Prediction of hematocrit decline and the impact of peri-operative fluid use in lumbar spinal fusion surgery

How Much Does Intravenous Fluid Cause Hematocrit to Drop? Improving Interpretation of Hematocrit Toward Better Post-Hemorrhage Care

INTRODUCTION

Hemorrhage is assessed, at least in part, via hematocrit testing. To differentiate unexpected drops in hematocrit because of ongoing hemorrhage versus expected drops as a result of known hemorrhage and intravenous fluid administration, we model expected post-operative hematocrit values accounting for fluid balance and intraoperative estimated blood loss (EBL) among patients without substantial post-operative bleeding.

MATERIALS AND METHODS

We reviewed patient-level data from the electronic health record of an academic medical center for all non-pregnant adults admitted for elective knee or hip arthroplasty from November 2013 to September 2022 who did not require blood products. We used linear regression to evaluate the association between post-operative hematocrit and predictor variables including pre-operative hematocrit, intraoperative net fluid intake, blood volume, time from surgery to lab testing, EBL, patient height, and patient weight.

RESULTS

We included 6,648 cases. Mean (SD) estimated blood volume was 4,804 mL (1023), mean net fluid intake was 1,121 mL (792), and mean EBL was 144 mL (194). Each 100 mL of EBL and 1,000 mL net positive fluid intake was associated with a decrease of 0.52 units (95% CI, 0.51-0.53) and 2.4 units (2.2-2.7) in post-operative hematocrit. Pre-operative hematocrit was the strongest predictor of post-operative hematocrit. Each 1-unit increase in pre-operative hematocrit was associated with a 0.70-unit increase (95% CI, 0.67-0.73) in post-operative hematocrit. Our estimates were robust to sensitivity analyses, and all variables included in the model were statistically significant with P <.005.

CONCLUSION

Patient-specific data, including fluid received since the time of initial hemorrhage, can aid in estimating expected post-hemorrhage hematocrit values, and thus in assessing for the ongoing hemorrhage.

Assessment of Blood Loss during Neuroendovascular Procedures

(1) Background: Neuroendovascular procedures have generally been considered to have minor or inconsequential blood loss. No study, however, has investigated this question. The purpose of this study is to quantify the blood loss associated with neuroendovascular procedures and identify predictors of blood loss, using hemoglobin change as a surrogate for blood loss. (2) Methods: A retrospective review of 200 consecutive endovascular procedures (diagnostic and therapeutic) at our institution from January 2020 to October 2020 was performed. Patients had to have pre- and post-operative hematocrit and hemoglobin levels recorded within 48 h of the procedure (with no intervening surgeries) for inclusion. (3) Results: The mean age of our cohort was 60.1 years and the male representation was 52.5%. The mean pre-operative hemoglobin/hematocrit was significantly lower among females compared to males (12.1/36.2 vs. 13.0/38.5, p = 0.003, p = 0.009). The mean hemoglobin decrease was 0.5 g/dL for diagnostic angiograms compared to 1.2 g/dL for endovascular interventions (p < 0.0001), and 1.0 g/dL for all procedures combined. In a multivariate linear regression analysis, pre-operative antiplatelet/anticoagulant use was associated with a statistically significant decrease in hemoglobin. (4) Conclusions: Our data support that blood loss from diagnostic angiograms is marginal. Blood loss in endovascular interventions, however, tends to be higher. Pre-operative blood antiplatelet/anticoagulant use and increasing age appear to increase bleeding risk and may require closer patient monitoring.