Blood loss quantification during major abdominal surgery: prospective observational cohort study

Background Blood loss during major abdominal surgery is an essential parameter in the evaluation of strategies aimed at reducing perioperative bleeding. However, blood loss quantification remains unreliable and inaccurate. The aim of this study was to compare several methods of blood loss quantification—visual estimation by surgeon and anesthesiologist, the gravimetric method, the calculation method with spectrophotometric measurement. The spectrophotometric measurement is considered as the most accurate method. Methods The study was designed as a prospective observational cohort single-center study. We analyzed 61 patients who underwent elective liver or pancreatic resection. The anesthesiologists’ and surgeons' estimate of blood loss was based on a visual assessment. The gravimetric method was based on weighing the suction canister and surgical drapes before and after use. The basis of calculation method was anthropometric and hematological parameters, we used López-Picado's formula. The spectrophotometric method was based on the spectrophotometric determination of hemoglobin mass in the lost blood. We compared the methods using paired t-test, non-parametric Wilcoxon test and Bland–Altman analysis. Results Visual estimation by surgeons and anesthesiologists, gravimetric measurement, and calculation method were significantly different from spectrophotometric measurement at the significance level α = 0.05. All methods overestimated blood loss which was measured by spectrophotometric method. The estimate by surgeons was the closest to the spectrophotometric measurement, difference 68.7 ml (95% confidence interval [CI]: -129.3–-8.2). Conclusions We conclude that the estimate of blood loss by surgeons and anesthesiologists, as well as gravimetric method and calculation method are all significantly inaccurate in real surgical setting. We did not confirm the commonly accepted assumption that surgeons underestimate the blood loss. Trial registration The study was registered under the title " Blood Loss Quantification During Major Abdominal Surgery" at ClinicalTrials.gov with the registration number NCT05316649. Date of the first registration was 20/3/2022.


Introduction
Blood loss estimate is an important parameter of quality of the surgical procedure, along with postoperative morbidity and mortality, radicality of the surgical resection, number of retrieved lymph nodes, duration of the surgical procedure, duration of the hospital stay, or some other parameters (e.g.postoperative pancreatic fistula in case of pancreatic resections).In addition, intraoperative blood loss is an essential component of many scoring systems used to predict outcomes such as morbidity, mortality, and readmission rates.Quantified blood loss also plays a key role in blood transfusion decisions, along with other information such as hemoglobin values and individual transfusion triggers.Negative impact of intraoperative blood loss on outcomes has been well characterized in many studies.However, blood loss estimation remains unreliable and inaccurate [1][2][3][4][5].Therefore, it is very important to have more reliable methods to measure the blood loss.The methods used to estimate or to measure blood loss are as follows: visual estimation, gravimetric method, direct measurement, photometry, calculation methods, colorimetric analysis, and miscellaneous [6,7].
Visual estimation of blood loss by surgeon and anesthesiologist is still the standard method of choice in most institutions [8].Irrigation fluids, lymph, bile, serum, ascites, urine, and other fluids often combine with lost blood, but do not alter its appearance to an extent that is typically appreciable visually, which can affect estimated blood loss [9].Visual estimation of operative blood loss is unreliable and inaccurate.No provider specialty, level of experience, or self-assessment of ability is associated with improved estimation [10].
The gravimetric method was first described by Wangensteen and depends on weighing surgical sponges before and after surgical use [11].Estimated blood loss is determined by assessing the weight difference before and after use, with every gram of weight equivocal to 1 mL of blood loss [12].This method is easy but neither precise nor accurate [13], especially with increased dilution by rinsing fluid [14,15].
Direct measurement of blood loss is a simple and long-established method that is mainly used in the field of obstetrics [6].A calibrated collection bag with a scale is designed for vaginal deliveries.Current blood loss can be immediately read from the scale.Study results show significant deviations from real blood loss [16,17].
Spectrophotometry is the most precise, but also the most expensive method and the most complex method to measure the blood loss [7].Hemoglobin mass loss is assessed in the lost blood using the spectrophotometric method.Spectrophotometric measuring of hemoglobincyanide is the reference method for hemoglobinometry in human blood recommended by the International Council for Standardization in Hematology since 1967 (ICSH) [18].
Calculation Method is based on anthropometric and hematological parameters.There are several mathematical formulas and they have been modified over time: Liu, Mercuriali, Bourke, Ward, Gross, Lisander, Meunier, Camarasa, Lopez-Picado [19].All calculation formulas require an estimation of the total blood volume of the patient.The formulas take into account height, weight, body surface area and gender of the patient.All blood loss estimation formulas showed a significant tendency to overestimate blood loss [6].
The aim of the study is to compare several methods of blood loss quantification-visual estimation by a surgeon and an anesthesiologist, gravimetric method, calculation method and the spectrophotometric method in real surgical settings.
There is still no gold standard in blood loss measurement, because accurate measurement of blood loss is difficult.The most accurate and reference methods are based on "hemoglobin extraction analysis" using spectrophotometry [6].Visual estimation method was chosen because this method is most frequently used in a real surgical setting and it is the easiest method.Gravimetric method was chosen because it is a standard method to measure blood loss in certain surgical fields (gynecological surgery, orthopedic surgery, etc.).
We hypothesized that visual estimation, gravimetric measurement, and calculation method will significantly differ from measured hemoglobin loss by spectrophotometry.

Participants
The study was designed as a single-center prospective cohort study.All patients who were scheduled for elective HPB surgery were screened and assessed for eligibility.A total of 61 adults patients undergoing elective liver or pancreas surgery between May 2021 -June 2022 at Department of Surgery, University Hospital Hradec Kralove, Czech Republic were enrolled to the study.We decided for patients undergoing elective HPB surgery because expected estimated blood loss over 200 ml was needed for the statistical analysis [12].The inclusion criteria were adult patients undergoing elective liver or pancreas surgery, age of patient ≥ 18 years, signed informed consent.The exclusion criteria were patient's coagulation disorder (congenital or iatrogenic due to the chronic use of anticoagulants), use of cell saver suctioning during operation, clotting/damage blood samples, lack of compliance, informed consent not provided or refusal to participate.
The study was registered under the title " Blood Loss Quantification During Major Abdominal Surgery" at Clini-calTrials.gov with the registration number NCT05316649.

Study variables
The basic clinical data of the 61 patients were collected, namely age, height, weight, body mass index (BMI), use of anti-coagulants, comorbidities, physical status (American Society of Anesthesiologists classification system score, ASA), operative time, perioperative catecholamine use, blood transfusions, fluids intake, volume of irrigation fluids, diuresis, hospital stay.

Study outcome
Patients venous blood samples for blood count (including hemoglobin concentrations) were drawn before incision, and after surgery at 1, 24 and 48, 72 h, on the postoperative day 5 and at the end of hospital stay using an automated hematology analyzer XN-10 (Sysmex, Kobe, Japan).
The suction canister and surgical drapes are weighed before and after the surgical procedure with a Kern, PCB 6000-0 with an accuracy of ± 1 g (Balingen, Germany).Gravimetric blood loss (vGBL) is determined by assessing the weight difference after subtracting weight of added fluids [11].Every gram of weight difference equivocal to 1 mL of blood loss [12].
The suction canister was heparinized before surgery (10,000 IU of heparin in 100 ml saline solution) to prevent clotting.The total volume contained in the canister was measured after the end of the surgical procedure by a system capable of determining differences up to ± 10 mL, weighted with a Kern, PCB 6000-0 with an accuracy of ± 1 g (Balingen, Germany) and analyzed for hemoglobin concentrations by spectroscopy using cell count in "body fluid" mode, which is more sensitive to lower cell counts in fluids.
The volume of irrigation fluids used during surgical procedure was carefully recorded as well as infusions, injections and transfusion volume.If necessary, vasoactive agents were titrated to obtain a mean blood pressure of > 65 mmHg during procedure.
Surgery time duration (skin to skin) was recorded.A daily visit of the study patients was made by clinical investigators or a delegated physician.All protocolrequired information collected during the trial were entered into the patient's record form.
Hemoglobin mass loss (hbMBL) for each case is calculated using the spectrophotometric measured hemoglobin concentration from the suction canister.This value is multiplied by the total volume of the suction canisters and the calculated fluid volume from surgical drapes.
To obtain total lost hemoglobin mass loss in g: where fluid volume from surgical drapes is calculated as: The spectrophotometric measured blood loss (vMBL) in mL is calculated using measured hemoglobin mass loss (hbMBL) and patient's average pre-and postoperative hemoglobin, the vMBL is obtained: For calculated blood loss (vCBL) was used López-Picado's formula [19] based on anthropometric and hematological parameters: where Hct i (initial hematocrit) is the patient's preoperative hematocrit, Hct f (final hematocrit) is the patient's postoperative hematocrit.Postoperative time point of the final hematocrit is not specified in the original formula, therefore in accordance with another studies (1) Hct f in this trail is determined 48 h after surgery or when hematocrit reached the nadir level after operation.The transfused RBC volume is calculated as follows: 1 Unit packed homologous blood = 450 mL × hematocrit of the transfused blood; 1 Unit packed autologous blood = 450 mL × hematocrit in the pre-surgical anesthesia consultation.Hct mean is the mean hematocrit (between Hct i and Hct f ).EBV is the estimated blood volume determined using the ICSH formula [18]: At the end of the operation, a blood loss estimate was obtained from surgeon (sEBL) and anesthesiologist (aEBL).

Statistical analysis
Statistical analyses were performed using NCSS statistical software (NCSS, Kaysville, UT, USA).Categorical data are expressed as absolute and relative frequencies.Quantitative data are expressed as means and standard deviations if normally distributed or medians and interquartile ranges in case of unnormal distributed data.Normality was tested with the Shapiro-Wilk test.
In pairwise comparison of methods, when normality was not rejected, the paired t-test was used.When normality was rejected the non-parametric Wilcoxon test was used.A P-value < 0,05 was considered statistically significant.Bland-Altman analysis with 95% confidence interval was used for the degree of agreement between the spectrophotometry method and other methods.

Sample size calculation
The sample size calculation was based on the data from a previous study [9].Power calculations revealed that a sample size of 54 pairs would be needed to detect a twofold difference between two methods with 95 percent power to detect the mean difference of 100 mL between these two methods.With an expected dropout rate of 10%, we planned to enroll 60 patients into the study.
All methods overestimated spectrophotometric measurement of blood loss as shown Table 3.The surgeons' estimate was the closest to the spectrophotometric measurement.Estimated blood loss by surgeon differ from spectrophotometric measurement less than ± 463mL with 95% probability as shown in the Fig. 1.

Discussion
There are several methods for blood loss quantification in surgical procedures.In this study, we compared all the methods in a real surgical setting.We hypothesized that visual estimation by surgeons and anesthesiologists, as well as gravimetric measurement and calculation method would be significantly different from the most accurate method, e.g.measured hemoglobin loss by spectrophotometry (p < 0.05).All the methods overestimated the blood loss compared to the accurately measured blood loss by spectrophotometry.
Visual estimation of blood loss is still the standard method of choice in most institutions [8].Even though the visual estimation by surgeon in this study was the closest to the measured hemoglobin loss by spectrophotometry; -68.7 ml (95% confidence interval [CI]: -129.3--8.2), the surgeons' estimate also shows significant variations.Irrigation fluids, lymph, bile, serum, ascites, urine, and other fluids often combine with lost blood, but do not alter its appearance to an extent that is typically appreciable visually, which can affect estimated blood loss [9].We proved that the visual estimation of blood loss is unreliable and inaccurate.No specialty, level of experience, or self-assessment of ability is associated with improved estimation [10].We did not confirm the commonly accepted assumption that the surgeons underestimate blood loss and the anesthesiologists overestimate it [13].In our study, even surgeons slightly overestimated blood loss, however, not as much as the anesthesiologists.
This study confirms that one of the oldest methods of quantifying blood loss gravimetric method described by Wangensteen [11] is easy but neither precise nor accurate [13], especially with increased dilution by rinsing fluid [14,15].The mathematical approach to evaluate blood loss is based on anthropometric and hematological parameters [19].It is important to note that all blood loss estimation formulas showed a significant tendency to overestimate blood loss [6].Although this method of measuring blood loss eliminates the need for perioperative weighing, measuring surgical drapes and canisters, the results are significantly inaccurate compared to the spectrophotometric measurements, as we demonstrate in this study using the Lopez-Picado formula.Fig. 1 Bland-Altman plot of differences between Spectrophotometric measurement and Estimated blood loss by surgeon (sEBL).In the Bland-Altman plot, bias is represented by a solid red line, 95% limits of agreement are indicated by solid blue lines It is very important to quantify the blood loss more accurately because blood loss is an important parameter of quality of the surgical procedure.There are numerous studies which corelate blood loss with postoperative morbidity or mortality.However, with inaccurate blood loss estimate, unreliable findings can be reached.Blood loss quantification also play an important role in blood transfusion decisions.Inaccurate blood loss estimate can cause unnecessary blood transfusion application.Inappropriate transfusion of blood products is associated with increased risks and negatively influences patient´s outcome and long-term survival in oncological patients.
Similar results were reached by Perri et al. [20].The authors performed a systematic review of original studies published between 2006 and 2021 reporting the blood loss in patients undergoing pancreatic or hepatic resection.The authors concluded that standardization of intraoperative blood loss quantification is urgently needed in HPB surgery to ensure the consistency of reporting the results of blood loss comparison with other factors [20].
The main limitation of this study is that fluid volume from surgical drapes was indirectly calculated through the weight before and after use and not directly measured.It is possible that the concentration of hemoglobin in surgical drapes is slightly different from the concentration of hemoglobin in suction canister.In our study, we assumed that these concentrations are alike, as it was shown in the study by Thomas et al. [9].However, it would be more accurate if hemoglobin losses from surgical drapes were measured directly separately.Nevertheless, it would make the measurement methodology even more difficult in real surgical conditions.

Conclusion
In agreement with previous studies [6,[8][9][10]13] we conclude that the assessment of blood loss using surgeon and anesthesiologist estimates, gravimetric methods, and calculation methods are all significantly inaccurate.All studies which use simple blood loss estimate and corelate it with other parameters should take it into account.We did not confirm the commonly accepted assumption that the surgeons underestimate blood loss and the anesthesiologists overestimate it [9].In our study, even surgeons slightly overestimated blood loss.Surprisingly, this study showed that calculation method and gravimetric method are even less accurate than simple surgeon's visual estimate.
Spectrophotometric measurement of blood loss in real surgical setting is difficult, and is not suitable for routine use due to the economic or methodological reasons.Therefore, despite inaccuracy, surgeon's estimation is very likely to remain the most used method of blood loss in most departments.
• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year

•
At BMC, research is always in progress.

Learn more biomedcentral.com/submissions
Ready to submit your research Ready to submit your research ?Choose BMC and benefit from: ? Choose BMC and benefit from: hbMBL = hemoglobin concetration from canister * (suction canister volume+fluid volume from surgical drapes) fluid volume from surgical drapes = (suction canisters volume) × weight difference of surgical drapes before and after use weight of suction canisters volume vMBL = hbMBL (measured hemoglobin mass loss in g) mean (pre and postoperative) hemoglobin (in g/dL) vCBL = EBV × Hct i − Hct f + transfused RBC volume Hct mean

Table 1
Descriptive statistics ASA American Society of Anesthesiologist physical status classification, BMI body mass index, ICSH International Council for Standardization in Haematology, IQR interquartile range, SD standard deviation

Table 2
Pairwise comparison of the methods of blood loss quantification

Table 3
Bland-Altman analysis of the methods