Estimating Blood Loss Using TritonTM in Vaginal Deliveries: A Validation Trial

Overview

This study will be a prospective cohort study. Patients who meet criteria for inclusion in the study will be approached for participation at same day of admission. Written informed consent will be obtained from the patients by the Co-principal investigator and by the study collaborators. If patients agree to participate, a CBC (complete blood count) will be obtained via venous puncture routine in our facility as part of the admission labs which will be around 10 cc of blood. The device will be used during the delivery in laboring room. The device will be used to assess QBL (quantitative blood loss) by the research staff only and results/ QBL assessment will be masked to the clinical team. Unmasking will only occur following study completion with purpose to perform data analysis. Patient management will be according to the clinical team without the knowledge of the QBL. All patients undergo a CBC postpartum as part of post-partum evaluation, this will also be performed by venipuncture where 10 cc of blood will be collected. The drop in Hgb (hemoglobin) between the pre and post partum CBCs will be calculated for each patient. The post-partum CBC will be collected approximately 24-30 hours from delivery as standard in our unit. The blood will be collected from each patient by the nursing staff who are experienced in withdrawing blood. Patients will be divided into quartiles of Hgb. Cases will be those patients whose Hgb is in the upper quartile, while controls will be those patients whose Hgb is in the lower 3 quartiles. We will be comparing visual EBL (estimated blood loss) by standard clinical assessment versus the QBL result from the device between cases and controls. The Triton L\&D (labor and delivery) system which comprises of the device, software analysis and staff training will be supplied by the manufacturer free of charge. Research staff will be trained by the manufacturer. We will be offering our skills, fellows, midwifes and residents, who will be collecting data and we will be performing the data analysis. Results will be available to the manufacturer after results are completed. The results of this study will be presented in conferences or published in a peer-review journal. Demographic information will be obtained from the electronic medical record. The data will be kept on a password secured University of Texas Medical Branch (UTMB) computer. An encrypted USB flash drive will be used to transfer data. The data will be identified and linked to the patient using the medical record number. During data analysis, all patient identifiers will be deleted.

Postpartum hemorrhage (PPH) is the leading cause of maternal mortality in low-income countries and the primary cause of nearly one quarter of all maternal deaths globally (1). An estimated blood loss (EBL) in excess of 500 mL following a vaginal birth has often been used for the definition of PPH, but the average volume of blood lost at delivery can approach these amounts when actually measured rather than estimated (2). More than half of all maternal deaths occur within 24 hours of delivery, most commonly from excessive bleeding (3). Worldwide, 140,000 women succumb to postpartum hemorrhage each year. The most common antecedents to postpartum hemorrhage are uterine atony, placental disorders, and trauma during delivery. Improving maternal health worldwide is one of the WHO's 8 Millennium Developmental Goals. The prevention and treatment of PPH is an essential step towards the achievement of that goal (4). Estimates of blood loss at delivery are notoriously inaccurate, with under-estimation more common than over-estimation (5). Traditionally, the clinicians performing the vaginal delivery would estimate the blood loss by visually assessing the blood collected in the delivery drape drain and counting the number of lap sponges used thru out the delivery. Current detection and management of hemorrhage is heavily based on clinical judgment, which often leads to delay in recognition and intervention. Often, interventions such as fluid resuscitation and blood transfusion are not initiated until significant hemorrhage has already taken place. The traditional method for estimating blood loss is based on the clinician and nursing staff's subjective assessment that is severely limited by human error and the presence of large volumes of amniotic fluid (6). Early detection and treatment of this potentially life-threatening obstetric complication is of utmost importance in the field of obstetrics. Simulations and didactic training have been shown to improve visual estimations, but there are still poor associations between experience level and accuracy, and a significant decay in blood loss estimation skills over time (7). The Triton L\&D system (Gauss Surgical, Inc., Menlo Park, CA) is an FDA-cleared mobile application on a tablet computer (iPad) that facilitates quantification of blood loss (QBL) by providing an easy to use process and user interface. Dry weights of all potential blood containing substrates are built-in to the device allowing batch weighing with automatic subtraction of dry weights. There is also a V-drape simulator accounting for collected fluids with automatic subtraction of a measured amount of amniotic fluid. QBL is not regularly used at UTMB. Although widely recommended, little data is available to support its use in the obstetrical population. We believe that gathering further evidence regarding its value is appropriate. Our hypothesis is that use of this device for QBL will enable clinicians to objectively measure blood loss in real-time. This study will be a prospective cohort study, in which we will evaluate two methods of evaluating blood loss during vaginal delivery (usual visual EBL assessment versus Device QBL). Of note, the subjects consented will be used as self-controls.