Non-Invasive Hemorrhage Monitoring: A Pilot Study to Assess the Feasibility and Tolerability of the AccuFlow Perfusion Sensor for the Detection of Intrapartum Hemorrhage

Hemorrhage remains the leading cause of direct pregnancy-related death in both the developing world and the developed world. Young, healthy patients compensate for hemorrhage via peripheral vasoconstriction, delaying the emergence of vital sign abnormalities until 15-30% of total blood volume has been lost (1-2L in a term pregnant patient). A significant drop in blood pressure and marked tachycardia (>120bpm) are not seen until 30-40% of blood volume has been lost. After 40% loss, however, patients may rapidly decompensate, leaving a narrow window between the emergence of vital sign abnormalities and rapid clinical decompensation. In facilities where obstetricians and anesthesiologists are not in-house at all times and blood product availability is limited, this narrow window may prove catastrophic.

Furthermore, management of obstetrical blood loss presents many unique challenges. First, during delivery, when hemorrhage is expected and occurs acutely and visibly, blood is often mixed with amniotic fluid, making it difficult to establish actual blood loss. Second, laboratory measurement of hemoglobin concentrations can be falsely normal during an acute episode of bleeding, as the hemoglobin measured does not reflect actual hemoglobin concentration until several hours after the hemorrhage event, when equilibration occurs. Third, unlike other surgical patients, postpartum women are expected to experience ongoing vaginal bleeding after their delivery. Bleeding which is persistently slightly heavier than average can lead to significant blood loss over a period of several days, without either the patient or the staff caring for her recognizing the hemorrhage until the patient becomes symptomatic or vital sign instability occurs. The postpartum uterus can serve as a reservoir for retained blood and clot, allowing a patient to continue bleeding into her uterus after delivery without any external evidence of bleeding. In a thin patient this is frequently recognized on "fundal checks" where the size of the uterus is evaluated, but in obese patients the uterine fundus may be difficult to appreciate, and a significant volume of blood can accumulate in the uterus before it is recognized.

Due to the high risk of mortality related to postpartum hemorrhage and the difficulties providers face quantifying blood loss accurately after delivery,, it is imperative to develop a tool that can detect and quantify hemorrhage before vital sign changes occur. This tool should be simple to use, allowing healthcare providers at all levels to identify hemorrhage early and to mobilize appropriate resources. This will facilitate earlier intervention to treat blood loss before clinical decompensation. As stated, vasoconstriction is the earliest physiologic responses to acute hypovolemia6 and, as such, would be an appropriate target for such a tool.

The AccuFlow Sensor is a device that is based on Combined Heat-Flux and Temperature Sensor (CHFT+) technology developed in the heat transfer laboratory at Virginia Tech and produced by ThermanSENSE Corp (Roanoke, VA). The device, which consists of a heat flux sensor, a thin-film thermocouple, and a heater element, makes direct, quantitative measurements of perfusion occurring at the surface of exposed tissue in real-time. The sensor is less than 1mm thick, covers one square inch of skin, and adheres to the patient's skin using a piece of medical-grade silicone adhesive tape. The heater applies a small amount of heat to the skin surface, raising its temperature to approximately 39 degrees C. The heat flux sensor measures the rate at which the heat dissipates into the tissue, and uses this to calculate perfusion using the Pennes Bio-Heat Transfer Equation, which was developed to describe patterns of heat distribution in the human forearm and which has been the standard tool to predict temperature distributions in living human tissue since the 1950s.10 The sensor measures both surface perfusion and deep perfusion (1-2cm below the skin surface).

The primary theoretical source of patient injury or discomfort from the device is thermal injury from the heater element. Under typical circumstances, the device heats up to approximately 39 degrees Celsius (102 Fahrenheit, the temperature of a typical hot tub), and to prevent thermal injury, the device automatically shuts off should it reach a temperature of 45 C (113F, just above the typical hot-shower temperature of 110F). The device has been tested for comfort and tolerability on 20 healthy volunteers, who reported no discomfort while wearing the device or afterwards. Prior studies have shown no evidence of thermal injury in tissues exposed to a temperature of 39C for prolonged periods, and that exposure to temperatures of 45C causes mild thermal injury only after 150 minutes of constant exposure.11 As this device only heats up while it is taking a measurement, and never higher than 45C, the risk of a thermal injury while wearing the device for the duration of a cesarean section as planned during this study is extremely low. However, should the sensor become uncomfortable to the patient, it can easily be removed. To date, the sensor has been used to evaluate viability of organs intended for transplantation, by measuring blood flow at the surface of the organ.7 Currently, the sensor is being used to evaluate differences in tissue perfusion between pediatric patients with sickle cell anemia and healthy controls in an ongoing research study at Virginia Tech. Of note, the device meets FDA criteria for a nonsignificant risk device, and has been determined to be a nonsignificant risk device by the Institutional Review Board at Virginia Tech/Carilion Clinic for use in adult and pediatric patients at Carilion Clinic facilities. As mentioned before, the patent-pending AccuFlow device is provided by ThermaSENSE Corp. (Roanoke, VA) and a number of sensors will be contributed for the purpose of this study. The device has not yet been studied for detection of obstetric hemorrhage.