European Journal of Cardiovascular Medicine

Prognostication and outcome prediction in ICU patient care  has traditionally been based on subjective and biochemical judgment by clinicians. However, the global expansion of intensive care units (ICUs) has necessitated the use of objective, quantitative, and clinically relevant measures to evaluate treatment effectiveness (1). Mortality remains the most sensitive, reliable, and meaningful measurement of outcome for ICU patients.

Tissue hypoxia is a significant factor leading to organ failure and increased mortality in critically ill ICU patients (2, 3). Despite advancements in understanding sepsis-related biochemical processes and clinical interventions, mortality remains at approximately 40% (4). Sepsis-induced hypoperfusion results in tissue hypoxia, cellular dysfunction, and ultimately multi-organ failure (5). Early and appropriate intervention—such as fluids and vasoactive drugs—may prevent this cycle and restore adequate oxygen delivery to cells, potentially allowing recovery of organ function (6). Recognizing and addressing signs of persistent tissue hypoperfusion is crucial, particularly in patients with compromised physiological reserves (7, 8).

When oxygen delivery is insufficient to meet demand in critically ill patients, a global tissue hypoxia and oxygen debt ensue. This leads to anaerobic metabolism and the production of lactate, a complex marker of microcirculation (9). Lactate levels depend on both its production and its metabolism, both of which are impaired in sepsis (10). An elevated lactate concentration is a sign of anaerobic metabolism and tissue hypoxia. Recent studies indicate that lactate clearance is significantly reduced in septic patients, even when hemodynamics are stable and liver function remains intact. This may be due to decreased pyruvate dehydrogenase activity . Increased lactate production or decreased clearance results in hyperlactatemia, with studies showing that faster lactate clearance correlates with better patient outcomes, and patients with rapid clearance are more likely to survive (11, 12, 13, 14). Lactate clearance is calculated as follows:

Lactate Clearance = Lactate Initial -  Lactate Later / Lactate Initial X 100

Another significant marker of metabolic disturbance due to inadequate perfusion is the veno-arterial difference in partial pressure of carbon dioxide (PCO2 GAP). The PCO2 gap is increasingly recognized as a reliable tool for assessing tissue perfusion and predicting poor outcomes in patients with circulatory shock. It reflects the relationship between cardiac output (CO) and global metabolic demand, indicating the adequacy of venous blood flow in eliminating CO2 produced by peripheral tissues (15). Under stable conditions, the PCO2 gap is influenced by the veno-arterial CO2 content difference, the CO2 dissociation curve, CO, and alveolar ventilation, with CO having the greatest impact (16).

Several scoring systems, such as APACHE II, have been developed to assess illness severity in critically ill patients. While moderately accurate in predicting individual survival, these systems are valuable for quality-of-care assessment and research, as they allow for outcome comparisons among groups of critically ill patients with similar disease severity. APACHE II (Acute Physiology and Chronic Health Evaluation II) is among the most widely used scoring systems. Developed in 1985, it assigns a score between 0 and 71 based on a variety of clinical and  arboratory parameters, including age, chronic health status, oxygenation, and GCS (17). A higher score indicates more severe disease and a greater risk of death (18).

One study based on pediatric critically ill patient showed us Lactate clearance > 16.5 at 6 hour of admission predicted high mortality rate.(19)

Aims and objectives  To assess and compare the prognostic value of the PCO2 gap, blood lactate clearance, and APACHE II score in critically ill ICU patients. Specifically, the study seeks to measure the PCO2 gap at admission and after 12 hours, assess lactate levels and clearance at the same intervals, and determine the APACHE II score at these times. These measures will be compared in terms of mortality to determine which is the best predictor.

A prospective observational study was conducted in the Intensive Care Unit of the Department of Anaesthesiology & Critical Care at S.N. Medical College, Agra, following approval from the Institutional Review Board and Ethics Committee (IEC/2022/125). Informed consent was obtained from the relatives of all patients. The study included 66 critically ill patients admitted to the ICU, aged 18 years or older, and expected to stay in the ICU for more than 48 hours. Exclusion criteria were patients younger than 18, ICU stays under 48 hours, pregnancy, acute coronary syndrome, psychiatric illness, endocrinal pathology, and terminal cancer.

Blood samples were collected via central venous and intra-arterial catheters to assess the PCO2 difference between the two samples at 0 and 12 hours, using an ABG analyzer. Initial blood lactate levels were also measured at admission and 12 hours later. APACHE II scores were calculated at admission and again after 12 hours using an online calculator (20). The outcomes of these parameters were analyzed in relation to mortality to determine their prognostic value.

Statistical Analysis: Data were analyzed using SPSS 20.0 software. The Kolmogorov-Smirnov test or Kruskal-Wallis rank sum test was applied for quantitative data, while classification variables were expressed as N (%). Chi-square tests were used to compare differences between groups. Binary logistic regression analysis assessed risk factors, and Hosmer-Lemeshow tests were applied. Receiver operating characteristic (ROC) curves determined the prognostic value of the parameters, with significance defined as a two-tailed p-value < 0.05.

The study population consist of 66 patients who were admitted in ICU during the specified time period; 40 males (60.6%) and 26 females (39.4%) with a mean age of 44.1 years (±14.8 years SD). Out of 66 cases, 41 patients (62.1%) required inotropic support while 53 (80.3%) patients required ventilatory support and 38 (57.6%) succumbed to death whereas 28 (42.4%) patients survived.( Table-1.)

The results of the measurements of the prognostic parameters are highlighted in Table-1. The mean Lactate at the time of admission (H0) was 4.81±1.25 and after 12 hours (H12) was 4.69±1.54. The mean Lactate clearance was found to be 3.70±14.367% (range -22.78% to 36.75%). The mean PCO2 GAP at H0 was 5.82±1.57 and at H12 it was 5.25±1.43. The mean APACHE II at H0 was 25.30±9.07 and at H12 it was 23.06±8.83.

We performed a comparative analysis between the prognostic parameters between the patients who survived and those who did not and a highly significant difference was found; including Mean lactate at H0& H12 of the patients who did not Survived includes 5.36 and 5.69 respectively and those who survived includes 4.05 and 3.32 respectively which showed a significant Difference at  H0 (p<0.001) and at H12 (p<0.001),  Mean Lactate clearance Of the patients who did not survived includes -7% and who did survive shows 18% which shows a difference (p<0.001)., Mean PCO2 gap at H0 & H12 of the patient Who did not survive includes 6.63 and 5.96 respectively & Those who survived includes 4.71 and 4.28 respectively which showed a difference at H0(p<0.001) and at H12 (p<0.001), Mean APACHE II at H0 & H12 of the patient Who did not survive includes 30.63 & 28.74 respectively & Those who survived includes 18.07 and 15.36 respectively which showed a difference at H0(p<0.001) and at H12 (p<0.001). All these parameters showed higher values in those who did not survive except Lactate clearance which showed lower mean in the same.(Table-:2)

No significant difference in Prognostic parameterswas found between patients who required inotropic support and those who did not, the parameters being  lactate at H0 (p=0.223) and at H12 (p=0.128), lactate clearance (p=0.186), PCO2 gap at H0 (p=0.499) and at H12 (p=0.490), APACHE II at H0 (p=0.284) and at H12 (p=0.132). Whereas, A highly significant difference was calculated in lactate at H0 (p=0.001) and lactate at H12 (p<0.001) between patients who required ventilatory support and those who did not. Similar observations were made for parameters including lactate clearance (p<0.001), pCO2 gap at H0( p<0.001) and at H12( p<0.001), APACHE II at H0(p<0.001) and at H12 (p<0.001) between patients who required and did not require ventilator.(Table -3, Table - 4.)

Although no significant association of patient requiring inotropic support with outcome was found (p= 0.081), a significant association was found with patients who expired requiring ventilator support (p< 0.001).

According to ROC analysis (figure.01), the LACTATE clearance was found to be best prognostic parameter forpredicting mortality (with highest AUROC value 0.979). The optimum cut off for predicting mortality was <6.19% for Lactate clearance which had sensitivity of 92.1% and specificity 100% .The next to the best prognostic parameter was found to be LACTATE at H12 (AUROC=0.970). 

Table-1: Demographic and prognostic variables of study subjects

H0 =At 0 hour

H12 = At 12 hours

N= Sample size

Figure-1: ROC Analysis For Comparing

Prognostic Accuracy and

Finding Optimum Cut off of Prediction of Survival

                   H0 = at 0 hour

                   H12 = At 12 hours

Table-2: Comparison of prognostic parameters between patients who survived and those who did not.

SD = standard division

H0 = at 0 hour

H12 = At 12 hours

Table – 3: Comparison of Prognostic Parameters between patients who required inotropic support and those who did not.

SD = standard division

H0 = at 0 hour

H12 = At 12 hours

Table – 4 : Comparison of Prognostic Parameters between patients who required ventilatory support and those who did not.

SD = standard division

H0 = at 0 hour

H12 = At 12 hours

In this study, lactate clearance at 12 hours of ICU stay was the strongest predictor of mortality (AUROC = 0.979), although both static lactate levels (AUROC = 0.970) at 12 hours and lactate clearance showed good diagnostic accuracy for predicting mortality. Understanding the significance of elevated lactate levels requires considering not only anaerobic production but also aerobic mechanisms and changes in lactate clearance. Despite these complexities, elevated lactate levels are generally associated with higher morbidity and mortality rates (21). Early indicators of prognosis are essential for clinicians, and both initial lactate levels and early lactate clearance provide useful tools for this purpose (22).

Badreldin et al. (23) compared the predictive value of lactate with complex physiological scores in a cohort of cardiothoracic surgery patients, finding that lactate had a better diagnostic performance (AUC of 0.88) compared to APACHE II (AUC of 0.76). This aligns with our findings, where APACHE II had an AUROC of 0.946 at 12 hours. Even in hemodynamically stable surgical patients, prolonged hyperlactatemia has been linked to increased mortality (24). Arterial lactate is a sensitive marker of tissue ischemia and hypoxia, and dynamic lactate monitoring is crucial for the early diagnosis of shock and tissue hypoxia. Successful resuscitation, indicated by improved oxygen delivery, should be reflected in a decrease in blood lactate concentration (25). Studies suggest that a lactate clearance of 10% or more signifies adequate oxygen delivery to tissues and serves as an independent predictor of survival in patients with septic shock (26, 27).

Dynamic assessment of metabolic parameters is likely more effective in predicting mortality than static values. Several studies, particularly in patients with severe sepsis, have highlighted the importance of lactate clearance within the first 6 hours of resuscitation for predicting 28-day survival (28). In contrast, our study assessed lactate clearance at 12 hours post-admission and found a significant association with ICU outcomes. A recent study on ICU patients with hyperlactatemia showed that lactate-guided therapy significantly reduced hospital mortality, emphasizing the clinical utility of early lactate monitoring (29).

Another important parameter assessed in this study was the venous-arterial difference in carbon dioxide, or the PCO2 gap. We found a significant association between the PCO2 gap at admission and 12 hours later with ICU mortality. The most sensitive variable in our study was the PCO2 gap at admission, with a sensitivity of 97.4%, followed by the PCO2 gap and APACHE II score at 12 hours, both showing sensitivities of 94.7%. The mortality prediction cut-off for the PCO2 gap at admission was >5 mmHg, which aligns with studies by Bakker, Emmanuel Robin, and HJ Androuge, who reported that a PCO2 gap of around 6 mmHg was associated with higher mortality risk (30).

The normal range for the P(cv-a)CO2 is typically 2-5 mmHg, with values above 6 mmHg indicating inadequate cardiac output and tissue hypoperfusion . This is consistent with reports by Varpuls and others, which have shown a negative correlation between the PCO2 gap and cardiac output. Other research has indicated that patients with a PCO2 gap greater than 6 mmHg often have higher lactate levels, lower cardiac index (CI), and lower central venous oxygen saturation (ScvO2) compared to those with a PCO2 gap below 6 mmHg . Fabrice Vallee et al. (2008) also demonstrated that a persistently elevated PCO2 gap at admission is associated with reduced lactate clearance and poorer outcomes (31), findings that are supported by our study.

Blood lactate clearance at 12 hours emerged as the most reliable indicator of ICU mortality. However, including the PCO2 gap at admission may serve as a valuable complementary tool for guiding resuscitation and therapeutic interventions aimed at improving outcomes in critically ill patients. Overall, hyperlactatemia and the APACHE II score, both at admission and after 12 hours, were also independent predictors of mortality.

1. Rapsang, A. G., and Shyam D. C. "Scoring Systems in the Intensive Care Unit: A Compendium." Indian Journal of Critical Care Medicine, vol. 18, no. 4, 2014, pp. 220-28. doi:10.4103/0972-5229.130573.
2. Jhanji, S., et al. "Microvascular Flow and Tissue Oxygenation after Major Abdominal Surgery: Association with Post-Operative Complications." Intensive Care Medicine, vol. 35, no. 4, 2009, pp. 671-77. doi:10.1007/s00134-008-1325-z.
3. Lobo, S. M., et al. "Early Determinants of Death Due to Multiple Organ Failure after Noncardiac Surgery in High-Risk Patients." Anesthesia & Analgesia, vol. 112, no. 4, Apr. 2011, pp. 877-83. doi:10.1213/ANE.0b013e3181e2bf8e.
4. Singer, M., et al. "The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3)." JAMA, vol. 315, no. 8, 2016, pp. 801-10. doi:10.1001/jama.2016.0287.
5. Cecconi, M., et al. "Sepsis and Septic Shock." Lancet, vol. 392, no. 10141, 2018, pp. 75-87. doi:10.1016/S0140-6736(18)30696-2.
6. Rhodes, A., et al. "Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016." Intensive Care Medicine, vol. 43, no. 3, Mar. 2017, pp. 304-77. doi:10.1007/s00134-017-4683-6.
7. Gurgel, S. T., and P. Jr. do Nascimento. "Maintaining Tissue Perfusion in High-Risk Surgical Patients: A Systematic Review of Randomized Clinical Trials." Anesthesia & Analgesia, vol. 112, no. 6, 2011, pp. 1384-91. doi:10.1213/ANE.0b013e3182055384.
8. Jhanji, S., et al. "Haemodynamic Optimisation Improves Tissue Microvascular Flow and Oxygenation after Major Surgery: A Randomised Controlled Trial." Critical Care, vol. 14, no. 4, 2010, p. R151. doi:10.1186/cc9220.
9. Evans, L., et al. "Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021." Intensive Care Medicine, vol. 47, no. 11, 2021, pp. 1181-1247. doi:10.1007/s00134-021-06506-y.
10. Vink, E. E., and J. Bakker. "Practical Use of Lactate Levels in the Intensive Care." Journal of Intensive Care Medicine, vol. 33, no. 3, 2018, pp. 159-65. doi:10.1177/0885066617708563.
11. Nguyen, H. B., et al. "Early Lactate Clearance Is Associated with Improved Outcome in Severe Sepsis and Septic Shock." Critical Care Medicine, vol. 32, no. 8, 2004, pp. 1637-42. doi:10.1097/01.ccm.0000132904.35713.a7.
12. Jansen, T. C., et al. "Early Lactate-Guided Therapy in Intensive Care Unit Patients: A Multicenter, Open-Label, Randomized Controlled Trial." American Journal of Respiratory and Critical Care Medicine, vol. 182, no. 6, 2010, pp. 752-61. doi:10.1164/rccm.200912-1918OC.
13. Arnold, R. C., et al. "Multicenter Study of Early Lactate Clearance as a Determinant of Survival in Patients with Presumed Sepsis." Shock, vol. 32, no. 1, 2009, pp. 35-39. doi:10.1097/shk.0b013e3181971d47.
14. Bakker, J., et al. "Serial Blood Lactate Levels Can Predict the Development of Multiple Organ Failure Following Septic Shock." American Journal of Surgery, vol. 171, no. 2, 1996, pp. 221-26. doi:10.1016/S0002-9610(97)89552-9.
15. McEvoy, J. D., et al. "Mixed Venous and Arterial PCO2." British Medical Journal, vol. 4, no. 5946, 1974, pp. 687-90. doi:10.1136/bmj.4.5946.687.
16. Bakker, J., et al. "Veno-Arterial Carbon Dioxide Gradient in Human Septic Shock." Chest, vol. 101, no. 2, 1992, pp. 509-15. doi:10.1378/chest.101.2.509.
17. "Critical Care Scoring Systems." MSD Manual Professional Edition, 2024, www.msdmanuals.com/en-in/professional/critical-care-medicine/approach-to-the-critically-ill-patient/critical-care-scoring-systems. Accessed 7 Oct. 2024.
18. Knaus, W. A., et al. "APACHE II: A Severity of Disease Classification System." Critical Care Medicine, vol. 13, no. 10, 1985, pp. 818-29. doi:10.1097/00003465-198603000-00013.
19. Kumar, R., and N. Kumar. "Validation of Lactate Clearance at 6 h for Mortality Prediction in Critically Ill Children." Indian Journal of Critical Care Medicine, vol. 20, no. 10, 2016, pp. 570-74. doi:10.4103/0972-5229.192040.
20. "APACHE II Score." MDCalc, 2024, www.mdcalc.com/apache-ii-score. Accessed 7 Oct. 2024.
21. Bakker, J., et al. "Clinical Use of Lactate Monitoring in Critically Ill Patients." Annals of Intensive Care, vol. 3, no. 1, 2013, p. 12. doi:10.1186/2110-5820-3-12.
22. Régnier, M. A., et al. "Prognostic Significance of Blood Lactate and Lactate Clearance in Trauma Patients." Anesthesiology, vol. 117, no. 6, 2012, pp. 1276-88. doi:10.1097/ALN.0b013e318273349d.
23. Badreldin, A. M., et al. "Mortality Prediction after Cardiac Surgery: Blood Lactate Is Indispensable." Thoracic and Cardiovascular Surgeon, vol. 61, no. 8, 2013, pp. 708-17. doi:10.1055/s-0032-1324796.
24. Meregalli, A., et al. "Occult Hypoperfusion Is Associated with Increased Mortality in Hemodynamically Stable, High-Risk, Surgical Patients." Critical Care, vol. 8, no. 2, 2004, pp. R60-65. doi:10.1186/cc2423.
25. Weil, M. H., and A. A. Afifi. "Experimental and Clinical Studies on Lactate and Pyruvate as Indicators of the Severity of Acute Circulatory Failure (Shock)." Circulation, vol. 41, no. 6, 1970, pp. 989-1001. doi:10.1161/01.cir.41.6.989.
26. Jones, A. E., et al. "Lactate Clearance vs Central Venous Oxygen Saturation as Goals of Early Sepsis Therapy: A Randomized Clinical Trial." JAMA, vol. 303, no. 8, 2010, pp. 739-46. doi:10.1001/jama.2010.158.
27. Vallée, F., et al. "Central Venous-to-Arterial Carbon Dioxide Difference: An Additional Target for Goal-Directed Therapy in Septic Shock?" Intensive Care Medicine, vol. 34, no. 12, 2008, pp. 2218-25. doi:10.1007/s00134-008-1199-0.
28. Tian, H. H., et al. "The Effect of Early Goal Lactate Clearance Rate on the Outcome of Septic Shock Patients with Severe Pneumonia." Zhongguo Wei Zhong Bing Ji Jiu Yi Xue, vol. 24, no. 1, 2012, pp. 42-45. doi:10.3760/cma.j.issn.1003-0603.2012.01.010.
29. Jansen, T. C., et al. "Early Lactate-Guided Therapy in Intensive Care Unit Patients: A Multicenter, Open-Label, Randomized Controlled Trial." American Journal of Respiratory and Critical Care Medicine, vol. 182, no. 6, 2010, pp. 752-61. doi:10.1164/rccm.200912-1918OC.
30. "Critical Care Scoring Systems." MSD Manual Professional Edition, 2024, www.msdmanuals.com/en-in/professional/critical-care-medicine/approach-to-the-critically-ill-patient/critical-care-scoring-systems. Accessed 7 Oct. 2024.
31. Vallée, F., et al. "Central Venous-to-Arterial Carbon Dioxide Difference: An Additional Target for Goal-Directed Therapy in Septic Shock?" Intensive Care Medicine, vol. 34, no. 12, 2008, pp. 2218-25. doi:10.1007/s00134-008-1199-0.