Pv-aCO2/ΔAVO2 ratio and lactate define high-risk hemodynamic profiles in cardiac surgery

Abstract

Introduction

Cardiac surgery with cardiopulmonary bypass (CPB) induces physiological stress that often results in hemodynamic mismatch. Assessment of macro- and microcirculatory variables is crucial for adequate tissue oxygenation. Conventional markers like lactate and ScvO₂ are limited, while CO₂-derived indices, including Pv-aCO₂ and the Pv-aCO₂/ΔAVO₂ ratio, may provide earlier detection of anaerobic metabolism. This study aimed to determine critical oxygen delivery (DO₂crit) and its association with hemodynamic profiles and outcomes.

Methods

A retrospective study analyzed 508 adults undergoing cardiac surgery with CPB at the Instituto Nacional de Cardiología “Ignacio Chávez” (June 2022–June 2024). Patients were stratified into Profile 1 (lactate < 2 mmol/L), Profile 2 (lactate ≥ 2 mmol/L and Pv-aCO₂/ΔAVO2 > 1), and Profile 3 (lactate ≥ 2 mmol/L and Pv-aCO₂/ΔAVO₂ ≤ 1). Clinical, surgical, and hemodynamic data were collected, and outcomes assessed using logistic regression adjusted for age and sex.

Results

Of 508 patients, 165 were Profile 1, 295 Profile 2, and 48 Profile 3. Profile 2 showed higher central venous pressure, elevated Pv-aCO₂, greater vasoactive support, and longer CPB and cross-clamp times. This group had higher risks of cerebrovascular events (OR 4.3), pneumonia (OR 2.0), and acute kidney injury (OR 1.8). DO₂crit ≤ 7.3 ml/min/kg was linked to Profile 2 and adverse outcomes.

Conclusions

Integration of macro- and microcirculatory variables identifies prognostically relevant subgroups. Profile 2, defined by lactate ≥ 2 mmol/L and Pv-aCO₂/ΔAVO₂ > 1, was associated with increased morbidity and DO₂crit ≤ 7.3 ml/min/kg. Early profiling may improve risk stratification and guide targeted interventions.

Resumen

Introducción

La cirugía cardíaca con circulación extracorpórea (CEC) induce un estrés fisiológico que con frecuencia resulta en un desacoplamiento hemodinámico. La evaluación de variables macro- y microcirculatorias es fundamental para garantizar una adecuada oxigenación tisular. Los marcadores convencionales como el lactato y la ScvO₂ presentan limitaciones, mientras que los índices derivados del CO₂, incluyendo la Pv-aCO₂ y la relación Pv-aCO₂/ΔAVO₂, pueden permitir una detección más temprana del metabolismo anaerobio. Este estudio tuvo como objetivo determinar la entrega crítica de oxígeno (DO₂crit) y su asociación con perfiles hemodinámicos y desenlaces clínicos.

Métodos

Se realizó un estudio retrospectivo en el que se analizaron 508 pacientes adultos sometidos a cirugía cardíaca con circulación extracorpórea en el Instituto Nacional de Cardiología “Ignacio Chávez” (junio de 2022–junio de 2024). Los pacientes se estratificaron en Perfil 1 (lactato < 2 mmol/L), Perfil 2 (lactato ≥2 mmol/L y Pv-aCO₂/ΔAVO₂ >1) y Perfil 3 (lactato ≥2 mmol/L y Pv-aCO₂/ΔAVO₂ ≤1). Se recopilaron datos clínicos, quirúrgicos y hemodinámicos, y los desenlaces se evaluaron mediante regresión logística ajustada por edad y sexo.

Resultados

De los 508 pacientes, 165 correspondieron al Perfil 1, 295 al Perfil 2 y 48 al Perfil 3. El Perfil 2 presentó mayor presión venosa central, valores elevados de Pv-aCO₂, mayor requerimiento de soporte vasoactivo y tiempos más prolongados de circulación extracorpórea y de pinzamiento aórtico. Este grupo mostró mayor riesgo de eventos cerebrovasculares (OR 4.3), neumonía (OR 2.0) y lesión renal aguda (OR 1.8). Un DO₂crit ≤7.3 ml/min/kg se asoció con el Perfil 2 y con desenlaces adversos.

Conclusiones

La integración de variables macro- y microcirculatorias permite identificar subgrupos con relevancia pronóstica. El Perfil 2, definido por lactato ≥2 mmol/L y una relación Pv-aCO₂/ΔAVO₂ >1, se asoció con mayor morbilidad y con DO₂crit ≤7.3 ml/min/kg. La estratificación temprana podría mejorar la evaluación del riesgo y orientar intervenciones dirigidas.

Introduction

Background

Patients undergoing cardiac surgery face significant physiological stress induced by tissue injury and the use of cardiopulmonary bypass (CPB).¹ This condition generates a state of hemodynamic mismatch, particularly during the first 24 postoperative hours.² Hemodynamic assessment includes both macrocirculatory and microcirculatory parameters, which are essential to ensure adequate cellular oxygenation. Macrocirculation guarantees global blood flow to the organs, while microcirculation maintains fine tissue perfusion; loss of coherence between the two is associated with worse outcomes and reduced therapeutic response.³

Macrocirculatory parameters include cardiac index (CI), indexed systemic vascular resistance (SVRi), central venous pressure (CVP), hemoglobin concentration, and oxygen saturation, which can be measured using noninvasive techniques such as ultrasound or invasive methods such as thermodilution catheters.^2–4 Microcirculatory parameters involve glycocalyx disruption, leukocyte activation, intravascular coagulation, and microthrombosis, resembling those observed in sepsis.^2,3 These can be assessed using videomicroscopy, perfusion indices (capillary refill time, lactate), central venous oximetry (ScvO₂), mixed venous oximetry (SvO₂), and CO₂-derived determinants, such as the veno-arterial PCO₂ difference (Pv-aCO₂) and the Pv-aCO₂/Arteriovenous oxygen difference (ΔAVO₂).^1–5

Hemodynamic mismatch may manifest as macrocirculatory dysfunction (cardiogenic shock, with reduced oxygen delivery [DO₂]) or microcirculatory dysfunction (distributive shock, with impaired oxygen extraction [O₂E]).⁵ When DO₂ fails to meet oxygen consumption (VO₂), the organism shifts to anaerobic metabolism with lactate production, defining the concept of critical oxygen delivery (DO₂crit).^5,6

Importance

Several parameters have been proposed as targets for hemodynamic resuscitation, including ScvO₂, lactate, and CO₂-derived indices. However, each presents limitations: ScvO₂ may be falsely normal in the early stages of septic shock due to reduced O₂E^5,7,8; lactate reflects anaerobic metabolism but lacks sensitivity in early phases and may rise from non-hypoxic causes in the postoperative setting of cardiac surgery.^1,5,9,10 In contrast, parameters such as Pv-aCO₂ and the Pv-aCO₂/ΔAVO₂ ratio may detect early microcirculatory alterations and provide prognostic insights.^11–13

The Pv-aCO₂/ΔAVO₂ ratio is an indirect marker of tissue anaerobiosis as it integrates the relationship between CO₂ production and VO₂. Under aerobic conditions, both processes maintain a stable proportion (respiratory quotient ≈ 1); however, when oxygen delivery becomes insufficient, metabolism shifts toward anaerobic glycolysis, leading to a relative reduction in O₂E (lower ΔAVO₂) and simultaneous increase in CO₂ production and retention (higher Pv-aCO₂) due to hypoperfusion. This uncoupling results in an elevated Pv-aCO₂/ΔAVO₂ ratio, with values > 1 indicating microcirculatory dysfunction, anaerobic metabolism, and worse clinical prognosis.^11,12,14,15

Recently, hemodynamic profiles combining macro- and microcirculatory variables have been described. In particular, a profile characterized by low microcirculatory flow and anaerobic metabolism has been associated with increased morbidity and mortality.² However, while DO₂crit has been studied in contexts such as septic and cardiogenic shock, its specific value in patients undergoing cardiac surgery remains unknown.

Goals of investigation

The primary objective of this study is to determine DO₂crit in patients undergoing cardiac surgery in the Cardiovascular Critical Care Unit at the Instituto Nacional de Cardiología “Ignacio Chávez.” Secondary objectives include classifying patients into three hemodynamic profiles—Profile 1, defined by lactate < 2 mmol/L (aerobic metabolism); Profile 2, with lactate ≥ 2 mmol/L and Pv-aCO₂/ΔAVO₂ ratio > 1 (anaerobic metabolism); and Profile 3, with lactate ≥ 2 mmol/L and Pv-aCO₂/ΔAVO₂ ratio ≤ 1 (lactate kinetics)—comparing DO₂crit across these profiles, and identifying clinical and hemodynamic factors associated with each profile.

Methods

This retrospective, cross-sectional observational study included adult patients of both sexes who underwent cardiac surgery with CPB at the Instituto Nacional de Cardiología “Ignacio Chávez” between June 1, 2022, and June 1, 2024. Patients with less than 12 h in the Cardiovascular Critical Care Unit, those who died intraoperatively, or with incomplete medical records were excluded, resulting in a final cohort of 508 patients. Data were obtained through a comprehensive review of physical and electronic medical records and imaging archives.

The study population was stratified into three metabolic profiles according to serum lactate levels and the Pv-aCO₂/ΔAVO₂ ratio. This classification was intended to differentiate patients with preserved aerobic metabolism from those who developed anaerobiosis as a consequence of tissue hypoxia.

The first profile included patients with normal lactate levels (<2 mmol/L) and a Pv-aCO₂/ΔAVO₂ ratio < 1, consistent with an aerobic metabolic state. At the opposite end, patients with elevated lactate (>2 mmol/L) and a Pv-aCO₂/ΔAVO₂ ratio > 1 were categorized as having anaerobic metabolism, reflecting an evident imbalance between oxygen delivery and consumption. Finally, a subgroup of patients exhibited lactate > 2 mmol/L but with a Pv-aCO₂/ΔAVO₂ ratio < 1. In these cases, the ratio indicated preserved aerobic metabolism, suggesting that lactate elevation was attributable to altered kinetics (impaired clearance, increased production from non-hypoxic pathways, or redistribution) rather than to true tissue hypoxia.

This stratification provides a more precise characterization of metabolic status, helps to elucidate the underlying mechanisms of hyperlactatemia, and prevents misinterpretation that could otherwise lead to inappropriate therapeutic interventions.

Variables were categorized as baseline characteristics (demographics, comorbidities, NYHA functional class, EuroSCORE II), surgical variables (bypass and aortic cross-clamp times, type of procedure), hemodynamic parameters (CI, SVRi, CVP, ScvO₂, Pv-aCO₂, O₂E, DO₂, VO₂, DO₂/VO₂ ratio, lactate levels, capillary refill time), vasoactive and inotropic drug use (measured 6 h after admission to the cardiovascular critical care unit following surgery) and postoperative outcomes, including bleeding, low cardiac output syndrome, vasoplegia, hypovolemia, delirium, cerebrovascular events, pneumonia, mediastinitis, transfusion, acute kidney injury, renal replacement therapy, hepatic failure, atrial fibrillation, in-hospital mortality, ICU stay, mechanical ventilation, and total hospital length of stay.

In the “other surgeries” group, which individually represented a small proportion of patients, the most frequent procedures included triple valve replacement, aortic surgery (other than the Bentall–De Bono procedure), tricuspid valve replacement, congenital heart disease surgery, heart transplantation, pulmonary thromboendarterectomy, endoventricular remodeling, post-infarction ventricular septal defect closure, and surgery for infective endocarditis.

Continuous variables were assessed for normality using the Shapiro–Wilk test and presented as mean ± SD or median (IQR) as appropriate. Comparisons were performed using Mann–Whitney U, Chi-square, or Fisher's exact tests. Logistic regression adjusted for age and sex was used to identify predictors of adverse events. Statistical significance was defined as p < 0.05, and analyses were conducted with STATA version 14 (StataCorp, College Station, TX, USA).

The requirement for formal approval was waived by the local institutional research and ethics committees. Written informed consent was obtained from all participants prior to study inclusion, including consent for the publication of clinical information and images, either directly from the patient or from a legally authorized representative.

Results

Between June 1, 2022, and June 1, 2024, a total of 508 patients meeting the inclusion criteria were recruited in the Cardiovascular Intensive Care Unit at the Instituto Nacional de Cardiología “Ignacio Chávez.” Of these, 165 patients had lactate < 2 mmol/L, 295 patients had lactate ≥ 2 mmol/L Pv-aCO₂/ΔAVO₂ ratio, and 48 patients had lactate ≥ 2 mmol/L and Pv-aCO₂/ΔAVO₂ ratio.

Demographic characteristics

The mean age of the cohort was 57 years, and 43.9% were women, with no significant differences among the three profiles. Weight, height, and body mass index were comparable across groups. Comorbidities and cardiovascular history were frequent: hypertension 42.3%, diabetes mellitus 22.4%, and hypothyroidism 10%. Cardiovascular diseases included heart failure in 27%, atrial fibrillation in 18.3%, and prior myocardial infarction in 11.6%. Previous cardiac surgery was present in 11% of patients, more frequently in Profile 2. The predominant functional class was NYHA II (62.2%) (Table 1).

Table 1.

Baseline characteristics.

Variable	Total (n = 508)	Profile 1 (n = 165)	Profile 2 (n = 295)	Profile 3 (n = 48)	p
Women n (%)	223 (43.9)	68 (41.2)	139 (47.1)	16 (34.4)	0.14
Men n (%)	285 (56.1)	97 (58.8)	156 (52.9)	32 (66.6)
NYHA functional class n (%)
I	63 (12.4)	30 (18.2)	28 (9.5)	5 (10.4)	0.21
II	316 (62.2)	99 (60)	188 (63.7)	29 (60.4)
III	113 (22.2)	31 (18.8)	70 (23.7)	12 (25)
IV	16 (3.2)	5 (3)	9 (3.1)	2 (4.2)
Previous cardiac surgery n (%)	56 (11)	11 (6.7)	43 (14.6)	2 (4.2)	0.01
Hypothyroidism n (%)	51 (10)	16 (9.7)	32 (10.9)	3 (6.3)	0.60
Myocardial infarction n (%)	59 (11.6)	25 (15.2)	28 (9.5)	6 (12.5)	0.18
Diabetes n (%)	114 (22.4)	47 (28.5)	54 (18.3)	13 (27)	0.03
Chronic obstructive pulmonary disease n (%)	4 (0.8)	1 (0.6)	2 (0.7)	1 (2)	0.56
Chronic kidney disease n (%)	28 (5.5)	14 (8.5)	12 (4.1)	2 (4.2)	0.12
Hypertension n (%)	215 (42.3)	78 (47.3)	117 (39.7)	20 (41.7)	0.28
Heart failure n (%)	137 (27)	44 (26.7)	80 (27.1)	13 (27.1)	0.99
Atrial fibrillation n (%)	93 (18.3)	24 (14.6)	62 (21)	7 (14.6)	0.17
Stroke n (%)	28 (5.5)	2 (7.3)	14 (4.8)	2 (4.2)	0.47
Weight (kg)Median (IQR)	68 (59.5–78)	70 (59.6–80)	68 (60–76.4)	65 (57.9–74)	0.44
Height (m)Median (IQR)	1.62 (1.55–1.7)	1.64 (1.55–1.72)	1.62 (1.55–1.7)	1.6 (1.57–1.68)	0.32
Body mass index (kg/m^2)Median (IQR)	25.9 (23.3–28.8)	25.9 (23.5–28.7)	26.1 (23.1–29)	24.5 (23.2–27.4)	0.49
Age	57 (45–65)	55 (43–66)	57 (46–65)	57.5 (44.5–67)	0.85

NYHA: New York Heart Association; IQR: interquartile range.

Surgical characteristics

The most common procedures were aortic valve replacement (29.9%) and coronary artery bypass grafting (16%). Patients in Profile 2 had a higher frequency of Bentall and Bono procedures (7.8% vs. 3% and 2.1%), as well as longer cardiopulmonary CPB and aortic cross-clamp times compared with the other profiles (Table 2).

Table 2.

Surgical characteristics.

Variable	Total (n = 508)	Profile 1 (n = 165)	Profile 2 (n = 295)	Profile 3 (n = 48)	p
EuroSCOREMedian (IQR)	1.85 (1–3.6)	1.6 (0.9–3.4)	2 (1–3.8)	2.1 (1–3.4)	0.72
Extracorporeal circulation time (min)Median (IQR)	143 (111–181)	133 (100–171)	149 (118–186)	137.5 (107.5–201.5)	0.05
Aortic clamping (min)Median (IQR)	99 (76–124)	91 (71.5–118)	101 (81–129)	100 (72–126.5)	0.03
Coronary artery bypass graft n (%)	81 (16)	29 (17.6)	45 (15.3)	7 (14.6)	0.78
Mitral valve replacement n (%)	47 (9.3)	11 (6.7)	31 (10.5)	5 (10.4)	0.38
Aortic valve replacement n (%)	152 (29.9)	16 (33.3)	79 (26.8)	57 (34.6)	0.19
Mitral valve replacement + tricuspid valve replacement n (%)	25 (4.9)	7 (4.2)	17 (5.8)	1 (2.0)	0.48
Aortic valve replacement + mitral valve replacement n (%)	36 (7.1)	9 (5.5)	25 (8.5)	2 (4.2)	0.73
Coronary artery bypass graft + aortic valve replacement n (%)	18 (3.5)	4 (2.4)	10 (3.4)	4 (8.3)	0.14
Bentall procedure n (%)	29 (5.7)	5 (3)	23 (7.8)	1 (2.1)	0.05
Others n (%)	124 (24.4)	45 (27.3)	67 (22.7)	12 (25)	0.55

IQR: interquartile range.

Hemodynamic characteristics and vasoactive use

Among macrocirculatory variables, CVP was significantly higher in Profile 2 (10 mmHg) compared with Profiles 1 and 3 (9 and 8 mmHg, respectively). Regarding microcirculatory parameters, Pv-aCO₂ was higher in Profile 2 (8 mmHg) than in Profiles 1 and 3 (6 and 3 mmHg). A trend toward lower DO₂ was also observed in this group, although without statistical significance.

Vasoactive drug use showed notable differences: norepinephrine was administered more frequently and at higher doses in Profile 2 (64.4%, median 0.09 μg/kg/min) compared with Profile 1 (49.1%, 0.06) and Profile 3 (52.1%, 0.06). Similarly, vasopressin was used more frequently in Profile 2 (28.5%) than in Profiles 1 and 3 (13.3% and 20.8%), although dose differences were not significant (Table 3).

Table 3.

Hemodynamic variables and vasoactive drugs.

Variable	Total (n = 508)	Profile 1 (n = 165)	Profile 2 (n = 295)	Profile 3 (n = 48)	p
6 h
Cardiac index (L/min/m^2) Median (IQR)	2.1 (1.6–2.6)	2.1 (1.8–2.6)	2 (1.6–2.6)	2.2 (1.6–2.8)	0.31
Central venous pressure (mmHg) Median (IQR)	9 (8–11)	9 (8–11)	10 (8–12)	8 (8–10)	0.27
Systemic vascular resistance index (dynes-sec/cm^5/m^2) Median (IQR)	2491 (1946–3128)	2423 (1885–3106)	2535 (1971–3158)	2584 (2008–3232)	0.85
Mixed venous O2 saturation (%) Median (IQR)	70 (63–76)	69.1 (64–75)	68.9 (61.8–76)	72.4 (69–80)	0.02
Arteriovenous O2 difference (ml/dl) Median (IQR)	4.3 (3.4–5.3)	4.2 (3.5–5.2)	4.5 (3.6–5.5)	3.9 (3.1–4.8)	0.47
O2 extraction ratio (%) Median (IQR)	30 (23–36)	29 (23–34)	30 (23–38)	26 (20–30)	0.03
Venous-to-arterial CO2 pressure difference (mmHg) Median (IQR)	7 (5–9)	6 (4–8)	8 (6–10)	3 (2–3)	0.63
DO2 (ml/min/m^2) Median (IQR)	518 (401.3–659.7)	541.96 (427.9–646)	494.6 (390–649.9)	574.4 (440.2–725.6)	0.07
VO2 (ml/min/m^2) Median (IQR)	129 (123–185)	135 (124.2–187.7)	128.6 (122.2–185)	125.3 (115.5–155)	0.93
DO2/VO2 Median (IQR)	3.4 (2.7–4.3)	3.4 (2.8–4.3)	3.3 (2.6–4.3)	3.7 (3.4–5.2)	0.11
Lactate Median (IQR)	2.5 (1.7–3.8)	1.5 (1.2–1.7)	3.3 (2.4–5.1)	3.4 (2.5–3.9)	0.0004
Capillary refill time (seg) Median (IQR)	2 (2–3)	2 (2–2.5)	2 (2–3)	2 (2–3)	0.01
Norepinephrine n (%)	296 (58.3)	81 (49.1)	190 (64.4)	25 (52.1)	0.04
Dose of norepinephrine Median (IQR)	0.08 (0.05–0.15)	0.06 (0.03–0.1)	0.09 (0.05–0.16)	0.06 (0.05–0.12)	0.01
Dobutamine n (%)	155 (30.5)	42 (25.5)	97 (32.9)	16 (33.3)	0.23
Dose of dobutamine Median (IQR)	3.7 (2.5–5)	3 (2.5–5)	4 (3–6)	3 (2.4–5)	0.26
Vasopressin n (%)	116 (22.8)	22 (13.3)	84 (28.5)	10 (20.8)	0.01
Dose of vasopressin Median (IQR)	0.05 (0.03–0.06)	0.04 (0.02–0.06)	0.06 (0.03–0.07)	0.03 (0.03–0.06)	0.13
Levosimendan n (%)	93 (18.3)	26 (15.8)	59 (20)	8 (16.7)	0.82
Dose of levosimendan Median (IQR)	0.1 (0.05–0.1)	0.1 (0.05–0.1)	0.1 (0.05–0.1)	0.1 (0.05–0.13)	0.35
Milrinone n (%)	15 (2.9)	6 (3.6)	9 (3.1)	0 (0)	0.41
Dose of milrinone Median (IQR)	0.3 (0.2–0.5)	0.3 (0.13–0.4)	0.5 (0.3–0.55)	–	0.27
Methylene blue n (%)	19 (3.7)	4 (2.4)	13 (4.4)	2 (4.2)	0.55
Steroids n (%)	17 (3.4)	4 (2.4)	11 (3.7)	2 (4.2)	0.7

IQR: interquartile range.

Outcomes

Profile 2 exhibited a higher incidence of clinically significant complications, including cerebrovascular events (5.8% vs. 1.8% and 0%), hospital-acquired pneumonia (12.6% vs. 5.5% and 10.4%), acute kidney injury (35.7% vs. 23.6% and 22.9%), and hepatic failure (14.9% vs. 6.6% and 8.3%). Other outcomes such as vasoplegic syndrome, postcardiotomy low cardiac output syndrome, higher SOFA scores at 24 h, need for renal replacement therapy, and in-hospital mortality were more frequent in Profile 2, though differences did not reach statistical significance (Table 4).

Table 4.

Outcomes.

Variable	Total (n = 508)	Profile 1 (n = 165)	Profile 2 (n = 295)	Profile 3 (n = 48)	p
Mediastinal bleeding n (%)	63 (12.4)	19 (11.5)	39 (13.2)	5 (10.4)	0.79
Postcardiotomy low cardiac output syndrome n (%)	63 (12.4)	13 (7.9)	44 (14.9)	6 (12.5)	0.09
Vasoplegic syndrome n (%)	37 (7.3)	6 (3.6)	28 (9.5)	3 (6.3)	0.07
Hypovolemia n (%)	180 (35.4)	49 (29.7)	109 (36.9)	22 (45.8)	0.085
Delirium n (%)	60 (11.8)	13 (7.8)	40 (13.6)	7 (14.6)	0.15
Stroke n (%)	20 (3.9)	3 (1.8)	17 (5.8)	0 (0)	0.03
In-hospital pneumonia n (%)	51 (10.1)	9 (5.5)	37 (12.6)	5 (10.4)	0.05
Mediastinitis n (%)	19 (3.8)	8 (4.9)	11 (3.7)	0 (0)	0.30
Transfusion n (%)	255 (50.3)	78 (47.3)	156 (53.1)	21 (43.8)	0.31
Acute kidney injury n (%)	155 (30.6)	39 (23.6)	105 (35.7)	11 (22.9)	0.01
Renal replacement therapy n (%)	32 (6.3)	8 (4.9)	22 (7.5)	2 (4.2)	0.44
Liver injury n (%)	59 (11.6)	11 (6.6)	44 (14.9)	4 (8.3)	0.02
Post surgical atrial fibrilation n (%)	81 (15.9)	17 (10.3)	55 (18.7)	9 (18.8)	0.05
Mortality n (%)	32 (6.3)	7 (4.2)	24 (8.1)	1 (2.1)	0.12
Days in intensive care unitMedian (IQR)	3 (2–4)	3 (2–3)	3 (2–4)	3 (2–4)	0.01
Days with mechanical ventilationMedian (IQR)	10 (7–18)	10 (7–18)	10 (7–19)	9 (7–17)	0.89
Total hospitalization timeMedian (IQR)	1 (1–1)	1 (1–1)	1 (1–1)	1 (1–1)	0.01
SOFA score at 24 hMedian (IQR)	5 (3–7)	4 (3–6)	5 (3–7)	4.5 (2.5–6)	0.07
SOFA score at 72 hMedian (IQR)	3 (2–4.5)	3 (2–4.5)	3 (2–5)	4 (2–5)	0.05

IQR: interquartile range.

Logistic regression analysis

Multivariate analysis adjusted for age and sex showed that patients in Profile 2 had a significantly higher risk of cerebrovascular events (OR 4.3, 95% CI 1.23–14.78), hospital-acquired pneumonia (OR 2.0, 95% CI 1.07–3.88), and acute kidney injury (OR 1.8, 95% CI 1.21–2.69). Profile 2 was also associated with a trend toward higher in-hospital mortality risk (OR 2.2, 95% CI 0.99–5.13), although this association did not reach statistical significance (Table 5).

Table 5.

Logistic regression model.

Variable	OR	CI 95%	p
Stroke	4.27	1.23–14.78	0.02
In-hospital pneumonia	2.04	1.07–3.88	0.002
Acute kidney injury	1.81	1.21–2.69	0.003
Renal replacement therapy	1.63	0.75–3.52	0.21
Liver injury	1.95	0.95–3.99	0.06
Mortality	2.25	0.99–5.13	0.05

OR: odds ratio; CI: confidence interval.

Determination of DO₂crit

Finally, analysis of the DO₂/VO₂ ratio identified that cardiac surgery patients with a DO₂crit ≤ 7.3 ml/min/kg were more likely to develop Profile 2 (Fig. 1).

Fig. 1.

Relationship between oxygen delivery (DO₂) and oxygen consumption (VO₂). A decrease in DO₂ is initially compensated by increased oxygen extraction, maintaining VO₂. Beyond the critical DO₂ (7.3 ml/min/kg in our population), VO₂ becomes delivery-dependent, leading to anaerobic metabolism, identified by a Pv-aCO₂/ΔAVO₂ ratio > 1 and lactate >2 mmol/L (original image by the authors).

Discussion

Patients undergoing cardiac surgery exhibit hemodynamic profiles characterized by invasive and non-invasive variables that have been employed in other hemodynamic states. Among these, arterial blood gases represent an easily accessible tool from which oxygen- and carbon dioxide-derived indices can be calculated. These parameters allow identification of hemodynamic profiles that have been associated with postoperative adverse events, particularly when a critical imbalance exists between DO₂ and VO₂, a condition known as DO₂crit.

In the study population, classification of patients into hemodynamic profiles proved highly relevant. Patients with elevated lactate but a Pv-aCO₂/ΔAVO₂ ratio ≤ 1 behaved similarly to those with normal lactate. In contrast, patients with a Pv-aCO₂/ΔAVO₂ ratio > 1 (Profile 2) exhibited a higher number of postoperative complications. Among demographic variables, the only factor significantly associated with this profile was a history of prior cardiac surgery.

Surgically, Profile 2 was significantly associated with Bentall and Bono procedures, as well as longer operative times, particularly CPB and aortic cross-clamp durations. Patients undergoing Bentall and Bono procedures represent a particularly complex clinical subgroup, as they frequently present with ascending aortic dissection or aneurysm associated with acute or acutely decompensated chronic aortic regurgitation, often accompanied by varying degrees of ventricular dysfunction. In this setting, the surgical approach—involving aortic valve replacement combined with replacement of the aortic root and ascending aorta—constitutes a highly complex procedure, typically associated with prolonged operative and CPB times.

These factors may increase susceptibility to cardiopulmonary bypass-related adverse effects, including a dysregulated systemic inflammatory response with subsequent impairment of both macrocirculatory and microcirculatory function. Collectively, these mechanisms may contribute to a more unfavorable hemodynamic profile and worse postoperative outcomes in this subgroup of patients.

From a hemodynamic perspective, this profile was characterized by higher CVP levels and increased Pv-aCO₂, along with a trend—though not statistically significant—toward lower DO₂. Other macrocirculatory and microcirculatory variables, including CI, SVRi, ScvO₂, O₂E, and DO₂/VO₂ ratio, showed patterns consistent with impaired oxygen delivery and utilization. Regarding vasoactive support, Profile 2 required norepinephrine and vasopressin more frequently and at higher doses; however, only the increased use of vasopressin reached statistical significance, with no differences in administered doses.

Clinical outcomes in Profile 2 were significantly associated with major complications, including cerebrovascular events, hospital-acquired pneumonia, acute kidney injury, and hepatic failure. Other outcomes, such as postcardiotomy low cardiac output syndrome, vasoplegic syndrome, need for renal replacement therapy, in-hospital mortality, and 24-h SOFA score, were more frequent in this group but did not reach statistical significance. Logistic regression analysis confirmed the relevance of this profile, showing a 4.3-fold increased risk of cerebrovascular events, a 2-fold higher risk of hospital-acquired pneumonia, a 1.8-fold higher risk of acute kidney injury, and a trend toward higher in-hospital mortality (2.2-fold).

Finally, we propose that the hemodynamic Profile 2, defined by lactate ≥ 2 mmol/L and Pv-aCO₂/ΔAVO₂ ratio ratio > 1, is associated with a higher risk of adverse outcomes. Based on these findings, we suggest a DO2crit threshold of 7.3 ml/min/kg as clinically relevant. We consider it essential that all patients undergoing cardiac surgery be categorized according to hemodynamic profiles, which would allow early identification of those at higher risk of complications and enable timely therapeutic interventions and preventive measures.

Conclusion

In patients undergoing cardiac surgery, the integration of macrocirculatory and microcirculatory parameters enables the identification of distinct hemodynamic profiles with prognostic relevance. Profile 2, characterized by lactate ≥ 2 mmol/L and a Pv-aCO₂/ΔAVO₂ ratio > 1, was associated with impaired DO₂, higher CVP, increased Pv-aCO₂, greater reliance on vasoactive support, and a significantly higher risk of major postoperative complications, including cerebrovascular events, hospital-acquired pneumonia, acute kidney injury, and hepatic failure. Additionally, a DO₂crit threshold ≤7.3 ml/min/kg emerged as a potentially useful marker for identifying patients at increased risk.

These findings underscore the potential value of early hemodynamic profiling using variables such as CI, SVRi, ScvO₂, O₂E, DO₂/VO₂ ratio, and Pv-aCO₂/ΔAVO₂ ratio to guide individualized management strategies. However, given the sample size of the present study, these results should be interpreted with caution. Larger, adequately powered studies are required to validate these associations and confirm the clinical utility of this hemodynamic classification approach in the postoperative setting.

Ethical disclosures

Protection of human and animal subjects

The authors declare that no experiments were performed on humans or animals for this investigation.

Confidentiality of data

The authors declare that no patient data appears in this article.

Right to privacy and informed consent

The authors declare that no patient data appears in this article.

Author contributions

JSSG: original draft writing; RGN: methodology, analysis; GRV: review; DMS: original idea, methodology, analysis, original draft writing, review, and editing.

Ethical considerations

The local research and institutional ethics committees waived approval for this study.

Consent for publication

Written informed consent was obtained for the publication of patient information and images, either from the patient or a legally authorized representative.

Use of artificial intelligence

No artificial intelligence tools were used in the preparation, writing, or analysis of this manuscript.

Financial support

This research did not receive any specific grant from funding agencies in the public, commercial, or non-profit sectors.

Conflict of interest

The authors declare that there are no conflicts of interest to disclose.

Availability of data and materials

The data supporting the findings of this study are available upon request from the corresponding author [DMS].

Acknowledgements

To the Electromechanical Instrumentation Department, the Immunology Department and Outpatient Clinic of the National Heart Institute Ignacio Chávez.