key: cord-0021738-o9i53lr2
authors: Ahmed, Walid; Laimoud, Mohamed
title: The Value of Combining Carbon Dioxide Gap and Oxygen-Derived Variables with Lactate Clearance in Predicting Mortality after Resuscitation of Septic Shock Patients
date: 2021-09-25
journal: Crit Care Res Pract
DOI: 10.1155/2021/6918940
sha: 6e6075a3197e702603dc0155ebefc604957ee970
doc_id: 21738
cord_uid: o9i53lr2

BACKGROUND: Achieving hemodynamic stabilization does not prevent progressive tissue hypoperfusion and organ dysfunction during resuscitation of septic shock patients. Many indicators have been proposed to judge the optimization of oxygen delivery to meet tissue oxygen consumption. METHODS: A prospective observational study was conducted to evaluate and validate combining CO(2) gap and oxygen-derived variables with lactate clearance during early hours of resuscitation of adults presenting with septic shock. RESULTS: Our study included 456 adults with a mean age of 63.2 ± 6.9 years, with 71.9% being males. Respiratory and urinary infections were the origin of about 75% of sepsis. Mortality occurred in 164 (35.9%) patients. The APACHE II score was 18.2 ± 3.7 versus 34.3 ± 6.8 (p < 0.001), the initial SOFA score was 5.8 ± 3.1 versus 7.3 ± 1.4 (p=0.001), while the SOFA score after 48 hours was 4.2 ± 1.8 versus 9.4 ± 3.1 (p < 0.001) in the survivors and nonsurvivors, respectively. Hospital mortality was independently predicted by hyperlactatemia (OR: 2.47; 95% CI: 1.63–6.82, p=0.004), PvaCO(2) gap (OR: 2.62; 95% CI: 1.28–6.74, p=0.026), PvaCO(2)/CavO(2) ratio (OR: 2.16; 95% CI: 1.49–5.74, p=0.006), and increased SOFA score after 48 hours of admission (OR: 1.86; 95% CI: 1.36–8.13, p=0.02). A blood lactate cutoff of 40 mg/dl at the 6th hour of resuscitation (T6) had a 92.7% sensitivity and 75.3% specificity for predicting hospital mortality (AUROC = 0.902) with 81.6% accuracy. Combining the lactate cutoff of 40 mg/dl and PvaCO(2)/CavO(2) ratio cutoff of 1.4 increased the specificity to 93.2% with a sensitivity of 75.6% in predicting mortality and with 86.8% accuracy. Combining the lactate cutoff of 40 mg/dl and PvaCO(2) gap of 6 mmHg increased the sensitivity to 93% and increased the specificity to 98% in predicting mortality with 91% accuracy. CONCLUSION: Combining the carbon dioxide gap and arteriovenous oxygen difference with lactate clearance during early hours of resuscitation of septic shock patients helps to predict hospital mortality more accurately.

Early detection of tissue hypoperfusion and rapid, efficient resuscitation is fundamental in the successful management of patients presenting with septic shock [1] . Recent guidelines from the Surviving Sepsis Campaign for management of septic shock patients focused on hemodynamic support through a systematic protocol of fluids and vasopressor therapy. e goal was to improve tissue perfusion and meet tissue oxygen demands [2] . e guidelines recommended continuing resuscitation and restoring mean arterial pressure (MAP) ≥ 65 mmHg with lactate clearance.

is was based on the understanding that lactate clearance could serve as a surrogate for the reversal of global tissue hypoxia [3, 4] . Many indicators have been proposed to determine optimization of oxygen delivery (DO 2 ) to meet tissue oxygen consumption (VO 2 ) [5, 6] . However, achieving hemodynamic stabilization does not prevent progressive tissue hypoperfusion and multiorgan dysfunction [7] . No consistent advantages have been found for lactate guided resuscitation over using oxygen indicators [8] [9] [10] . e venoarterial carbon dioxide difference (PvaCO 2 ) has been proposed as an indicator of tissue hypoperfusion [11] [12] [13] . A persistently high PvaCO 2 gap despite resuscitation efforts may anticipate lactate changes and adverse outcomes [14] as CO 2 production does not exceed O 2 availability during aerobic conditions. us, the ratio between the PvaCO 2 and the arteriovenous oxygen content difference (CavO 2 ) may detect patients with anaerobic metabolism. Mekontso-Dessap et al. [15] reported that a PvaCO 2 to CavO 2 ratio of 1.4 was superior to PvaCO 2 in predicting hyperlactatemia. e goal of this prospective observational study was to evaluate predictors of outcomes by combining the CO 2 gap to CavO 2 , PvaCO 2 /CavO 2 ratio, and lactate clearance during early hours of resuscitation of adult patients presenting with septic shock.

is study was conducted at Cairo University Hospitals between January 2018 and February 2020 after getting the approval of the ethical committee. Informed consents were obtained from the enrolled patients. All 18-year-old or older consecutive patients who presented with septic shock were enrolled in the study. Septic shock was diagnosed as sepsis with persisting hypotension and a blood lactate level >2 mmol/L (18 mg/dl) despite adequate fluid resuscitation and necessitating vasopressors to get a mean arterial blood pressure ≥65 mm Hg [2] . Sepsis was defined as critical organ dysfunction due to a dysregulated body response to infection and associated with an acute change in Sequential Organ Failure Assessment (SOFA) score ≥ 2 [2] .

Exclusion criteria included patients below 18 years of age, patients with chronic obstructive pulmonary disease or interstitial pulmonary diseases, refusal to consent, conditions that could affect lactate clearance such as chronic liver disease or alcoholism, diabetic patients on metformin therapy, and patients with different ventilation parameters during the 2 points of measurements.

All enrolled patients were resuscitated according to the Surviving Sepsis Campaign guidelines [2, 16] . Resuscitation started with crystalloids (30 ml/kg), and norepinephrine was started in the first hour as the first-choice vasopressor to maintain MAP ≥65 mm Hg. Cultures were obtained before starting broad-spectrum antibiotics. Central venous catheterization was performed via a jugular or subclavian approach, and the tip of the catheter was confirmed in the upper part of the right atrium by a chest X-ray. Arterial catheterization was performed via a radial or femoral approach, and the catheter position was confirmed by arterial waveform and blood gases analysis. Simultaneous arterial and central venous blood gases samples were obtained and analyzed.

(i) Oxygenation and ventilation parameters were compared at the start of resuscitation (T0) and after 6 hours of resuscitation (T6) including arterial oxygen tension (PaO 2 ), arterial carbon dioxide tension (PaCO 2 ), arterial oxygen saturation (SaO 2 %), central venous oxygen tension (PcvO 2 ), central venous carbon dioxide tension (PcvCO 2 ), central venous oxygen saturation (ScvO 2 %), the arterial oxygen content (CaO 2 ), and central venous oxygen content (CvO 2 ). e difference between arterial and venous oxygen content (CavO 2 ), oxygen extraction ratio (ER O 2 %), CO 2 gap between venous and arterial samples (PvaCO 2 ), and PvaCO 2 /CavO 2 ratio were calculated as follows: ER O 2 % � (CaO 2 − CvO 2 )/CaO 2 and PvaCO 2 � PvCO 2 − PaCO 2 [17] . (ii) Lactate was recorded at 3 points: before (T0), after 6 hours (T6), and after 12 hours (T12) of resuscitation. Lactate clearance and delta changes were calculated from the difference between the second or the third reading and the first reading, divided by the first reading. is determined maximal change in the second or third reading. (iii) APACHE II and SOFA scores were calculated upon admission, as well as SOFA scores after 48 hours. (iv) Hospital mortality was the primary outcome, while the secondary outcomes included ICU stay and the need for hemodialysis.

(i) Data are reported as mean ± standard deviation (±SD), median with interquartile range (IQR), or the number of cases and relative frequencies (percentages) when appropriate. Comparison of quantitative variables between the study groups was calculated using the t-test or Mann-Whitney U test for independent samples. For categorical data, a chisquare test was performed. Correlation between various variables was done using the Pearson correlation coefficient for continuous variables. A probability value (p value) less than 0.05 was considered statistically significant. Receiver operator characteristic (ROC) analysis was used to determine the optimum cutoff values for determining mortality. (ii) We developed the lactate-PvaCO 2 score as follows:

after multivariate logistic regression analysis to identify independent risk factors associated with mortality, the probability of using a stepwise analysis model was 0.05 for entry and 0.1 for removal. We tested various regression models using four variables (lactate, PvaCO 2 , CavO 2 , and PvaCO 2 /CavO 2 ). We evaluated the final model (lactate-PvaCO 2 ) for the goodness of fit using the Hosmer-Lemeshow test (p > 0.05). e variables were classified in the final model into clinically meaningful categories, and the estimated risk of mortality was recorded in each category. Binary logistic regression was run to explore potential predictors (following fluid resuscitation) for mortality. A score was calculated according to the regression output as follows: mortality probability � 1/(1 + exp(− (− 8.968 + (0.119 * lactate) + (0.379 * PvaCO 2 )))). "exp" is an exponential value (Tables 1  and 2 3 ± 6.8 (p < 0.001) and the initial SOFA score was 5.8 ± 3.1 versus 7.3 ± 1.4 (p � 0.001), while SOFA score after 48 hours was 4.2 ± 1.8 versus 9.4 ± 3.1 (p < 0.001) in the survivors and nonsurvivors, respectively. e nonsurvivors had significant hypoalbuminemia (3.2 ± 1.6 versus 3.6 ± 1.3, p � 0.02) and a higher mean serum creatinine level (1.6 ± 0.6 versus 1.2 ± 0.2, p � 0.04) compared to the survivors. e survivors had a significantly longer length of ICU stay (6.8 ± 2.1 versus 3.6 ± 1.7, p < 0.001) but a lesser need for hemodialysis (10.3% versus 23.2%, p � 0.02) compared to the nonsurvivors (Table 3) .

e mean arterial blood pressure (MAP), heart rate, and temperature measurements were similar in both groups. Before starting resuscitation, the nonsurvivors had a significantly lower mean ScvO 2 % (52.6 ± 8.8 versus 61.4 ± 10.3, p � 0.003) and PvaCO 2 /CavO 2 ratio (1.8 ± 0.15 versus 2.2 ± 0.81, p � 0.013) with a higher oxygen extraction ratio (40 ± 7.3% versus 35 ± 8.1%, p � 0.002) compared to the survivors. However, there were no significant differences between both groups in blood lactate levels, arterial PO 2 , arterial oxygen content, arterial PCO 2 , or carbon dioxide gap (PvaCO 2 ) at the initiation of resuscitation (Table 4 and Figure 1 ).

After 6 hours of resuscitation, the survivors showed a significantly higher MAP (69. 8 4) increased the specificity to 93.2%, with a sensitivity of 75.6% in predicting mortality and 86.8% accuracy. Combining cutoffs for lactate (40 mg/dl) and PvaCO 2 gap (6 mmHg) increased the sensitivity to 93% and increased the specificity to 98% in predicting mortality with 91% accuracy (Table 7 and Figures 3 and 4) .

Our study's main results are that septic shock patients who do not survive have worsening lactate levels with increased PvaCO 2 and PvaCO 2 /CavO 2 ratio, even after resuscitation. Mortality is predicted by hyperlactatemia, increasing PvaCO 2 gap, incrementing PvaCO 2 /CavO 2 ratio, and high SOFA score after 48 hours. Combining the PvaCO 2 gap and PvaCO 2 /CavO 2 ratio with lactate measurements can help in resuscitation and mortality prediction of septic shock patients. Many parameters are measured during septic shock resuscitation, including arterial and central venous pressures, urine output, cardiac output, blood lactate level, and central venous oxygen saturation (ScvO 2 %). e goal of a MAP at least 65 mmHg by the guidelines of Surviving Sepsis Campaign (SSC) has been challenged by many studies because targeting a predefined MAP by fluids and vasopressors did not lead to increased survival [18] [19] [20] . Hernandez et al. [21] reported that sepsis-induced hypotension without hyperlactatemia was associated with low mortality and less risks of organ dysfunction. Houwink et al. [22] reported the greater importance of first 24 hours' lactate over the MAP during septic shock resuscitation. ey divided patients into 4 subgroups according to blood lactate cutoff of 2 mmol/L and MAP cutoff of 65 mmHg and reported lower mortality in the groups with low blood lactate regardless of MAP [22] .

Our study showed significantly higher oxygen extraction and lower ScvO 2 % in the nonsurvivors compared to the survivors during early resuscitation without significant differences in arterial oxygen content (CaO 2 ), arterial oxygen pressure (PaO 2 ), or hemoglobin level in both groups.

is could explain the increasing oxygen demands and failure of the aerobic metabolism with a subsequent anaerobic pathway activation and hyperlactatemia. After 6 hours of resuscitation, oxygen extraction increased and was associated with lactate clearance and improving markers of anaerobic metabolism in the survivors compared to the nonsurvivors.

ScvO 2 % was widely recommended targeting ≥70% during the first 6 hours of septic shock resuscitation [23] [24] [25] [26] and ScvO 2 % less than 70% was predictive of mortality [23, 27, 28] . However, ScvO 2 % was unable to differentiate survivors from nonsurvivors in other studies [29] [30] [31] .

Our study showed that both the survivors and nonsurvivors had similar initial lactate levels, but, after resuscitation, lactate clearance was evident in the survivors. Progressive hyperlactatemia has been associated with mortality and other negative clinical outcomes in septic and nonseptic critically ill patients [3, 4, 10, 29, [32] [33] [34] . Despite the proven beneficial role of lactate clearance in guiding [35] [36] [37] . We report that the nonsurvivors received more fluids during the first 6 hours of resuscitation compared to the survivors. Boyd et al. [35] reported that more fluids transfused during the first 12 hours of septic shock resuscitation were linked to morbidity and mortality. Sadaka et al. [36] linked more fluid balance at 24 hours after resuscitation to increased mortality and documented increased mortality from 42 to 58% if there were more than 6 liters of positive fluid balance. e PvaCO 2 gap has been used as an important marker of tissue perfusion and cardiac output [11-14, 38, 39] . Its value persists in septic shock resuscitation, especially after getting MAP to at least 65 mmHg and normalization of ScvO 2 % [11, 12, 14] . Our results showed a significant decrease of the PvaCO 2 gap in the survivors after 6 hours of resuscitation compared to the nonsurvivors without a significant difference between their initial values at T0. PvaCO 2 gap was also a predictor of mortality in the multivariate regression analysis. Moreover, ROC analysis showed the PvaCO 2 gap cutoff of 6 mmHg had an AUROC of 0.79 but a lesser predictive value compared to blood lactate in predicting mortality. Combining lactate with PvaCO 2 gap increased the sensitivity and specificity with AUROC of 0.93 and accuracy of 91% in predicting mortality. Ospina-Tascón et al. [14] reported that a persistently high PvaCO 2 gap could independently predict adverse outcomes and lactate changes regardless of other oxygen-derived variables. e PvaCO 2 gap is a simple and attractive goal in resuscitation but it is a physiologically complex tool and should be studied with oxygen changes as it may be normal despite tissue hypoperfusion if associated with high blood flows preventing CO 2 accumulation. Also the PvaCO 2 gap may increase without tissue hypoperfusion in aerobic and anaerobic conditions due to the Haldane effect as the relation between the CO 2 partial pressure and the CO 2 content is affected by O 2 saturation, pH differences, and hemoglobin changes [40] .

As CO 2 production does not exceed O 2 production during aerobic metabolism, correcting the PvaCO 2 gap by using the ratio of PvaCO 2 to the arteriovenous oxygen content difference (CavO 2 ) may detect patients with a risk of anaerobic metabolism [14, 40, 41] . Mekontso-Dessap et al. [15] reported that a PvaCO 2 to CavO 2 ratio of 1.4 was superior to PvaCO 2 in predicting hyperlactatemia and also reported the agreement between PvaCO 2 /CavO 2 ratio and blood lactate levels. Our study showed a difference in the PvaCO 2 /CavO 2 ratio combined with lactate clearance in the survivors compared to the nonsurvivors who develop increased ratio and lactate elevation after 6 hours of resuscitation. Our results showed that PvaCO 2 /CavO 2 ratio was a predictor of mortality in the multivariate regression analysis. e PvaCO 2 /CavO 2 ratio cutoff of 1.4 had an AUROC of 0.793 in differentiating mortality. Our ROC analysis to determine possible cutoffs for lactate, PvaCO 2 , and PvaCO 2 / CavO 2 to predict mortality showed that blood lactate had a superior performance differentiating mortality. However, combining the cutoffs for lactate and PvaCO 2 or PvaCO 2 / CavO 2 provided higher specificity (98% and 93.2%, respectively) than blood lactate levels alone (75.3%) with higher accuracy. Our results were consistent with Ospina-Tascón et al.'s [42] study that described lactate and PvaCO 2 / CavO 2 ratio as independent predictors of mortality after 6 hours of resuscitation. Interestingly, Ospina-Tascón et al. [42] reported that, after getting the MAP ≥65 mmHg and ScvO 2 % ≥ 65% in most enrolled patients, 48% of the studied patients had a PvaCO 2 /CavO 2 ratio ˃1 and 62% of patients had a blood lactate ˃2 mmol/L. Ospina-Tascón et al. [42] reported that patients with combined high lactate and PvaCO 2 /CavO 2 ratio at T6 had the worst outcomes. Patients with normalized lactate level and PvaCO 2 /CavO 2 ratio had the best outcomes, while patients with normalized lactate but still high PvaCO 2 /CavO 2 ratio had similar unfavorable outcomes to those with high lactate levels with normalized PvaCO 2 /CavO 2 ratio. We used SOFA score to assess the clinical severity of the studied patients and the degree of multisystem dysfunction after resuscitation. e SOFA score has been previously used in different critically ill patients and is linked to mortality and outcomes [43] [44] [45] . Our results revealed that the nonsurvivors had higher initial SOFA and increasing scores trend after 48 hours of admission compared to the survivors. Also, the increasing trend of SOFA score was an independent predictor of mortality after resuscitation. Mesquida et al. [46] used SOFA scoring and reported increased SOFA trend and PvaCO 2 /CavO 2 ratio in nonsurvivors after resuscitation. Ospina-Tascón et al. [42] divided the septic shock patients into four groups based on the blood lactate and PvaCO 2 /CavO 2 ratio and reported the highest scores in the group with combined lactate ≥2 mmol/L and PvaCO 2 /CavO 2 ratio ˃1 and the lowest scores in patients with low lactate ˂2 mmol/L and PvaCO 2 / CavO 2 ratio ˂1.

Finally, our data showed that septic shock patients who were unlikely to survive persistently had worsening lactate levels with high PvaCO 2 and PvaCO 2 /CavO 2 ratios, despite resuscitation. Increased oxygen extraction was associated with increased lactate clearance and improving markers of anaerobic metabolism. Adding PvaCO 2 gap and PvaCO 2 / CavO 2 ratio to lactate measurements can increase the accuracy of mortality prediction.

Combining the carbon dioxide gap and arteriovenous oxygen difference with lactate clearance during the early hours of resuscitation of adult patients with septic shock helps to predict hospital mortality more accurately.

e data used in this study are available from the corresponding author upon request.

A PvaCO 2 gap is a simple tool, but its validity is debatable in high-cardiac-output septic patients. Our study was conducted before the COVID-19 pandemic, and we do not know if these results can apply to patients with COVID-19 presenting with septic shock. We planned to include the cardiac output and response during resuscitation but were unable to do so because of resource limitations. Most of the studied patients had echocardiographic assessments after resuscitation. So, we included only the systolic function of the left ventricle EF.

e authors declare that they have no conflicts of interest. Critical Care Research and Practice

Consensus on circulatory shock and hemodynamic monitoring. Task forceof the European Society of Intensive Care Medicine

Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016

Early lactate clearanceguided therapy in patients with sepsis: a meta-analysis with trial sequential analysis of randomized controlled trials

Early lactate-guided therapy in intensive care unit patients: a multicenter, open-label, randomized controlled trial

Oxygen delivery in septic shock

Use of central venous oxygen saturation to guide therapy

Outcomes of patients undergoing early sepsis resuscitation for cryptic shock compared with overt shock

Lactate clearance vs central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial

Outcome effectiveness of the severe sepsis resuscitation bundle with addition of lactate clearance as a bundle item: a multi-national evaluation

Early lactate-guided therapy in intensive care unit patients: a multicenter, open-label, randomized controlled trial

Resuscitation of patients with septic shock: please "mind the gap"!

Central venous-arterial PCO2 difference as a tool in resuscitation of septic patients

e forgotten hemodynamic (PCO2 gap) in severe sepsis

Persistently high venous-to-arterial carbon dioxide differences during early resuscitation are associated with poor outcomes in septic shock

Combination of venoarterial PCO2 difference with arteriovenous O 2 content difference to detect anaerobic metabolism in patients

e surviving sepsis campaign bundle: 2018 update

Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016

Arterial pressure optimization in the treatment of septic shock: a complex puzzle

High versus low blood-pressure target in patients with septic shock

Association of arterial blood pressure and vasopressor load with septic shock mortality: a post hoc analysis of a multicenter trial

Persistent sepsisinduced hypotension without hyperlactatemia: a distinct clinical and physiological profile within the spectrum of septic shock

e association between lactate, mean arterial pressure, central venous oxygen saturation and peripheral temperature and mortality in severe sepsis: a retrospective cohort analysis

Early goal-directed therapy in the treatment of severe sepsis and septic shock

Prospective external validation of the clinical effectiveness of an emergency department-based early goal-directed therapy protocol for severe sepsis and septic shock

Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock

Translating research to clinical practice: a 1-year experience with implementing early goal-directed therapy for septic shock in the emergency department

Hemodynamic variables related to outcome in septic shock

Prevalence of low central venous oxygen saturation in the first hours of intensive care unit admission and associated mortality in septic shock patients: a prospective multicentre study

Lactate clearance for death prediction in severe sepsis or septic shock patients during the first 24 hours in Intensive Care Unit: an observational study

High central venous oxygen saturation in the latter stages of septic shock is associated with increased mortality

e pursuit of a high central venous oxygen saturation in sepsis:growing concerns

Effective lactate clearance is associated with improved outcome in postcardiac arrest patients

e clinical significance of blood lactate levels in evaluation of adult patients with venoarterial extracorporeal membrane oxygenation

Lactate clearance time and concentration linked to morbidity and death in cardiac surgical patients

Fluid resuscitation in septic shock: a positive fluid balance and elevated central venous pressure are associated with increased mortality

Fluid resuscitation in septic shock

Etiology and therapeutic approach to elevated lactate levels

Central venous O 2 saturation and venous-to-arterial CO 2 difference as complementary tools for goal-directed therapy during high-risk surgery

Central venous-to-arterial carbon dioxide difference: an additional target for goal-directed therapy in septic shock?

e Haldane effect-an alternative explanation for increasing gastric mucosal PCO2 gradients?

Lactate and venoarterial carbon dioxide difference/arterial-venous oxygen difference ratio, but not central venous oxygen saturation, predict increase in oxygen consumption in fluid responders

Combination of arterial lactate levels and venous-arterial CO 2 to arterial-venous O 2 content difference ratio as markers of resuscitation in patients with septic shock

e SOFA (sepsisrelated organ failure assessment) score to describe organ dysfunction/failure. On behalf of the working group on sepsis-related problems of the European Society of intensive care medicine

e validity of SOFA score to predict mortality in adult patients with cardiogenic shock on venoarterial extracorporeal membrane oxygenation

Serial evaluation of the SOFA score to predict outcome in critically ill patients

Central venous-to-arterial carbon dioxide difference combined with arterial-to-venous oxygen content difference is associated with lactate evolution in the hemodynamic resuscitation process in early septic shock