Effect of Red Blood Cell Transfusion on Central Venous-to-Arterial Carbon Dioxide Difference in Anemic Surgical Patients A Pilot Study

Special Article- Unusual Bleeding

Ann Hematol Onco. 2023; 10(4): 1431.

Effect of Red Blood Cell Transfusion on Central Venous-to-Arterial Carbon Dioxide Difference in Anemic Surgical Patients – A Pilot Study

Anton Alpatov1; Maria Wittmann1; Heidi Ehrentraut1; Achilles Delis1; Jochen Hoch2; Andreas C Strauss3; Holger Bogatsch4; Patrick Meybohm5; Markus Velten1#; Tobias Hilbert1#*

1Department of Anesthesiology and Intensive Care Medicine, University Hospital Bonn, Bonn, Germany

2Institute for Experimental Hematology and Transfusion Medicine, University Hospital Bonn, Bonn, Germany

3Department of Orthopedics and Trauma Surgery, University Hospital Bonn, Bonn, Germany

4Clinical Trial Centre Leipzig, Leipzig, Germany

5University Hospital Würzburg, Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, Würzburg, Germany

*Corresponding author: Tobias Hilbert Department of Anesthesiology and Intensive Care Medicine, University Hospital Bonn, Venusberg-Campus 1, 53127 Bonn, Germany Email: thilbert@uni-bonn.de

#These authors have contributed equally to this article.

Received: June 10, 2023 Accepted: July 10, 2023 Published: July 17, 2023

Abstract

Background: Biochemical markers for monitoring adequacy of cardiac output and tissue perfusion such as blood lactate and central venous oxygen saturation (ScvO2) are meanwhile well established in clinical routine. In addition, in recent years, the central venous-to-arterial carbon dioxide difference (dCO2) has been evaluated as a further marker, and studies meanwhile have demonstrated the validity of an increased dCO2 to identify capillary perfusion mismatches. However, results from animal studies suggest that dCO2 may be influenced by altered hemoglobin values during severe bleeding. It was the aim of our study to evaluate if dCO2 changes upon Red Blood Cell (RBC) transfusion in humans.

Methods: Patients of the ongoing LIBERAL trial were prospectively evaluated. Participants were aged ≥70 years and scheduled for elective intermediate or high risk orthopedic or trauma surgery with the clinical need for invasive blood pressure monitoring and central venous catheterization. During surgery, drop of hemoglobin triggered administration of one single RBC unit, together with arterial and central venous blood analysis immediately before as well as after transfusion.

Results: In total, 46 patients were analyzed. Baseline median hemoglobin before RBC transfusion was 8.35 (7.48–8.73)g/dl, while dCO2 was 6.2 (3.4–9.6)mmHg. According to Spearman correlation, there was a linear association between pre-transfusion dCO2 and ScvO2. Transfusion of one RBC unit resulted in a significant increase of median hemoglobin by 1.2 (0.7–1.63)g/dl (p<0.0001), and hemoglobin increase was more pronounced when pre-transfusion hemoglobin was low, as evidenced by a significant negative association between both parameters (r=-0.61, p <0.0001). Neither lactate nor ScvO2 nor dCO2 were significantly influenced by transfusion. When the whole cohort was divided according to pre-transfusion dCO2 levels using a cut-off value of 6 mmHg, median dCO2 decreased significantly more pronounced following RBC transfusion when pre-transfusion values were high (>6 mmHg), compared to those patients with a pre-transfusion dCO2 below 6 mmHg.

Conclusions: The results of our study suggest that crude dCO2 is not influenced by moderate hemoglobin increases in orthopedic and trauma surgery patients. However, including dCO2 into the decision whether to administer RBC or not may be an interesting reasonable approach for further investigations on the way towards more individualized transfusion regimens.

Keywords: Transfusion; Bleeding; Capillary perfusion; Microcirculation; Cardiac output; Central venous oxygen saturation

Background

The purpose of hemodynamic optimization is to maintain adequate tissue perfusion. During capillary perfusion, oxygen is delivered to organs and tissues, while Carbon Dioxide (CO2) produced during cell metabolism is washed out. Impaired microcirculation disrupts this exchange with the risk of tissue loss and organ death due to a critical mismatch between oxygen demand and delivery.

It is meanwhile recognized that basic and also advanced monitoring solely taking systemic hemodynamic variables (e.g., arterial blood pressure, cardiac output, global volume status) into account may only poorly reflect actual tissue perfusion [1,2]. Administration of inotropics, vasopressors and fluids in reaction to hypotension or cardiac output failure irrespective of knowing actual capillary perfusion may further worsen the situation. This concept of dissociation of micro- and macrocirculation is called the loss of hemodynamic coherence, reflecting that optimizing systemic surrogate variables does not necessarily result in restored microcirculation but sometimes rather leads to the opposite [2-4]. Therefore, ways to assess capillary perfusion and tissue oxygenation to guide hemodynamic optimization are mandatory.

In addition to technically more advanced approaches such as side stream dark field imaging or intravital tissue oxygen assessment electrodes that are complex and therefore only available in limited form [4], biochemical markers for monitoring adequacy of cardiac output and tissue perfusion are well established in clinical routine. As such, lactate and mixed (or central) venous oxygen saturation (SmvO2 and ScvO2) have become widely used as easily assessable parameters to validly guide fluid and catecholamine support and transfusion in critical situations such as bleeding, sepsis or cardiogenic shock [5]. In addition to that, in recent years, the central venous-to-arterial carbon dioxide difference (dCO2) has been introduced and evaluated as a further marker for capillary perfusion mismatches. CO2 produced during cell metabolism is either combined with water to form bicarbonate, bound to hemoglobin or physically dissolved in blood. As these, it is carried from the capillary bed via venules and larger veins to the pulmonary alveolar system for exhalation. As the principle of Fick not only applies for mixed venous oxygen saturation but also for dCO2, CO2 production is proportional to cardiac output and capillary perfusion, and therefore so is the difference between CO2 partial pressure on the arterial and the one on the venous side. Studies meanwhile have demonstrated the validity of an increased dCO2 to identify capillary perfusion mismatches, even when ScvO2 is normal [6,7]. In critically ill patients, a cut-off value of 6 mmHg has been suggested to reflect the adequacy of tissue perfusion [8]. However, dCO2 should be interpreted with caution in case of, e.g., respiratory alkalosis [9]. Moreover, results from animal studies suggest that dCO2 may be influenced by altered hemoglobin values during severe bleeding [10]. Therefore, it was the aim of our study to evaluate if dCO2 changes upon Red Blood Cell (RBC) transfusion in humans. Orthopedic and trauma surgery patients with both arterial and central venous catheters and clinical indication for RBC transfusion were evaluated.

Methods

The present study recruited consecutive patients from orthopedic or trauma surgery as part of a subgroup of the ongoing LIBERAL trial during August 2019 to January 2023. The LIBERAL trial is a prospective multicenter, randomized, open phase IV trial, investigating the effects of a liberal transfusion strategy of RBCs on mortality and anemia-associated ischemic events in elderly patients undergoing non-cardiac surgery. It is funded by the German Research Foundation (DFG, protocol no. ME 3559/3-1). For a detailed description of the protocol of the LIBERAL trial, see Meybohm et al [11]. All analyses were performed in accordance with the Declaration of Helsinki. The leading ethics committee (University Wuerzburg 87/17_ff) and local ethics committee (University Hospital Bonn, Germany; protocol number 096/17-AMG) considered the study (clinical trials: NCT03369210) to be compliant with the applicable professional codes and regulations and thereby approved the study protocol. Written informed consent was obtained from all included patients. In brief, patients aged =70 years and scheduled for elective intermediate or high risk orthopedic or trauma surgery with the clinically indicated need for invasive blood pressure monitoring and central venous catheterization were included.

Exclusion criteria comprised emergency surgery, refusal or inability to provide written informed consent, preoperative severe anemia with hemoglobin levels below 9.0g/dl, chronic kidney injury requiring dialysis, participation in other interventional trials, and preoperative autologous blood donation. Arterial as well as central venous catheterizations were performed by the attending anesthetist who was not part of the study personnel.

Hemoglobin levels were monitored during surgery. If they dropped below =9.0g/dl, patients were randomized either to a liberal (receiving one single RBC unit each time hemoglobin reaches =9.0g/dl) or restrictive transfusion regime (a single RBC unit each time hemoglobin reaches =7.5g/dl). Immediately before and 10 minutes following transfusion of one single RBC unit, a central venous and an arterial blood sample were taken simultaneously and the following parameters were obtained using a Siemens Rapidpoint 500 blood gas analyzer (Siemens Healthineers, Erlangen, Germany):

- Hemoglobin

- Hematocrit

- Blood lactate

- Carbon dioxide partial pressure (pCO2)

- Central venous oxygen saturation (ScvO2)

The central venous-to-arterial carbon dioxide difference (dCO2) was calculated as follows:

dCO2 = pCO2 (central venous) - pCO2 (arterial)

Additional data recorded comprised:

- Body weight

- Body height

Body Mass Index (BMI) was calculated using the formula: BMI [kg/m²] = body weight [kg]/body height [m]².

Statistical analyses and visualizations were performed using MS Excel 2019 (Microsoft Corp., Redmond, CA, USA) and GraphPad PRISM 8 (La Jolla, CA, USA). Data are presented as median values with interquartile range (25-75) and were analyzed using Mann-Whitney test or Wilcoxon signed rank test, respectively, and Spearman correlation. The alpha level was set to 0.05. A sample size calculation prior to recruiting of the first participant revealed that at least 46 subjects would be required for sufficient power (given an effect size of 40%, a type-I error probability of 0.05 and a power of 85%). All data sets are available from the author upon reasonable request. This investigation did neither unblind any data nor analyze any group differences of the large LIBERAL trial.

Results

In total, 50 patients were prospectively included to participate in the study. Four cases were excluded due to incomplete sampling, resulting in 46 arterial and venous blood samples before and 46 arterial and venous samples after RBC transfusion (184 blood samples in total). Median patient age was 79 (75–83) years. The cohort comprised 46 patients, receiving upper or lower limb surgery in 25 and spine surgery in 21 cases. Median body weight was 76 (64–89) kg, height was 168 (162–178) cm, and Body Mass Index (BMI) was 26.6 (23.7–31.1) kg/m².

Baseline median hemoglobin before RBC transfusion was 8.35 (7.48–8.73) g/dl in arterial blood samples (Figure 1), while hematocrit was 24.5 (22–26) %. Blood lactate was 0.97 (0.84-1.38) mmol/l. Median arterial carbon dioxide partial pressure (pCO2) was 40.7 (37-44.6) mmHg. In central venous blood samples, median pCO2 was 47.2(43.6–50.6) mmHg, while oxygen saturation (ScvO2) was 78 (71–82) %. Before transfusion, median dCO2 was 6.2 (3.4-9.6) mmHg. According to Spearman correlation, there was a linear association between pre-transfusion dCO2 and ScvO2, with higher dCO2 values being associated with a lower ScvO2 (r=-0.37, p=0.01). There was no correlation between pre-transfusion hemoglobin and dCO2, ScvO2 or lactate.