Title: Arterial blood gas, laboratory and portable point-of-care testing for perioperative haemoglobin measurement in patients undergoing major non-cardiac surgeries: A comparative study

Introduction:

Worldwide, close to 85 million red blood cell (RBC) units are transfused annually, with surgical patients accounting for up to 44% of transfusions.1–3 Clinically indicated RBC transfusions in surgery can be life- saving interventions, while inappropriate transfusions provide no clinical benefit and can cause harm.(1) Perioperative transfusions are guided by blood loss assessment and hemoglobin (Hb) values. Hb measured by arterial blood gas (ABG) analyser is convenient, easily available and results are obtained faster than the conventional laboratory methods (Lab). (2) It is well established that hemoglobin measurement plays a central role in any decision to transfuse RBCs. A recent survey of Canadian anaesthesiologists reported intraoperative hemoglobin levels to be the most important parameter for transfusion decision making- more important blood loss or hemodynamics.(3) It is common practice to perform daily blood analysis in intensive care units (ICUs), the results are delayed by a few hours when processed by a central laboratory (Lab) . Blood gas analyzers have been widely into modern ICUs and offer a unique opportunity to determine measurements at the point-of-care (POC), and in a time frame of 2 minutes. (4) Current practice guidelines suggest that the decision to transfuse RBCs in surgery should be generally limited to those with a haemoglobin (Hb) of 60–100 g/L and informed by important clinical parameters. Implicit in this recommendation is the emphasis on measuring Hb at critical times during surgery. Traditionally, Hb measurement during surgery was done through submission of a complete blood count specimen to the main laboratory. More recently, point- of- care testing (POCT) instruments capable of measuring or calculating haemoglobin (POCT- Hgb) have become commonplace in surgery due to their ease of use and rapid delivery of a result. Few clinical trials have compared intraoperative transfusion strategies, the majority of which have limited sample sizes. There are three main classes of POCT-Hgb methods currently in use in operating rooms. The first class yields a Hgb value from 10 µL of whole blood in less than 60 secs. Second class can yield results using 65 µL to 100 µL of whole blood in 120 s using conductometric methods and third class involves non-invasive monitoring of capillary Hgb through multiwavelength sensors and pulse co-ox imetry.(1) As per the quality requirements based on 2024 CLIA (Clinical Laboratory Improvement Amendment) Acceptance Limits for Proficiency Testing (Hematology CLIA 2024), the new criteria for AP for Hemoglobin & Hematocrit is (Test Value) TV ± 4% for both.

To date, POCT- Hgb instruments have been validated with static and normal Hgb values, such as in healthy blood donors. In clinical care, they have been examined primarily in the non- operative setting, such as the outpatient clinic or emergency department, reporting conflicting results in relation to bias and accuracy of POCT-Hgb devices. Their value in surgery, where Hgb can change rapidly due to bleeding and hemodilution from concurrent intravenous fluid administration, are relatively untested. More importantly, there is also a paucity of published validation data based on intraoperative Hgb values within the transfusion zone of 60–100 g/L used for clinical decision- making, highlighting a major gap in existing evaluations.(1)

In our study we plan to determine the reliability and accuracy of Hgb measurements by ABG analyser and POCT-portable Hgb analyser compared to the laboratory method during major non-cardiac surgeries and having a significant amount of blood loss.

Research Question: Is ABG analyser & Point of care portable Hgb analyser accurate in assessing Hgb levels in comparison to Central laboratory method?

Objectives:

Primary Objective:

To compare the accuracy of Hgb measurements obtained by ABG analyser, POCT Portable Hgb analyser with central laboratory method during major non-cardiac surgeries

Secondary Objective:

To evaluate the level of agreement between the measurement techniques of ABG, POCT portable Hgb analyser and laboratory method

Hypothesis:
The hemoglobin values obtained by ABG analyser and POCT portable Hgb analyser might show significant differences compared to the values obtained by the laboratory method in patients having significant amount of intraoperative bleeding.

Methods:

Study design:

Non-randomised

Interventional prospective comparative study

Study Setting:

Department of Anaesthesiology, Critical care & Pain medicine, AIIMS Guwahati.

Study Population:

Inclusion criteria:

·        Age: 18-60 years

·        Undergoing Major non-cardiac surgery with anticipated significant blood loss^*

·        Preoperative Hb>10g/dl

·        Patients belonged to ASA Physical status Class I to III

^*Operational definition for significant intraoperative blood loss for the present study is minimum of 70% of maximum allowable blood loss.

Exclusion criteria:

·        Patients with known hematologic disorders

·        patients with severe coagulopathies

·        Uncontrolled Hypertension

·        Patients on diuretics and anticoagulants

·        Ischaemic Heart disease

Sample Size:

Based on Sunil Ranjan et al(2) using mean differences and standard deviation and Marianne Johnson et al (5) where comparison of Hgb measurements by 3 Point of care devices with standard laboratory values sample size for our study was calculated.

In order to achieve a power of 95%, with an alpha error of 0.05, and a 95% confidence interval, with an effect size of 0.29, we will need approximately 548 samples. We added a 10% drop-out to make the sample number to 603.75(Rounded off to 604). Given that each participant provides 4 samples, 151 participants will be required to get 684 samples. The sample size was calculated for cross-sectional study with Z alpha adjusted value of 2.41 and Z beta of 1.96 for 95% power  comparing three techniques of Hgb estimation was 137.  The sample size was calculated using open epidemiological tool www.openepi.com.

In our study, the Bonferroni correction will be applied to the p-values obtained from the pairwise comparisons between the three techniques at each time point. If there are multiple time points or additional comparisons, the total number of comparisons will be considered.

Example Calculation

1. Number of Techniques: 3
2. Number of Comparisons: 3 (ABG vs. POCT, ABG vs. Lab, POCT vs. Lab)
3. Initial Alpha Level: 0.05

Applying the Bonferroni correction:

Corrected alpha: 0.05/3=0.017

This comprehensive approach ensures that the study is sufficiently powered and accounts for potential dropouts, enhancing the reliability and validity of the results.

After applying the Bonferroni correction, any p-value less than 0.017 would be considered statistically significant instead of the usual 0.05 threshold. This correction helps to maintain the overall Type I error rate at the desired level (5% in this case) across all comparisons. This will help ensure that the conclusions drawn are statistically valid and not due to chance.

In our study, We are enrolling 151 participants will be collecting  4 samples per participant.

This comprehensive approach ensures that the study is sufficiently powered and accounts for potential dropouts, enhancing the reliability and validity of the results.

Procedure:

After satisfying the inclusion criteria all patients are kept fasting for 6 hr preoperatively. After shifting the patient to the operation theatre, standard monitors such as pulse oximeter, non-invasive blood pressure monitor (NIBP), Temperature(^oC) and Electrocardiograph(ECG) are attached. Arterial cannulation under local anaesthesia will be performed using 20G cannula or BD arterial cannula and two large bore intravenous cannula will also be placed.(2) Arterial transducers will be used for invasive hemodynamic monitoring in all non-cardiac surgeries anticipating significant blood loss intraoperatively. Each time an intraoperative Hgb measurement is under-taken by an anaesthesiologist, the patient will undergo three concurrent Hgb measurements using the same sample.(5) The anaesthesiologist will draw a standard 3 ml of blood from the arterial line and run on the first device i.e POCT-portable Hgb analyser (AGAPPE, MispaHbX portable hemoglobin analyzer) which assess Hb/Hct by photometric detection method and about 0.5 to 1 ml of blood is collected in a heparinised syringe and sent for ABG analysis on ABG analyzer (ESCHWEILER Combi line 2, ESCHWEILER GmbH & Co.KG, Kiel, Germany) and remaining 2 ml of blood is sent to the central laboratory in an EDTA vacutainer and tested using Hematology analyzers used in our Institute laboratory i.e (SYSMEX XP-300 & SYSMEX XN-1000) which uses non-cyanide method and Sulfolyser method for hemoglobin estimation respectively. These samples are collected at four time points i.e preoperatively before induction of anaesthesia, Intraoperatively at a point of about 70% MABL(Maximal allowable blood loss), Postoperatively immediately in the PACU/ICU and at 6 hrs in the PACU/ICU.

All patients will receive anaesthesia following a standard protocol. After induction patients will be hydrated with Ringer’s Lactate/Kabilyte at the maintenance rate based on the weight of the patient using the 4-2-1 Estimation of maintenance requirements based on data by Holliday & Segar. During episodes of excess loss, such as when body cavities are open or bleeding occurs, the fluid volumes prescribed are then based on perceived knowledge of the movement of fluids between compartments

Intraoperative blood loss will be assessed by weighing blood soaked surgical swabs, measuring volume of blood loss in suction apparatus, and loss at surgical site will also be taken into account. Up to MABL, intravascular volume will be replaced with either Ringer’s lactate or Kabilyte (based on lactate levels) 3:1 ratio and colloids like Hydroxy ethyl starch  or albumin(if existing hypoalbuminemia) in 1:1 ratio as appropriate to replace blood loss. The Hgb levels <8 g/dl is used as trigger point for transfusion of blood products when there is significant intraoperative blood loss. Intraoperative use of vasopressors and packed red blood cells/Fresh frozen plasma/Platelets transfusion will be documented accordingly. Intraoperative hemodynamic parameters like HR/min, MAP mm/Hg will be documented.(2) The time of sample collection T1, T2, T3 & T4 are documented and also the time of report collection is also documented. The demographic data of the participant i.e Age in years , Sex, Weight in Kg, Height in cm, BMI(Kg/m², Type of Surgery, Emergency/Elective,  ASA class, Preoperative Hgb, Arterial or Venous sample and Type of Anaesthesia are documented. The number of IV fluids used intraoperatively (Crystalloids & Colloids), Use of blood preventative methods (e.g., tranexaemic acid) and amount of blood loss are thoroughly documented. The time of start of surgery and time of end of surgery and also the duration of surgery are documented.

Missing Data:

Missing data is expected with all the three methods of testing Hgb. In very rare instances, there may be missing data for the lab-Hgb value due to clotted blood sample or an insufficient blood sample. As the Lab-Hgb value is the reference standard to assess the precision of POCT-Hgb device & ABG analyser, any Hgb measurement missing data will be excluded from the analysis. Furthermore, during the data collection phase of the study, blood samples with missing lab-Hgb values will not be considered in the total sample size.

Missing data for the POCT portable Hgb analyzer (Mispa HbX, AGAPPE) should be rare, but might occur in instances of device failure (e.g.,if microcuvette is filled inappropriately) or unavailability of device being used in another operating room, or due to insufficient quantity to run all the tests. Considering the very rare instance of missing data in this device, they will be excluded in the analysis.

Reference test(Gold standard) test:

The central lab Hgb sample will be collected in an EDTA vacuum collection tube and delivered to the laboratory for processing in the usual fashion. The POCT portable Hgb analyzer(AGAPPE, MispaHbX portable Hemoglobin Analyzer) and Arterial Blood Gas analyzer(ESCHWEILER Combi-line 2) are considered Index tests. For each device, the manufacturer’s instructions and institutional procedures will be strictly adhered to, including device handling, training , storage, calibration and quality control.(5,6)

Blood gas analysis:

To determine the Hgb level of either arterial or venous heparinised blood samples, ESCHWEILER Combiline 2 Analyzer will be used in our Institute. Hemoglobin level was quantified with using the law of Lambert-Beersch in a secondary analysis of RCT study by Tanner L et al(7) in which If Hb was <0.16 g/dl (0.1 mmol/l) or >40.26 g/dl (25 mmol/l), results were classified as outlier and not considered in the analysis. Blood samples, both arterial and venous, were taken by trained medical staff and transferred to the ABG analyzer immediately. Results were displayed in the patients medical file & documented.

Standardization Techniques

• Calibration and Quality Control:
• All devices (ABG Analyzer, POCT Hemoglobin Analyzer, Central Laboratory Analyzers) will undergo regular calibration and quality control as per the manufacturer’s and institutional guidelines.
• Blood Sample Handling:
• Samples will be drawn and processed uniformly to minimize preanalytical variability.
• Protocol Adherence:
• Strict adherence to study protocols for sample collection, handling, and analysis to ensure consistency.
• Training:
• Personnel involved in sample collection and analysis will receive standardized training to ensure uniformity in procedures.

Outcomes:

Primary outcomes:

• To compare the accuracy of hemoglobin (Hgb) measurements obtained by the ABG analyzer, POCT portable Hgb analyzer with central laboratory method during major non-cardiac surgeries

Secondary outcomes:

1.      Level of Agreement:

o   Assessment of the agreement between the Hgb measurement techniques (ABG, POCT, and central lab) using statistical methods such as Bland-Altman plots or Intraclass Correlation Coefficient (ICC).

Data Collection:

Data will be recorded and documented in standardised forms by blinded outcome assessor

Statistical Analysis:

1. Descriptive Statistics:
• Mean, median, standard deviation, range for Hgb values from each method.
2. Paired t-test or Wilcoxon Signed-Rank Test:
• To compare the Hgb values between two measurement techniques (e.g., ABG vs. POCT, ABG vs. laboratory, POCT vs. laboratory).
• Use the paired t-test if data are normally distributed; otherwise, will use Wilcoxon Signed-Rank test.
3. ANOVA or Friedman Test:
• To compare Hgb values across all three measurement techniques simultaneously.
• Use ANOVA if the data are normally distributed; otherwise, will use the Friedman test for non-parametric data.

Secondary Objective

1. To evaluate the level of agreement between the measurement techniques of ABG, POCT portable Hgb analyzer with central laboratory method

• Bland-Altman Plots was used to assess agreement between two measurement techniques.
• Intraclass Correlation Coefficient (ICC)was used to assess the reliability and agreement among the three measurement techniques.

Summary of Statistical Tests

1. Descriptive Statistics:
• Mean, median, standard deviation, range
2. Comparative Tests:
• Paired t-test or Wilcoxon Signed-Rank Test
• ANOVA or Friedman Test
3. Agreement Tests:
• Bland-Altman Plots
• Intraclass Correlation Coefficient (ICC)