Multicenter, retrospective, observational study at 7 level 1 trauma centers. trauma patients (356 included in final analysis). This study is a post hoc analysis of a prospective study (number 3 in this table).
Nonprobability sampling. Inclusion criteria: patients who(1) had a SBP <90 mm Hg and BD ≥6 mEq/L within 60 minutes of admission; (2) received red blood cell (RBC) transfusions hours within 6 hours of admission; (3) were inured with at least 1 of the following: fractures (pelvic, long bone, multiple rib), pulmonary contusion, major blunt, or penetrating torso trauma.
Exclusion criteria: GCS ≤ 4, cardiac arrest before or upon arrival, advanced directives limiting aggressive care, penetrating or nonsurvivable brain injury, and bilateral upper extremity fractures.
Determine the relationship between hypothermia and patient outcome. Hypothermia defined as <35°C within the first 6 hours of admission. Outcome variables at 28 days of hospitalization: (1) multiple organ dysfunction syndrome (MODS); (2) mortality.
Overall incidence of early hypothermia was 43% (155/359). Multivariate logistic regression analysis determined that the minimum StO2 (<75%) within the first hour of admission was one of several significant predictors of MODS and mortality in both normothermic and hypothermic patients. Base deficit was able to predict mortality , but not MODS, among both hypothermic and normothermic patients.
Study StrengthsInclusion and exclusion criteria were clearly specified. Prolonged data collection period. Multicenter sampling increases the external validity of the findings.
Study Limitations (1) The retrospective nature and post hoc analysis of the study design limit the internal validity of the study. (2) Study funded by the manufacturer. (3) Authors do not disclose whether data was shared with the manufacturer prior to submission of paper.
Convenience sampling. Inclusion Criteria(a) Injury Severity Score (ISS) of at least 25; or (b) ISS = 15, plus need to transfusion of >6 units packed red blood cells (RBCs) within 6 hours of admission. Exclusion Criteria Glasgow Coma Scale <8.
To determine whether there is a relationship between early supply independent mitochondrial dysfunction and multiple organ dysfunction syndrome. All patients received the same resuscitation protocol with a specific goal for oxygen delivery (DO2) of 600 mL/min/m2: (1) SaO2 > 90%; (2) pulmonary artery wedge pressure (PAWP) 15–18 mm Hg; (3) hematocrit (hct) 35%. Serum lactate sampled at baseline and postresuscitation at 12 hours of admission. Near-infrared spectroscopy (NIRS) was used to measure tissue HbO2 and cytochrome a, a3. An additional fiberoptic probe was used to supply infrared light for absorption spectrum of cytochrome a, a3. The definition of mitochondrial oxygen consumption dysfunction, referred to as decoupling between tissue HbO2 and a, a3, was operationalized as the “absolute rate of change of the cytochrome a, a3 redox state relative to that of HbO2 was >0.03 absorbance units per hour” (p. 534).
The primary outcome variable was the incidence of multiple organ failure (MOF). MOF scores obtained within the first 48 hours of admission were not included in the data analysis. At 12 hours within admission, lactate levels were significantly higher in 9 (38%) who developed MOF. Evidence of decoupling was identified by NIRS in 8 (89%) of patients who developed MOF. Supranormal levels of DO2 and VO2 did not discriminate between patients who did and did not develop MOF, although, in clinical terms, the rates for both were considerably higher in patients who did not develop MOF.
Study StrengthsTwo of the investigators were blinded during the coding of the NIRS data. The interrater agreement was reported to be high (, CI 0.61, 1.00). Study Limitations (1) Small sample size. (2) Two authors were employed by the manufacturer. (3) In addition to grants from the National Institutes of Health and a Center of Excellence Award from the Emergency Medicine Foundation, the manufacturer provided a grant for this study. (4) The authors did not report evidence for an objective method for ensuring resuscitation protocols were consistently followed. (5) The authors do not report when MOF scores were obtained. Level of Evidence. C
Multicenter, observational study at 7 level 1 trauma centers. A power analysis based on a 14% incidence of MODS estimated that sample should consist of at least 278 patients in order to achieve a sample size of 35 patients who would likely develop MODS. patients; 381 included in final analysis.
Nonprobability sampling. Inclusion criteria: patients who (1) had a SBP <90 mm Hg and BD ≥6 mEq/L within 60 minutes of admission; (2) received red blood cell (RBC) transfusions hours within 6 hours of admission; (3) were inured with at least 1 of the following: fractures (pelvic, long bone, multiple rib), pulmonary contusion, major blunt, or penetrating torso trauma. Exclusion criteria: GCS ≤ 4, cardiac arrest before or upon arrival, advanced directives limiting aggressive care, penetrating or nonsurvivable brain injury, and bilateral upper extremity fractures.
(1) Evaluate predictive capability of monitoring for MODS or 28-day mortality in severely injured patients. (2) Identify threshold value for predicting MODS. StO2 sensor was placed within 30 minutes of ED arrival and remained in place for 24 hours or until death or discharge, if earlier. The minimum StO2 value within the first hour of ED arrival was used as the primary variable for data analysis.
MODS: when the minimum cutoff StO2 value is set at 75% within the first hour of admission: (a) StO2 sensitivity to MODS: 78%. (b) StO2 specificity for MODS: 39%. (c) StO2 positive predictive value for MODS: 18%. (d) StO2 negative predictive value for MODS: 91%. Mortality: when the minimum cutoff StO2 value is set at 75% within the first hour of admission: (a) StO2 sensitivity to mortality: 91%. (b) StO2 specificity for mortality: 37%. (c) StO2 positive predictive value for mortality: 20%. (d) StO2 negative predictive value for mortality: 96%. StO2 predicted which patients will not develop MODS, but poorly in those who do develop it. A minimum StO2 of 75% performed significantly better than SBP >90 mm Hg (39% versus 32%) in predicting MODS.
Study StrengthsThere were no adverse events related to the use of the monitor. Investigators were blinded to the StO2 data. Power analysis performed a priori. Study Limitations (1) No standardized resuscitation protocol reported. As a result, physician treatment bias may confound results. (2) The positive predictive values for both MODS and mortality are poor. (3) Authors state that, due to limitations in the availability of monitors and research staff at all hours, the study sample was a “random sample” of consecutive patients. The method of randomization is not stated. (4) StO2 collected on the thenar eminence adjacent to the site of radial artery cannulation in 34 patients was removed from the data analysis. (5) Study funded by the manufacturer. (6) Authors do not disclose whether data was shared with the manufacturer prior to submission of paper. Level of Evidence. C
Adult patients receiving mechanical ventilation for greater than 48 hours and considered ready to wean by physician: (a) partial or complete recovery from the underlying cause of acute renal failure; (b) adequate gas exchange, as indicated by partial pressure of arterial oxygen >60 torr; (c) FiO2 < 40%; (d) PEEP less than 5 cm; core temperature <38°C; Hgb greater than 8 grams/dL; (e) no further need for vasoactive and sedative agents. Exclusion criteria: (a) any injury to extremities that could hinder placement of NIRS sensor probe; (b) altered level of consciousness that could lead to central hypoventilation and/or excessive secretions.
To determine whether associated with an ischemic challenge is related to weaning outcome. The vascular occlusion test (VOT) was performed at the beginning of the weaning trial and at 30 minutes of weaning, providing paired data for changes in StO2.
An unsuccessful weaning process was associated with higher increases in StO2 deoxygenation rate (DeO2) and in local skeletal muscle oxygen consumption (nirVO2) after a 30-minute weaning trial. In contrast, successful weaning was not associated with these changes. Both groups of patients showed a significant increase in respiratory rate and heart rate, with no other changes in their hemodynamic, respiratory, and oximetry parameters.
Physician decision-maker for extubation had no access to the StO2 data. Standardized weaning trial protocol in place. Level of Evidence. C
Prospective observational pilot study. trauma patients.
Convenience sampling was used to enroll adult patients who were admitted to the ICU after trauma. Inclusion criteria: Abbreviated Injury Score ≥23. Requiring mechanical ventilation Assessed to be adequately resuscitated defined by protocol:(1) mean arterial pressure MAP ≥ 70 mm Hg, (2) heart rate HR ≥ 110 bpm, (3) base deficit ≤ 2, (4) partial pressure of arterial oxygen (PaO2) = 80-I 50 mm Hg. Average time from hospital admission to placement of devices was 10 hours. Exclusion criteria: patients with head injury who required craniotomy and postoperative vasopressor therapy were excluded; patients with spinal cord injuries.
(1) To evaluate the relationship between and implanted probe with other measures of tissue perfusion; (2) to determine values of and in resuscitated injured patients; (3) to identify the relationship between and values with subsequent infection and MODS within the first 24 hours of intensive care unit(ICU) admission. The Licox monitor, equipped with a polarographic tissue probe, was used to measure the partial tissue pressure of oxygen in 27 patients. The InSpectra monitor was used to noninvasively measure StO2 in 25 patients.
The relationship between StO2 and PmO2values in the same deltoid muscle bed was statistically significant but weak (, ). The mean PmO2 value in fully resuscitated patients was mm Hg. The mean value in fully resuscitated patients was %. Low PmO2 values (≤25 mm Hg) during the first 24 hours of ICU stay were significantly associated with ICU length of stay, MODS, infection, and mortality , while low StO2 values (≤35%) were not associated with these outcomes, with one exception: low StO2value (≤35%) persisting for longer than 2 hours on the first day of admission was able to predict MODS (odds ratio 12.8 CI 1.3–131, ).
Study StrengthsClinical criteria of adequate resuscitation were defined a priori.
Study Limitations (1) Small sample size. (2) Power analysis was not performed to determine adequate sample size. (3) The mean PmO2 and StO2 values reported in this study differ significantly from those published in other studies. Level of Evidence. C
Retrospective analysis of prospective, within-subjects cohort study,.
Samples size estimated by power analysis. Nonprobability sampling. Inclusion criteria: convenience sample of patients undergoing major abdominal or groin bypass surgery.
Determine whether measurements can predict postoperative surgical site infections (SSIs). Measured StO2 preoperatively over the biceps and at the intended incision site and at 12, 24, and 48 h after surgery. Postoperative StO2 measurements were taken through transparent dressings at the surgical site. Blinded to StO2 readings, an infection control nurse assessed surgical incisions at 7 and 30 days postoperatively. According to the authors, a noninvasive measure of SSI is needed because the National Nosocomial Infection Surveillance (NNIS) risk index has limited accuracy.
17 (29%) subjects developed SSIs. No statistical difference related to age, sex, body mass index, or triceps skinfold thickness was found between groups. All StO2 values increased postoperatively, regardless of postoperative SSI outcome. There was a significant difference in StO2 at the surgical site between patients who developed an SSI (43% ± 18.1) and those who did not (55% ± 22.0) at 12 and 48 hours after operation (; ). Sensitivity: 70.6%. Specificity: 76.2%. StO2 performed significantly better at predicting SSIs than the NNIS risk index.
Study StrengthsInnovative, well-designed study controlled with baseline measurements. Blinded outcome assessment. Power analysis performed a priori. Study Limitations Standard deviations for both groups were large with overlap. The mean values are lower than the mean StO2 values reported in the literature. The validity of findings is limited by the retrospective design. Postoperative measurements of surgical-site infections (SSIs) were performed at low frequency. Loaned equipment from manufacturer is acknowledged. Authors do not disclose whether data was shared with the manufacturer prior to submission of paper. Level of Evidence. C
Nonprobability sampling. Inclusion criteria: convenience sample of consecutively admitted patients diagnosed with septic shock according to the International Sepsis Forum Definition of Infection in the ICU Consensus Conference (2005).
(1) Determine the relationship between values and mortality in patients with septic shock after early resuscitation. Data collection initiated after optimization of macro hemodynamic variables (MAP, urine output, and ScvO2) according to the Consensus Conference recommendations.
Sample analyzed by groups: entire sample, survivors on day 3, and survivors on day 28. Nine (64%) patients whose StO2 was <78% did not survive to day 28 when compared to 5 patients who did . Overall, StO2 values were significantly lower in nonsurvivors than in survivors (73% 68–82% versus 84% 81–90%, ). A receiver operating curve analysis confirmed that StO2 was associated with mortality with an area under the curve at 71% (52–91%,), whereas elevated lactate levels were not significantly related to mortality.
Study StrengthsResuscitation treatment protocol clearly stated. Study Limitations (1) Endpoint of optimization of hemodynamic variables was determined but may have been modified by the attending physician. (2) Duration of StO2 monitoring is unclear. (3) No intervention is based on unblinded StO2 values. (4) Power analysis is not performed. (5) Authors do not refer to conflict of interest. Level of Evidence. C
Convenience sample. Inclusion criteria: elevated lactate levels (>3.0 mmol/L) upon admission to the ICU; absence of history of peripheral vascular disease.
To test the hypothesis that persistence of low values less than 70% is related to organ failure and mortality. Early goal-directed therapy (EGDT) based on the Surviving Sepsis Guidelines was used to guide resuscitation. StO2 measurements were recorded for 8 hours every 2 hours.
Among the 12 patients who at baseline had low values, the 10 patients who were unable to normalize low values were more likely to develop MODS and have an increased mortality and persistence of elevated serum lactate .
Study Strengths(1) StO2 monitoring was performed at baseline and every 2 hours during the first 8 hours following resuscitation. (2) Data collector of NIRS data blinded to treatment and clinical information. Study Limitations (1) Small sample size. (2) Publication of this information in a journal supplement was supported by the manufacturer. Level of Evidence. C
Inclusion criteria: early stage of septic shock (authors state “less than 24 hours” but do not specify when measurements began); decision to insert PAC catheter; stabilized MAP at least 65 mm Hg for 30-minute period.
(1) Test the significance of the relationship between and global .
The correlation between StO2 and global DO2 was 0.78 . There were also strong statistical correlations with SvO2 (0.77, ), CvO2 (0.8, ), and O2ER (−0.76, ). A cutoff StO2 value of 75% predicted an extremely low DO2 (<450 mL/min/m2) with a sensitivity and specificity of 90%. Serum lactate values among patients with high DO2 (4.1 ± 2.5) versus low (5.2 ± 2.9) were clinically but not statistically different. Mean StO2 value among patients with high DO2 was 84% ± 8. The mean StO2 value among patients with a low DO2 was 68% ± 5 . StO2 monitoring did not correlate with moderately low DO2 (450–600 mL/min/m2). There was no relationship between DO2 level and mortality. The authors do not report whether there was a significant relationship between StO2 levels and mortality.
Study StrengthsThese findings suggest that, in the absence of a PAC catheter, StO2 and serum lactate might accurately substitute for DO2 measurements.
Study Limitations (1) Hemodynamic management protocol was not described. (2) Patient recruitment occurred at different points during resuscitation. (3) The authors point out that StO2 values did not correlate with moderately low DO2, thereby suggesting that StO2 has greater sensitivity to only extremely low DO2. (4) Small sample size. Level of Evidence. C
Case controls matched to age and sex: 9 healthy volunteers and 10 patients.
Nonprobability sampling. Inclusion criteria: convenience sample of patients admitted to ICU with severe sepsis and one episode of hypotension (SBP < 90 mm Hg) within 24 hours of study entry as well as criteria based on the SCCM/ACCP Consensus Conference (1995): at least one organ dysfunction, positive culture results, and 2 of the following 4: WBC >12,000 or <4000/mm3, fever >38 C or <36 C, tachycardia >90/min, and tachypnea >20 breaths/min. Exclusion criteria: patients in whom a PA catheter was contraindicated.
Determine the magnitude of the relationship between measurements and severity of sepsis and invasive hemodynamic measurements.
Volunteers had higher MAP, lower serum lactate concentration, and higher StO2 than patients with sepsis . Near-infrared spectroscopy-derived mixed venous oxygen saturation (NIRSSvO2) and StO2 correlated with SvO2 measured by blood drawn from the PA catheter (, ).
Study StrengthsThe correlations between NIRS-related data and SvO2 were found to be statistically significant, but weak . Caregivers were blinded to StO2 data. Study Limitations (1) Small sample size. (2) Relationship to patient outcome was not analyzed. (3) One author was employed by the manufacturer. (4) No statement regarding whether data was shared with manufacturer prior to submission for publication. Level of Evidence. C
Nonprobability sampling. Inclusion criteria: septic shock and MODS.
Determine whether degree of edema confounds accurate measurement of . InSpectra probes were placed at multiple muscle sites: bilateral deltoid muscles, brachial and thenar eminence, and bilateral vastus and gastrocnemius muscles. Tissue thickness and edema were measured by echography.
The lowest variance (within-subject variability) of StO2 was identified in the thenar eminence (22%). The degrees of edema and skin thickness were lowest in the thenar eminence. StO2 values significantly correlated with degree of edema (, ) and with total tissue thickness from skin to muscle , ).
Study StrengthsPilot study examining methodological issues of accuracy and precision of StO2 measurement. Study Limitations (1) Accuracy and precision of echocardiography not reported. (2) Small sample size. Level of Evidence. C
Nonprobability sampling. Inclusion criteria: convenience sample of all adult patients who underwent cardiopulmonary bypass (CPB) procedures.
(1) To determine the ability ofto detect subtle changes in tissue oxygenation in a controlled model of altered perfusion. (2) Compare with standard hemodynamic variable (MAP and ) (3) Examine the temporal relationship between and standard biochemical variables in real-time CPB. Measured vital signs and central pressures at stages of CPB: preinduction, postinduction, sternotomy, initiation of CPB, after 30 minutes of CPB at start of rewarming, when CPB was terminated, and at sternal closure. Data analysis focused on StO2, MAP, CO, DO2, lactate, BD, and temperature.
StO2 changes: initiation of CPB: decreased by 12.9% ; this decrease preceded maximum lactate level by on average 94 minutes; this decrease also correlated with a decrease in DO2. StO2 increased from the rewarming period to termination of CPB . A threshold value of 79% was identified as an indicator of hypoperfusion.
Study StrengthsPerioperative caregivers were blinded to the StO2 data. Study Limitations (1) A power analysis was not performed to optimal sample size. (2) A table presenting the variables of interest at each stage of surgery would have been helpful in the interpretation of real-time changes in key variables. Level of Evidence. C
Convenience sampling. Inclusion criteria: first-time CABG or cardiac valvular surgery requiring cardiopulmonary bypass. Exclusion criteria: patients were excluded for the following circumstances: emergent surgery; congenital heart condition; current participation in a clinical intervention trial; or having dermatological contraindications for use of the InSpectra probe.
To determine the degree of association of changes in tissue oxygen saturation with postoperative outcome in cardiac surgery patients. The primary outcome was postoperative morbidity. Data was collected at the induction of anesthesia at a rate of 1/min during first 20 minutes in ICU and at 2, 6, and 12 hours after surgery until extubation in the ICU or completion of total monitoring time of 24 hours.
Mean StO2 measures during the first minutes of anesthesia and at 20 minutes in the ICU were lower in patients with postoperative morbidity scores than in patients without such morbidity on day 3 (72.9% versus 85.5%; ) and day 15 (81.1% versus 87.6%; ).
Study Strengths A priori power analysis was performed to determine sample size. Study Limitations (1) This study was supported by a grant from the manufacturer. (2) The authors do not report whether the manufacturer had access to the data prior to publication. Level of Evidence. C
Convenience sampling. Inclusion criteria: patients estimated to be at risk for hemorrhagic shock, including any of the following: SBP ≤ 90 mm Hg; injury to neck, chest, abdomen, flank, buttock, or proximal extremities; uncontrollable hemorrhage; or suspected ongoing internal hemorrhage. Exclusion criteria: any of the following: isolated head injury; prehospital traumatic cardiopulmonary arrest; transfers from other hospitals; hand injury preventing placement of StO2 sensor.
(1) Determine the threshold value predictive of need for blood transfusion. (2) Evaluate the potential relationship between values and hospital length of stay and mortality. StO2 measured for 1 hour after arrival to trauma bay. Independent variable: minimum StO2 used as marker for trigger to blood transfusion within first hour of admission (dependent outcome variable).
Sample categorized as: HI-StO2 ≥ 70% or LO-StO2 < 70%. No statistical differences between 2 groups based on gender, age, ISS, RTS, SBP, HR, serum lactate, BD, hgb, or 30-day mortality were found. LO-StO2 patients 9 times more likely to receive transfusion (64%) than HI-StO2 group (7%) . Survivor length of hospital stay longer in LO-StO2 patients than in HI-StO2 patients .
Study Strengths(1) Caregivers blinded to the StO2 data. (2) Inclusion/exclusion criteria clearly stated. Study Limitations (1) Authors acknowledge that selection bias may have occurred, as decision to treat with transfusion made by attending trauma surgeon or chief surgical resident. (2) Small sample size. (3) Authors do not refer to conflict of interests. Level of Evidence. C
StO2: tissue oxygen saturation; DO2I: oxygen delivery indexed to body surface area; VO2I: oxygen consumption determined by metabolic cart, indexed to BSA: body surface area; SvO2: mixed venous oxygen saturation; hgb: hemoglobin; hct: hematocrit; ABG: arterial blood gas; NIRS: near-infrared spectroscopy; PA: pulmonary artery; Sensitivity: proportion of actual positives which are correctly identified; Specificity: proportion of negatives which are correctly identified; SSI: surgical-site infection; PAD-IC: peripheral arterial disease with intermittent claudication; ABI: ankle brachial index; BMI: body mass index; MODS: multiple organ dysfunction syndrome; SBP: systolic blood pressure; ED: emergency department; CPB: cardiopulmonary bypass; CO: cardiac output; DO2 mL/minute: oxygen delivery measured in milliliters per minute; ISS: Injury Severity Score; RTS: Revised Trauma Score; BD: base deficit; ScvO2: central venous oxygen saturation of oxygen; ROC: receiver operating characteristic curves; 50: time in seconds from peak exercise to 50% StO2 recovery; 100: time elapsed from peak exercise to 100% StO2 recovery time.
*Definition of terms is as follows. Sensitivity: percentage of subjects who have both the condition (MODS) and a positive test. Specificity: percentage of patients who did not have the condition (MODS) and a negative test. Positive predictive value: percentage correctly diagnosed with a positive test. Negative predictive value: percentage correctly diagnosed with a negative test.
