IntroductionShock is circulatory insufficiency, inadequate oxygen delivery, and cellular hypoxia. Intravenous fluids are essential for shock management. Despite treatment, patients can face persistent shock with ongoing hypotension, contributing to higher mortality. This analysis aims to quantify hypotensive non-traumatic cases in an Australian ambulance service, determine persistent hypotension prevalence, and assess paramedic-administered intravascular fluids' impact on blood pressure changes.MethodsThis study is a retrospective analysis of prehospital fluid resuscitation by New South Wales Ambulance paramedics during 2022. Hypotension is defined as a systolic blood pressure of ≤ 90 mmHg, and persistent hypotension is a systolic blood pressure consistently below 90 mmHg across all observations, with a final blood pressure below 90 mmHg. This study aimed to determine the volume of fluid resuscitation at which a plateau in population-level systolic blood pressure response is observed, by calculating the derivative of the fitted logistic regression model. Moreover, this analysis identified the relative contribution of factors influencing the probability of an attempt at intravenous or intraosseous access using machine learning.ResultsAmong 796,865 attendances, 23,049 (2.9%) involved non-traumatic patients with hypotension. In total 7,388 (32.1%) of the hypotensive cases resulted in persistent hypotension, of which 3,235 (43.8%) received Hartmann's solution and 1,745 (53.9%) received at least 500 ml of fluids but still had hypotension. The model showed that systolic blood pressure tends to stop increasing after 500-600 millilitres of fluid are given. This suggests that, on average, giving more fluid than this may not raise blood pressure further in a prehospital setting, though individual patient needs can differ. The top four factors from the machine learning shows that as initial respiratory rate goes up, the probability of intravascular access rises. Transport times less than 20 minutes are associated with a smaller chance of access and younger patients are less likely to receive an attempt. Finally, extremes of systolic blood pressure are more likely to receive access attempts.ConclusionThis study found that three percent of non-traumatic attendances have at least one episode of hypotension, and that more than half of these have persistent hypotension. Only 44% of persistently hypotensive received fluids, and half of persistently hypotensive patients stayed hypotensive despite a reasonable volume of prehospital crystalloids.
Keyphrases
- blood pressure
- hypertensive patients
- heart rate
- end stage renal disease
- cardiac arrest
- spinal cord injury
- machine learning
- chronic kidney disease
- heart failure
- prognostic factors
- left ventricular
- blood glucose
- type diabetes
- patient reported outcomes
- cardiovascular disease
- high dose
- artificial intelligence
- skeletal muscle
- risk factors
- coronary artery disease
- metabolic syndrome
- patient reported
- deep learning
- data analysis
- big data
- insulin resistance
- smoking cessation
- combination therapy
- replacement therapy