.964 (95 % CI 2.211, 28.686)). The lowest upslope in the ICU was independently associated with survival after ICU discharge (odds ratio 0.970 (95 % CI 0.945, 0.996)).

Conclusions: In critically ill patients, NIRS with a VOT enables identification of alterations in tissue oxygen extraction capacity and microvascular reactivity that can predict mortality.

Trial registration: NCT02649088, www.clinicaltrials.gov , date of registration 23rd December 2015, retrospectively registered.

Keywords: Critical illness; Microcirculation; Near-infrared spectroscopy; Tissue oxygenation; Vascular occlusion test.

Figures

Fig. 1

Tissue O2 saturation (StO2) and tissue hemoglobin index (THI) in healthy volunteers, ICU survivors and ICU non-survivors (on the first 3 days and on the day of death/discharge). *p < 0.05, **p < 0.01, ***p < 0.001, versus healthy volunteers, Kruskal-Wallis test with Dunn’s test for multiple comparisons. ##p < 0.01, Mann–Whitney U test. Number of patients is indicated below the error bars

Fig. 2

Downslope 1, downslope 2 and delta-downslope in healthy volunteers, ICU survivors and ICU non-survivors (on the first 3 days and on the day of death/discharge). *p < 0.05, **p < 0.01, ***p < 0.001, versus healthy volunteers, Kruskal-Wallis test with Dunn’s test for multiple comparisons. #p < 0.05, ##p < 0.01, Mann–Whitney U test. Number of patients is indicated below the error bars. StO2 tissue O2 saturation

Fig. 3

Upslope and area under the curve of the hyperaemic response (AUC StO2) in healthy volunteers, ICU survivors and ICU non-survivors (on the first 3 days and on the day of death/discharge). *p < 0.05, **p < 0.01, ***p < 0.001, versus healthy volunteers, Kruskal-Wallis test with Dunn’s test for multiple comparisons. #p < 0.05, ##p < 0.01, Mann–Whitney U test. Number of patients is indicated below or above the error bars. StO2 tissue O2 saturation