ORIGINAL ARTICLE |
https://doi.org/10.5005/jp-journals-10089-0075 |
Prognostic and Therapeutic Value of Venous to Arterial Carbon Dioxide Difference during Early Resuscitation in Critically Ill Nosocomial Septic Shock Patients
1,2Department of Critical Care Medicine, Manipal Hospital, Bengaluru, Karnataka, India
3Department of Critical Care Medicine, Aster Whitefield Hospital, Bengaluru, Karnataka, India
Corresponding Author: Mohan Kumar Narava, Department of Critical Care Medicine, Manipal Hospital, Bengaluru, Karnataka, India, Phone: +91 7042711090, e-mail: mohannarava1988@gmail.com
Received on: 25 April 2023; Accepted on: 22 June 2023; Published on: 30 October 2023
ABSTRACT
Introduction: Septic shock is a medical emergency. Various clinical biomarkers have been used to either prognosticate or use them for goal-directed management of the same. The venoarterial difference in the partial pressure of carbon dioxide (Pv-aCO2 gap) has been used as an alternate marker for assessing tissue hypoperfusion and to predicting mortality.
Aim: To determine the therapeutic and prognostic value of central venous to arterial carbon dioxide difference during early resuscitation of critically ill nosocomial septic shock patients.
Objectives: The primary objective was to predict 28-day mortality using the Pv-aCO2 gap. The secondary objectives were to compare the accuracy of lactate clearance, sequential organ failure assessment (SOFA) score against Pv-aCO2 gap as a predictor of 28-day mortality and to determine the association of fluid resuscitation and its effects on the Pv-aCO2 gap.
Materials and methods: A prospective observational cohort study was performed in a tertiary care intensive care unit (ICU). A total of 50 nosocomial septic shock patients were recruited. They are from either ward admissions or those who stayed in ICU beyond 48 hours. A Pv-aCO2 gap was measured serially over 0, 3, and 6 hours. Lactate clearance at 6 hours was measured. SOFA score on days 1 and 2 of admission, fluid resuscitation in the first 6 hours, and cumulative fluid balance over 24 hours and 7 days were calculated. The patients were divided into survivors and nonsurvivors according to the outcome at 28 days. Pv-aCO2 gap was assessed in both groups. The receiver operating characteristic (ROC) curve was plotted to analyze the prognostic value of these variables in predicting 28-day mortality. Data analysis was carried out using the Statistical Package for the Social Sciences (SPSS) version 18.5 package.
Results: The median values of the Pv-aCO2 gap had progressively increased in nonsurvivors (7.17, 7.70, and 8.06 mm Hg) over 0, 3, and 6 hours, respectively, whereas it progressively narrowed (6.84, 6.45, and 6.03 mm Hg) in survivors during the first 6 hours of the resuscitation period. Persistently high Pv-aCO2 gap at the end of 6 hours of resuscitation was observed in nonsurvivors, which were statistically significant (85.3 vs 43.8%, p = 0.004). Survivors and nonsurvivors received a mean crystalloid volume of 1430.8 ± 431.6 mL, irrespective of their Pv-aCO2 gap of < or >6 mm Hg. The discriminatory capacity at predicting 28-day mortality for SOFA score on days 1 and 2, lactate clearance at 6 hours, and Pv-aCO2 gap at 0, 3, and 6 hours were compared. ROC curve analysis showed that SOFA scores on days 1 and 2, lactate clearance at 6 hours, and Pv-aCO2 gap at 3 and 6 hours had predictive value to prognosticate 28-day mortality. The area under the ROC curve (AUROC) for SOFA score on days 1 and 2 was 0.907 [95% confidence interval (CI) was 0.791–0.971, p < 0.001] and 0.943 (95% CI was 0.839–0.989, p < 0.001) respectively. The AUROC for lactate clearance at 6 hours was 0.938 (95% CI was 0.743–0.947, p < 0.001). AUROC for Pv-aCO2 gap values at 3 and 6 hours were 0.814 (95% CI was 0.679–0.910, p < 0.001) and 0.865 (95% CI was 0.738–0.945, p < 0.001), respectively.
Conclusion: Persistent high Pv-aCO2 gap can be used as a prognostic marker for predicting 28-day mortality in nosocomial septic shock patients. Pv-aCO2 gap at 6 hours has almost the same discriminatory capacity as SOFA score on days 1 and 2, and lactate clearance at predicting 28-day mortality. More studies are required to ascertain the value of Pv-aCO2 gap values in estimating the adequacy of fluid resuscitation in nosocomial septic shock patients.
How to cite this article: Narava MK, Gopaldas JA, Gupta KVV. Prognostic and Therapeutic Value of Venous to Arterial Carbon Dioxide Difference during Early Resuscitation in Critically Ill Nosocomial Septic Shock Patients. J Acute Care 2023;2(2):46–53.
Source of support: Nil
Conflict of interest: Dr Justin A Gopaldas is associated as the Editorial Board member of this journal and this manuscript was subjected to this journal’s standard review procedures, with this peer review handled independently of this editorial board member and his research group.
Keywords: Central venous to arterial carbon dioxide difference, Fluid resuscitation, Nosocomial sepsis, Sepsis, Septic shock, Venoarterial difference in the partial pressure of carbon dioxide gap, 28-day mortality
INTRODUCTION
Septic shock is associated with considerable morbidity and mortality. Indian literature notes gram-negative microbes as the most common cause of sepsis, with hospital mortality, 28-day mortality, and severe sepsis-attributable mortality being 63.6, 62.8, and 85%, respectively.1,2 The Surviving Sepsis Campaign (SSC) reemphasizes bundles of care to address this medical emergency, with resuscitation being the immediate task. Nosocomial, in comparison to community-acquired sepsis, is associated with increased mortality, longer length of hospital stay, and greater healthcare costs and is typically caused by antibiotic-resistant pathogens. Initial admission diagnosis (septic or nonseptic) and the severity of illness influence the outcome.3 Despite the differences in the patient characteristics (respiratory and bloodstream infections being common with nosocomial sepsis), nosocomial septic shock is managed with the same principles utilized for community sepsis as per SSC. Compliance with these guidelines for nosocomial sepsis is poorly investigated, and where it is audited, it shows poor compliance.2,4-6
The goals of resuscitation are to optimize intravascular volume, improve tissue perfusion, and, in turn, tissue oxygenation with the hope of limiting or reversing organ dysfunction. Many goal-directed therapies are undertaken in this regard to assist optimal resuscitation [e.g., early-goal-directed therapy (EGDT) with central venous oxygen saturation (ScvO2), lactate, pressure variables, etc.]. Though clinician-driven care is adequate for improved outcomes compared to EGDT, over the years, clinicians have become accustomed to using various parameters alone or in combination to guide their resuscitative endpoints.6,7 Such practice is predominantly undertaken in emergency rooms and later continued in intensive care unit (ICU) where utilization of lactate and later dynamic indices of circulation to assess for adequacy of resuscitation remain the cornerstone.8 With the ever-increasing use of point-of-care ultrasound, nosocomial septic shock is predominantly assessed and managed with dynamic indices [inferior vena cava (IVC) variability > velocity time integral (VTi) variability and pulse pressure variation (PPV)/stroke volume variation (SVV)]. Lactate and base excess come second best in these circumstances.9
When evidence is compiled, it is mounting favorably for markers of peripheral perfusion (temperature, mottling, capillary refill time (CRT), sublingual microcirculation, etc.) but falls short in their ability to prognosticate or be utilized for adequacy of resuscitation. The clinical utility of CRT is low in the ICU, unlike in the emergency department, in part due to shock more likely to be mixed in a critically ill patient. Markers of central perfusion (ScvO2 and lactate) driven resuscitation have not been able to improve outcomes in various studies. Though lactate is used extensively in emergency departments, it lacks specificity as a marker of dysoxia. While a ScvO2 at >80 or <50% is more indicative of severe alteration in oxygen extraction with poor correlation for values in between. This brings us to discuss the novel biomarker venoarterial difference in the partial pressure of carbon dioxide (Pv-aCO2) gap as another marker of central perfusion.10
Venoarterial difference in the partial pressure of carbon dioxide (Pv-aCO2) gap is a more practical measure, as CO2 is more diffusible than O2 and reflects global tissue blood flow relative to metabolic demand. The association between the Pv-aCO2 gap and sepsis is dependent on myocardial contractility, volume status, and tissue perfusion. Studies show a correlation between the increased gap and microcirculatory impairment.11,12 A Pv-aCO2 gap >6 mm Hg indicates persistent shock and may benefit from fluid or inotropic support. High Pv-aCO2 gap postresuscitation is consistently shown to be associated with increased mortality. Based on similar research in the care of community sepsis, a “ScvO2–central venous-to-arterial CO2 (CvaCO2) gap-guided protocol could be used for septic shock management.”11
The majority of studies assessing the clinical value of the Pv-aCO2 gap have focused on its therapeutic and prognostic potential in patients with septic shock who were admitted from emergency departments, with a significant proportion of these cases attributed to community-acquired sepsis. We wish to evaluate its prognostic power in nosocomial sepsis and also indirectly note its effectiveness in identifying the adequacy of fluid resuscitation.
MATERIALS AND METHODS
The prospective observational cohort study was conducted in the multidisciplinary tertiary care ICU in South India, where 50 patients suspected of or diagnosed with nosocomial infection and septic shock (acquired either in the hospital wards or ICU) were recruited. Participants were followed up until death or ICU discharge, and if discharged earlier than 28 days, their status at 28 days was determined by phone. The study passed the scrutiny of the institutional ethics and scientific committee. The study utilized a de-identified method of data collection, and it did not involve any intervention based on the values collected.
Patients aged 18 or above diagnosed with septic shock were included in the study, and those who were admitted with pregnancy, liver cirrhosis, advanced metastatic cancers, moribund patients, do not attempt resuscitation order, those without appropriate central venous and direct arterial catheter access, shock due to bleeding from injury or surgery and admission with acute myocardial infarction were excluded from the study.
Data collection was carried out in accordance with the hospital sepsis protocol for each patient, including a detailed history and a focused physical examination. The sequential organ failure assessment (SOFA) score was calculated at admission and 24 hours, and routine laboratory investigations were performed in accordance with the unit protocol. Arterial blood gas and central venous samples were obtained simultaneously at 0, 3, and 6 hours of admission, and the Pv-aCO2 gap was calculated in all patients, irrespective of spontaneous or mechanical ventilation. Lactate levels were measured at 0, 3, and 6 hours, and lactate clearance was calculated. Blood and urine cultures were collected on admission, and chest X-rays and an electrocardiogram were performed for every patient. Septic shock was diagnosed in patients with systolic blood pressure of <90 mm Hg as those suspected or confirmed to have infection along with 2 or more criteria of sytemic inflammatory response syndrome (SIRS) and after excluidng obstructive and cardiogenic causes.
Resuscitation was carried out at the discretion of the treating clinician with the use of biomarkers and dynamic indices (lactate, base excess, IVC variability index, VTi respiratory variation, PPV, and SVV).
Other parameters, such as the need for mechanical ventilation and its duration, positive fluid balance at 24 hours and 7 days, level of vasopressors required, days of vasopressor needed, need for renal replacement therapy (RRT), time to antibiotics, and days of ICU stay, were recorded. The cause of healthcare-associated infection was noted in accordance with the Centers for Disease Control and Prevention definition. The maximal vasoactive-inotropic score (VISmax) was used to calculate the maximum level of vasopressors used during patient care. The patients were retrospectively classified into two groups, the survivors and nonsurvivors group, and the Pv-aCO2 difference was evaluated in the two groups.
Statistical Analysis
This study employed statistical analyses, including Chi-squared tests for proportions and student’s t-test for continuous data to compare the study groups. Results are presented in a table and figure, which display numbers, percentages, mean, and standard deviation (SD) as appropriate for each parameter. A receiver operating characteristic (ROC) curve analysis was used to evaluate the accuracy of SOFA, Pv-aCO2, and lactate clearance in predicting mortality. Sensitivity, specificity, positive predictive value, and negative predictive value were determined for each parameter. A p-value of <0.05 was considered statistically significant. Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS) version 18.5.
RESULTS
The majority of patients (64%) were male and aged between 50 and 79 years. Central line-associated bloodstream infection (CLABSI) was seen in 52%, and hospital/ventilator-associated pneumonia (HAP/VAP) in 30% of the study cohort. 60% of patients had two or more comorbidities, with cardiovascular disease (CVD) (62%) and metabolic disorders (82%) being the most prevalent. The mean age of the study population was 57.9 ± 12.55 years, with a mean SOFA score on day 1 of 11.34 ± 2.44 and on day 2 of 11.58 ± 2.80. The mean length of stay (LOS) in the ICU was 19.2 ± 13.9 days, with mean durations of invasive mechanical ventilation and vasopressor support of 13.3 ± 10.53 and 6.3 ± 3.51 days, respectively. (Table 1). Patients were categorized as survivors or nonsurvivors based on their outcome status at 28 days of ICU admission. Of the patients, 68% were classified as nonsurvivors (Table 1). Two patients were identified to have grade I/II pressure ulcers, constituting approximately 4% of our study population.
| Variable | Survivors | Nonsurvivors | p-value* |
|---|---|---|---|
| Age (mean±) | 57 ± 13.2 | 58.4 ± 12 | 0.35 |
| Gender (M/F) | 9/7 | 23/11 | 0.22 |
| Comorbidities | |||
| CVD | 8 | 23 | 0.12 |
| Metabolic disorder | 13 | 28 | 0.14 |
| Cerebrovascular disease | 2 | 5 | 0.4 |
| CKD | 1 | 3 | 0.38 |
| Chronic respiratory disease | 0 | 6 | 0.03 |
| SOFA (mean±) | 8.9 ± 1.28 | 12.4 ± 2.0 | 0.00001* |
| Diagnosis | |||
| CLABSI | 7 | 19 | 0.4 |
| CAUTI | 4 | 5 | 0.37 |
| HAP/VAP | 4 | 11 | 0.6 |
| Antibiotics (under 1 hour) | 15 | 34 | 0.07 |
| Initial fluid resuscitation volume (mean± in mL) | 1518 ± 260 | 1390 ± 121 | 0.17 |
| Organ support requirement | |||
| Mechanical ventilation | 13 | 34 | 0.004* |
| Vasopressor support | 16 | 34 | 0.5 |
| RRT | 6 | 29 | 0.002* |
| Fluid balance at day 7 (mean± in mL) | 5014 ± 1473 | 10039 ± 1772 | 0.00001* |
| ICU/LOS (mean±) | 24.2 ± 21.06 | 16.9 ± 8.39 | 0.04 |
| 28-day mortality (%) | 32 | 68 | |
*Significance of p-value; CAUTI, catheter-associated urinary tract infections; CKD, chronic kidney disease; CLABSI, central line-associated bloodstream infections; CVD, cardiovascular disease; HAP, hospital-acquired pneumonia; ICU/LOS, intensive care unit/length of stay; RRT, renal replacement therapy;SOFA, sequential organ failure assessment; VAP, ventilator-associated pneumonia
Survivors had lower mean SOFA scores on days 1 and 2 (8.9 ±1.28 and 8.56 ± 1.96, respectively) compared to nonsurvivors (12.4 ± 2.0 and 13.00 ± 1.85, respectively) (Table A1). There was a statistical difference between the two groups as the SOFA score was higher in nonsurvivors compared to survivors on days 1 and 2.
Critical coronavirus disease of 2019 (COVID-19) pneumonia was the admission diagnosis in 38% of patients with high mortality (84.2%). CLABSI was the most common infection in the study cohort. nonsurvivors had a higher incidence of CLABSI (61.4%), with multidrug-resistant organism (MDRO) Klebsiella pneumoniae as the most prevalent isolate (33.3%) (Table A2). Survivors required less duration (mean number of days: 4.1 ± 2.55 vs 7.3 ± 3.47) and quantity of vasoactive medications (mean VISmax: 24.2 ± 7.94 vs 46.8 ± 8.34) compared to nonsurvivors. While 50% of survivors only required a single vasopressor, >75% of nonsurvivors required three vasopressors (Tables A3 and A4). Mechanical ventilation days (15.6 ± 16.44 and 12.2 ± 6.16) and ICU LOS (24.2 ± 21.06 and 16.9 ± 8.39) were not statistically significant between survivors and nonsurvivors.
Initial resuscitation volume and volume at 6 hours from onset of sepsis were not different between survivors and nonsurvivors, but significant differences were observed in the mean cumulative fluid balance at day 7 days (5014.1 ± 1472.5 vs 10038.7 ± 1771.5 mL, p < 0.001) (Table 1).
Evaluation of the Prognostic Accuracy of the Pv-aCO2 Gap in Predicting Mortality
This study evaluated the prognostic accuracy of serial Pv-aCO2 gap measurements at 0, 3, and 6 hours in predicting mortality. The mean Pv-aCO2 gap at 0 hours did not differ significantly between survivors and nonsurvivors. However, at 3 and 6 hours, a significant difference was observed between the two groups. The evolution of the Pv-aCO2 gap over 6 hours of fluid resuscitation showed a significant difference between survivors and nonsurvivors. Survivors had a decreasing trend, while nonsurvivors had an increasing trend (Table A5 and Fig. A1).
The percentage of survivors with Pv-aCO2 gap of >6 mm Hg at 0, 3, and 6 hours was 75, 62.5, and 43.8%, respectively. The percentage of nonsurvivors with Pv-aCO2 gap >6 mm Hg at 0, 3, and 6 hours was 73.5, 85.3, and 85.3%, respectively (Table 2). A statistically significant difference between the two groups was observed at 6 hours. The study utilized ROC curve analysis to evaluate the discriminatory capacity of the Pv-aCO2 gap in predicting mortality. Pv-aCO2 gap values above certain cutoff values were predictive of mortality at different time points. Specifically, a Pv-aCO2 gap of 5.3 mm Hg at 0 hours, 7.3 mm Hg at 3 hours, and 8.0 mm Hg at 6 hours were associated with increased mortality. Sensitivity and specificity varied depending on the time point. ROC analysis showed that the Pv-aCO2 gap was a good predictor of mortality at 3 and 6 hours (Fig. 1).
| Outcome | Total (N = 50) | p-value* | ||||||
|---|---|---|---|---|---|---|---|---|
| Nonsurvivors (N = 34) | Survivors (N = 16) | |||||||
| Hour | Pv-aCO2 gap | n | % | n | % | n | % | |
| 0 hour | ≤6.0 | 9 | 26.5% | 4 | 25.0% | 13 | 26.0% | 0.600 |
| >6.0 | 25 | 73.5% | 12 | 75.0% | 37 | 74.0% | ||
| 3 hours | ≤6.0 | 5 | 14.7% | 6 | 37.5% | 11 | 22.0% | 0.076 |
| >6.0 | 29 | 85.3% | 10 | 62.5% | 39 | 78.0% | ||
| 6 hours | ≤6.0 | 5 | 14.7% | 9 | 56.3% | 14 | 28.0% | 0.004 |
| >6.0 | 29 | 85.3% | 7 | 43.8% | 36 | 72.0% | ||
*Significance of p-value
Fig. 1: Discriminatory capacity of serial Pv-aCO2 gap levels in predicting mortality using ROC curve analysis
Comparing Outcome Associations of SOFA Score, Lactate Clearance, and Pv-aCO2 Gap
The study compared the association between the mean SOFA scores on day 1 and the Pv-aCO2 gap at 6 hours in critically ill patients. Among the 50 patients, 14 had a Pv-aCO2 gap of <6 mm Hg at 6 hours, with 35.7% nonsurvivors and a mean SOFA score of 11.0 ± 2.1. The remaining 36 patients had a Pv-aCO2 gap of >6 mm Hg at 6 hours, with 80.6% nonsurvivors and a mean SOFA score of 12.7 ± 1.9. It was statistically significant (Table A6).
This study analyzed the mean lactate clearance at 6 hours in relation to survival outcomes. The results demonstrated that survivors had a mean lactate clearance of 15.27 ± 11.81, while nonsurvivors had a mean lactate clearance of −15.31 ± 16.54 (Fig. A2). Statistical analysis revealed a significant difference between the two groups (p < 0.001).
Lactate clearance greater than 10% within the first 6 hours is considered an effective guide for fluid resuscitation.13 In this study, 11.8% of nonsurvivors had lactate clearance of >10%, while 62.5% of survivors had lactate clearance of >10% (Table A7), with a significant statistical difference observed between the two groups.
Assessment of Prognostic Value of SOFA Score, Pv-aCO2 Gap, and Lactate Clearance for Predicting 28-day Mortality
The study assessed the discriminatory capacity of the SOFA score on days 1 and 2, Pv-aCO2 gap at 0, 3, and 6 hours, and lactate clearance at 6 hours for predicting 28-day mortality in critically ill patients.
The results showed that the SOFA score at day 1 had a cutoff value of >10 and an area under the ROC curve (AUROC) of 0.907 [p < 0.0001, 95% confidence interval (CI) = 0.791–0.971], and the SOFA score at day 2 had a cutoff value of >10 and an AUROC of 0.943 (p < 0.0001, 95% CI = 0.839–0.989), indicating the good discriminatory capacity in predicting 28-day mortality. The Pv-aCO2 gap at 0 hours had a cutoff value of >5.3 mm Hg and an AUROC of 0.579 (p = 0.37, 95% CI = 0.431–0.717), indicating the poor discriminatory capacity in predicting 28-day mortality. However, the Pv-aCO2 gap at 3 hours had a cutoff value of >7.3 mm Hg and an AUROC of 0.814 (p < 0.0001, 95% CI = 0.679–0.910), and the Pv-aCO2 gap at 6 hours had a cutoff value of >8.0 mm Hg and an AUROC of 0.865 (p <0.0001, 95% CI = 0.738–0.945), indicating the good discriminatory capacity in predicting 28-day mortality. Lactate clearance at 6 hours had a cutoff value of <3% and an AUROC of 0.938 (p < 0.0001, 95% CI = 0.743–0.947), indicating a good discriminatory capacity in predicting 28-day mortality (Fig. 2).
Fig. 2: Discriminatory capacity for different parameters to predict mortality using ROC analysis
DISCUSSION
The pv-aCO2 gap could predict nosocomial sepsis outcomes similar to its ability in community sepsis data based on the findings of the study. Study also noted motality prediction of Pv-aCO2 gap to be simialr to SOFA and delta lactate (changes over 6 hours) in community sepsis.14,15
The study population belonged to a tertiary care ICU with an average admission rate of 7.2 per day and with a mean Acute Physiology and Chronic Health Evaluation II score of 14.5. From this group, 50 consecutive nosocomial sepsis patients were identified, and data were collected prospectively. This group had a median duration from hospital or ICU admission to index sepsis deemed nosocomial of 10 days (mean 17.6 ± 19 days), which is higher compared to community sepsis.3 Consistent with the available literature, mean SOFA scores were significantly higher in nonsurvivors compared to survivors (12.47 ± 2.00 vs 8.94 ± 1.28).16 The type and incidence of infection are consistent with the epidemiology and outcomes of hospital-acquired bloodstream infections in ICU patients similar to the EUROBACT-2 international cohort study trial with MDRO (e.g., multidrug-resistant, extensively drug-resistant, and pandrug resistant gram-negative bacteria) sepsis accounting for 70% of infections. MDRO gram-positive sepsis and fungal sepsis accounted for 18 and 12%, respectively.4
The Pv-aCO2 gap increased in nonsurvivors (7.17–8.06 mm Hg) but reduced in survivors (6.84–6.03 mm Hg) during the initial 6 hours of resuscitation phase.17,18 In our current study, a high Pv-aCO2 gap (>6 mm Hg) at the end of 6 hours of resuscitation was associated with a significant increase in mortality (85.3% mortality rate in patients with a high Pv-aCO2 gap, n = 29/34, p < 0.001) as noted with other studies.19 Similarly, SOFA scores on day 1 [area under the ROC curve (AUROC) 0.907] and day 2 (AUROC 0.943) had higher discriminatory capacity for 28-day mortality. When the Pv-aCO2 gap at 6 hours is tested against another biomarker of central perfusion, like lactate at 6 hours, it is shown that both are deemed comparable in their prediction of mortality. Similar results were found in nonnosocomial sepsis studies. Lactate could still be used in nosocomial septic shock resuscitation as part of prognosticating the illness, but it is unclear if it can be used to limit resuscitation volume.12,20,21
Fluid resuscitation was assessed in the study through the quantity of fluid administered during the initial 6 hours from the time of initial bolus and cumulative fluid balance on days 1 and 7. There is no statistical difference between survivors and nonsurvivors with respect to initial resuscitative volume (552 vs 625 mL: nonsignificant) and similarly at the end of 6 hours since initiation of resuscitation (1390 vs 1517 mL: nonsignificant). This represented a practice of graded administration of fluid rather than preset quantity. The initial volume ordered, ranging between 500 and 1000 mL of crystalloid with concurrent vasoactive medication use, is consistent with personalised strategy of fluid resuscitation. An initial crystalloid fluid bolus of around 500 mL and reassess prior to escalating more fluid loading contrary to a standard prescription of 30 mL/kg at diagnosis as proposed by surviving sepsis guidance for the care of all septic shock patients.22
Cumulative fluid balance showed no significant difference in volume at the end of 24 hours between survivors and nonsurvivors (mean volumes of 2217 vs 2319 mL). The optimum amount of fluid required for early resuscitation of septic shock is between 15 and 45 mL/kg (U-shaped curve of dose-response). The longer the time from initial sepsis assessment and fluid administration, the lower the blood pressure response for further fluid administration.23 Our study clearly noted initial resuscitation volume during the first 6 hours to be much lower than the amount expected by the SSC guidance. This could be due to the higher use of dynamic indices to guide the resuscitation phase. Early use of an escalated dose of vasopressors in this group is likely the practice that may influence the amount of fluid administered. The limit on the volume is not just due to the improved pressure indices but partly because vasoactive medications alter dynamic indices and cardiac output.24 Nearly all patients had vasopressor demand, which is defined as refractory septic shock (>0.5 µg/kg/minute). Concomitant use of vasopressors shows that either the number of vasopressors required or the VISmax score in itself reflects a significant difference between survivors and nonsurvivors. Arterial pressure and resuscitation optimization do not equate to adequacy of perfusion. When we read the above interaction of SOFA, lactate change, and Pv-aCO2 curves along with fluid resuscitation volumes and vasopressor demand at 6 hours, it is clear that mortality is driven predominantly by the severity of illness but not due to lack of resuscitation. Lack of perfusion correction after pressure and resuscitation optimization is hence considered to be a marker of mortality.18
The strength of the study is it being a prospective and pragmatic resuscitation strategy that is assessed against the prognostic and potential therapeutic value of Pv-aCO2 levels in the phenotype of nosocomial septic shock patients. It is the first of its kind to document clinician-driven resuscitation of nosocomial septic shock in an Indian ICU and showed the practice pattern and the adequacy of the same. The results were similar to previous community sepsis studies and provided avenues for future research. The limitation of the study is it is a single-center and observational study conducted during the severe acute respiratory syndrome coronavirus 2 pandemic, leading to inclusion of greater than 35% of patients whose initial admission diagnosis was critical COVID pneumonia. Issues relevant to the pandemic, like higher nosocomial sepsis and mortality, may have influenced the outcome data. The study population represent extremely sick patients, with the majority suffering from refractory shock. Study group analysis noted that many nonsurvivors were admitted with critical COVID-19 pneumonia (>85% mortality) as their initial diagnosis. Resuscitation undertaken with the pragmatic use of dynamic indices resulted in lower volume of administration than expected from SCC guidelines. This could have been even less if stringent assessment at each step of administration was undertaken. But this also brings to the fore a phenotype of septic shock that is overlooked and still being treated along the lines of community sepsis. We were unable to follow up with the survivors for a period of over 4 weeks and couldn’t inform the long-term outcomes.
CONCLUSION
In this nosocomial sepsis management study, compliance with majority of SSC guidance (measure and remeasure lactate, administer early and appropriate antibiotics along with appropriate cultures, and initiate early vasopressor support) noted. Pragmatic critical care physician driven resuscitation of nosocimial septic shock with use of dynamic idices has resulted in significantly less crytalloid use comapred to 30 mL/kg. Nosocomial septic shock, hence, could be a subset that has a low preload deficit and a higher propensity for refractory shock despite adequate fluid resuscitation.
Disease severity rather than the adequacy of the resuscitation has defined mortality. The persistently high Pv-aCO2 gap after adequate resuscitation of nosocomial sepsis can identify the risk of death with a discriminatory capacity similar to the SOFA score and lactate change at 6 hours. However, further studies are needed to establish the value of adding the Pv-aCO2 gap in the armamentarium of the treating critical care physician as a surrogate of central perfusion marker when judging resuscitation adequacy in septic shock patients and particular nosocomial septic shock.
Appendix
| Quartiles | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| N | Mean | SD | Minimum | Maximum | Q1 (25%) | Q2 (50%) | Q3 (75%) | p-value | ||
| SOFA (Day 1) | Nonsurvivors | 34 | 12.47 | 2.004 | 8 | 15 | 11.00 | 13.00 | 14.00 | <0.001 |
| Survivors | 16 | 8.94 | 1.289 | 7 | 11 | 8.00 | 9.00 | 10.00 | ||
| Total | 50 | 11.34 | 2.446 | 7 | 15 | 9.00 | 11.00 | 14.00 | ||
| SOFA (Day 2) | Nonsurvivors | 34 | 13.00 | 1.859 | 9 | 16 | 11.00 | 14.00 | 14.00 | <0.001 |
| Survivors | 16 | 8.56 | 1.965 | 5 | 12 | 7.00 | 9.00 | 10.00 | ||
| Total | 50 | 11.58 | 2.807 | 5 | 16 | 10.00 | 12.00 | 14.00 | ||
| Organism isolated | Number | Percentage |
|---|---|---|
| Acinetobacter baumannii | 12 | 24 |
| Klebsiella pneumoniae | 12 | 24 |
| Escherichia coli | 6 | 12 |
| Pseudomonas aeruginosa | 3 | 6 |
| Vancomycin-resistant Enterococcus | 4 | 8 |
| Enterobacter aerogenes | 1 | 2 |
| Candida tropicalis | 3 | 6 |
| Candida glabrata | 2 | 4 |
| Candida auris | 1 | 2 |
| Not isolated | 6 | 12 |
| Total | 50 | 100 |
| Number of vasopressors | Outcome | Total (N = 50) | p-value* | |||||
|---|---|---|---|---|---|---|---|---|
| Nonsurvivors | Survivors (N = 16) | |||||||
| Vasopressor support | 1 | 0 | 0.0% | 8 | 50.0% | 8 | 16.0% | <0.001 |
| 2 | 8 | 23.5% | 8 | 50.0% | 16 | 32.0% | ||
| 3 | 26 | 76.5% | 0 | 0.0% | 26 | 52.0% | ||
| N | Mean VISmax | SD | Minimum VIS max | Maximum VISmax | p-value* | |
|---|---|---|---|---|---|---|
| Nonsurvivors | 34 | 46.8 | 8.3429 | 26.0 | 56.0 | <0.001 |
| Survivors | 16 | 24.2 | 7.9493 | 16.0 | 45.0 | |
| Total | 50 | 39.6 | 13.4278 | 16.0 | 56.0 |
*Significance of p-value
| N | Mean | SD | Minimum | Maximum | p-value | ||
|---|---|---|---|---|---|---|---|
| 0 hour | Nonsurvivors | 34 | 7.17 | 1.222 | 5.1 | 9.6 | 0.380 |
| Survivors | 16 | 6.84 | 1.277 | 4.8 | 9.0 | ||
| Total | 50 | 7.06 | 1.237 | 4.8 | 9.6 | ||
| 3 hours | Nonsurvivors | 34 | 7.70 | 1.238 | 4.9 | 10.0 | 0.001 |
| Survivors | 16 | 6.45 | 0.823 | 4.9 | 7.8 | ||
| Total | 50 | 7.30 | 1.260 | 4.9 | 10.0 | ||
| 6 hours | Nonsurvivors | 34 | 8.06 | 1.387 | 5.1 | 9.7 | <0.001 |
| Survivors | 16 | 6.03 | 1.191 | 4.0 | 8.0 | ||
| Total | 50 | 7.41 | 1.625 | 4.0 | 9.7 |
| Pv-aCO2 gap (6 hours) (mm Hg) | Outcome | Number | Total | Mean SOFA score (day 1) | p-value |
|---|---|---|---|---|---|
| ≤6.0 | Survivors | 9 (64.3%) | 14 (100%) | 8.6 ± 1.130 | 0.014 |
| Nonsurvivors | 5 (35.7%) | 11.0 ± 2.121 | |||
| >6.0 | Survivors | 7 (19.4%) | 36 (100%) | 9.43 ± 1.397 | <0.001 |
| Nonsurvivors | 29 (80.6%) | 12.72 ± 1.907 |
| Hour | Lactate clearance (%) | Outcome | Total | p-value | ||||
|---|---|---|---|---|---|---|---|---|
| Nonsurvivors (N = 34) | Survivors (N = 16) | |||||||
| 6 hours | <10 | 30 | 88.2% | 6 | 37.5% | 36 | 72.0% | <0.001 |
| ≥10 | 4 | 11.8% | 10 | 62.5% | 14 | 28.0% | ||
Fig. A1: Serial Pv-aCO2 gap measurements between two study groups in comparison with fluid resuscitation at 0, 3, and 6 hours
Fig. A2: Comparison of mean lactate clearance (%) values among the study groups
ORCID
Mohan Kumar Narava https://orcid.org/0009-0002-9022-2100
Justin A Gopaldas https://orcid.org/0000-0002-0420-6925
Venkatesha Gupta K V https://orcid.org/0000-0003-4344-3840
REFERENCES
1. Chatterjee S, Bhattacharya M, Todi SK. Epidemiology of adult-population sepsis in India: a single center 5 year experience. Indian J Crit Care Med 2017;21(9):573–577. DOI: 10.4103/ijccm.IJCCM_240_17
2. Markwart R, Saito H, Harder T, et al. Epidemiology and burden of sepsis acquired in hospitals and intensive care units: a systematic review and meta-analysis. Intensive Care Med 2020;46(8):1536–1551. DOI: 10.1007/s00134-020-06106-2
3. Paoli CJ, Reynolds MA, Sinha M, et al. Epidemiology and costs of sepsis in the United States-an analysis based on timing of diagnosis and severity level. Crit Care Med 2018;46(12):1889–1897. DOI: 10.1097/CCM.0000000000003342
4. Tabah A, Buetti N, Staiquly Q, et al. Epidemiology and outcomes of hospital-acquired bloodstream infections in intensive care unit patients: the EUROBACT-2 international cohort study. Intensive Care Med 2023;49(2):178–190. DOI: 10.1007/s00134-022-06944-2
5. Nassar AP Jr, Pires-Neto RC, de Carvalho WB, et al. Characteristics and outcomes of patients with community-acquired and hospital-acquired sepsis: a multicenter cohort study in Brazil. Rev Bras Ter Intensiva 2019;31(2):167–177. DOI: 10.5935/0103-507X.20190013
6. Rhodes A, Phillips G, Beale R, et al. The surviving sepsis campaign bundles and outcome: results from the international multicentre prevalence study on sepsis (the IMPreSS study). Intensive Care Med 2015;41(9):1620–1628. DOI: 10.1007/s00134-015-3906-y
7. Liu B, Ding X, Yang J. Effect of early goal directed therapy in the treatment of severe sepsis and/or septic shock. Curr Med Res Opin 2016;32(11):1773–1782. DOI: 10.1080/03007995.2016.1206872
8. Uffen JW, Oosterheert JJ, Schweitzer VA, et al. Interventions for rapid recognition and treatment of sepsis in the emergency department: a narrative review. Clin Microbiol Infect 2021;27(2):192–203. DOI:10.1016/j.cmi.2020.02.022
9. Pellegrini JAS, Cordioli RL, Grumann ACB, et al. Point-of-care ultrasonography in Brazilian intensive care units: a national survey. Ann Intensive Care 2018;8(1):50. DOI: 10.1186/s13613-018-0397-3
10. Hasanin A, Mukhtar A, Nassar H. Perfusion indices revisited. J Intensive Care 2017;5:24. DOI: 10.1186/s40560-017-0220-5
11. Vallet B, Pinsky MR, Cecconi M. Resuscitation of patients with septic shock: please ”mind the gap”! Intensive Care Med 2013;39(9):1653–1655. DOI: 10.1007/s00134-013-2998-5
12. Mallat J, Pepy F, Lemyze M, et al. Central venous-to-arterial carbon dioxide partial pressure difference in early resuscitation from septic shock: a prospective observational study. Eur J Anaesthesiol 2014;31(7):371–380. DOI: 10.1097/EJA.0000000000000064
13. Lyu X, Xu Q, Cai G, et al. Efficacies of fluid resuscitation as guided by lactate clearance rate and central venous oxygen saturation in patients with septic shock. Zhonghua Yi Xue Za Zhi 2015;95(7):496–500. PMID: 25916923.
14. Khwannimit B, Bhurayanontachai R, Vattanavanit V. Comparison of the performance of SOFA, qSOFA and SIRS for predicting mortality and organ failure among sepsis patients admitted to the intensive care unit in a middle-income country. J Crit Care 2018;44:156–160. DOI: 10.1016/j.jcrc.2017.10.023
15. Ahmed W, Laimoud M. The value of combining carbon dioxide gap and oxygen-derived variables with lactate clearance in predicting mortality after resuscitation of septic shock patients. Crit Care Res Pract 2021;2021:6918940. DOI: 10.1155/2021/6918940
16. Bitar ZI, Maadarani OS, El-Shably AM, et al. The forgotten hemodynamic (PCO2 Gap) in severe sepsis. Crit Care Res Pract 2020;2020:9281623. DOI: 10.1155/2020/9281623
17. Ospina-Tascón GA, Bautista-Rincón DF, Umaña M, et al. Persistently high venous-to-arterial carbon dioxide differences during early resuscitation are associated with poor outcomes in septic shock. Crit Care 2013;17(6):R294. DOI: 10.1186/cc13160
18. Helmy TA, El-Reweny EM, Ghazy FG. Prognostic value of venous to arterial carbon dioxide difference during early resuscitation in critically ill patients with septic shock. Indian J Crit Care Med 2017;21(9):589–593. DOI: 10.4103/ijccm.IJCCM_64_16
19. Shaban M, Salahuddin N, Kolko MR, et al. The predictive ability of PV-ACO2 gap and PV-ACO2/CA-VO2 ratio in shock: a prospective, cohort study. Shock 2017;47(4):395–401. DOI: 10.1097/SHK.0000000000000765
20. Marty P, Roquilly A, Vallée F, et al. Lactate clearance for death prediction in severe sepsis or septic shock patients during the first 24 hours in intensive care unit: an observational study. Ann Intensive Care 2013;3(1):3. DOI: 10.1186/2110-5820-3-3
21. Al Duhailib Z, Hegazy AF, Lalli R, et al. The use of central venous to arterial carbon dioxide tension gap for outcome prediction in critically ill patients: a systematic review and meta-analysis. Crit Care Med 2020;48(12):1855–1861. DOI: 10.1097/CCM.0000000000004578
22. Marik PE, Byrne L, van Haren F. Fluid resuscitation in sepsis: the great 30 mL per kg hoax. J Thorac Dis 2020;12(Suppl 1):S37–S47. DOI:10.21037/jtd.2019.12.84
23. Lat I, Coopersmith CM, De Backer D, et al. The surviving sepsis campaign: fluid resuscitation and vasopressor therapy research priorities in adult patients. Crit Care Med 2021;49(4):623–635. DOI:10.1097/CCM.0000000000004864
24. Meng L, Tran NP, Alexander BS, et al. The impact of phenylephrine, ephedrine, and increased preload on third-generation Vigileo-FloTrac and esophageal doppler cardiac output measurements. Anesth Analg 2011;113(4):751–757. DOI: 10.1213/ANE.0b013e31822649fb
________________________
© The Author(s). 2023 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and non-commercial reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.