https://doi.org/10.1007/s00392-024-02526-y
1Universitätsklinikum Essen Klinik für Kardiologie und Angiologie Essen, Deutschland
Intraoperative hypotension (IOH), defined as a low mean arterial pressure (MAP), is associated with deterioration of renal function, myocardial injury, and poorer clinical outcomes. Many studies examined the association of different blood pressure thresholds and duration of hypotension with clinical outcomes in patients undergoing noncardiac surgery. Although it is known that hypotension occurs during electrophysiological procedures, there is no sufficient information concerning its duration and severity. Novel non-invasive devices using artificial intelligence (AI) technology provide real-time information concerning blood pressure and upcoming hypotension.
Aim:
This observational study aims to investigate intraoperative hypotension events measured with continuous non-invasive blood pressure monitoring during invasive electrophysiological procedures.
Methods:
We included patients undergoing electrophysiological procedures such as 3D mapping and/or Ablation for Atrial Flutter or Atrial Fibrillation at the Department of Cardiology and Angiology at Essen University Medical Centre, Essen, Germany. The blood pressure was measured continuously and non-invasively with a novel cuff sensor technology. Episodes of hypotension were defined as MAP<65 mmHg for at least 20 seconds and the threshold for severe hypotension was MAP<50 mmHg. The severity of hypotension was measured by the median area under the curve (AUC) and timeweighted average (TWA) of AUC with MAP<65 mmHg. We examined the number, duration, and severity of hypotension events during electrophysiological procedures. Also, the accuracy of a hypotension prediction index (HPI) calculated by an AI algorithm was analyzed. In all procedures, the system worked as a bystander in addition to conventional blood pressure measurements.
Results:
In total 105 patients (median 64 years [IQR 51, 77], 62% male) were included in our analysis. Median monitoring time was 124 minutes [IQR 92, 154] and 269 hypotension events occurred in total. A higher incidence of hypotension events was seen during the first 20 minutes. 80% of patients faced hypotension events with a mean duration of 10 minutes ±13 min per event. Severe hypotension was found only in 12% of patients with a mean time of 1±3,7 minutes. The TWA of intraoperative hypotension, the area under the threshold divided by the total duration of the procedure, was 0,51 mmHg. HPI detected IOH in the cases above in a variable duration before the event.
Conclusion:
Intraoperative hypotension is not rare during electrophysiological procedures. The early diagnosis of these events is feasible with non-invasive devices implementing artificial intelligence and allows early reaction to minimize hypotension time and potential complications due to IOH.