Predicting Mechanical Ventilation Duration Using the Asynchrony Index in ICU Patients

Abstract

Background: Mechanical ventilation is a supportive measure for patients with respiratory failure. Patient-ventilator interaction must be harmonious to achieve the goals of mechanical ventilation. Mismatch between patient’s demand and ventilator’s delivery results in Patient-ventilator asynchrony (PVA). Though PVA is associated with adverse outcomes, it is less recognized. PVA is quantified by Asynchrony index (AI). The largest body of literature has focused only on ineffective triggering in the calculation of AI. The effects of other pattern of asynchrony on patient’s length of mechanical ventilation is not well-known. Objectives: To predict the duration of mechanical ventilation by evaluating the asynchrony index following assessment of various asynchrony type prevalence among mechanically ventilated ICU patients. Methods: This prospective, non-interventional cohort study was conducted in the Department of Anaesthesia, Analgesia, and Intensive Care Medicine at Bangladesh Medical University, Dhaka, over a one-year period from September 1, 2018, to August 31, 2019. Following approval by the Institutional Review Board, we enrolled seventy ICU patients receiving mechanical ventilation via orotracheal tube according to predefined inclusion and exclusion criteria. Each patient underwent a one-hour observation period during the first 24 hours of admission. We visually analyzed pressure-time and flow-time waveforms displayed on bedside mechanical ventilator monitors to identify various asynchrony types. The asynchrony index (AI) was calculated as the ratio of asynchronous events to total respiratory rate, expressed as a percentage. Patients were stratified into two groups based on AI: high asynchrony (AI ≥10%) and low asynchrony (AI <10%). All participants were followed for 28 days to assess the duration of mechanical ventilation. Data analysis was performed using SPSS version 25 (IBM Corp.) for Windows. Results: Mean age of 70 study subjects were 57.66 ± 11 years. Among them 55.7% were male and 44.3% were female. Twenty-seven (38.57%) patients had AI ≥ 10% and forty-three (61.4%) had AI < 10%. Age and sex distribution were similar between two groups. About Nineteen percent of total breath were asynchronous. Identified asynchrony includes Ineffective triggering 40.55%, Flow asynchrony 37.14%, Premature termination 11.45%, Auto-triggering 5.42%, Double-triggering 4.92% and Delayed termination 0.5%. At least one type of asynchrony was present in 55% patients and two or more combined asynchrony were present in 12.85% patients. Patients with AI ≥ 10% had 10 more days on mechanical ventilation than patients with AI < 10% [19 days (IQR 10-28) vs. 9 days (IQR 3-15), p = 0.0001]. At 28 days, 28% patients with AI ≥ 10% were on mechanical ventilation in contrast to no patients belonging AI < 10% group. Conclusion: Patients with a high asynchrony index (AI ≥10%) demonstrated significantly prolonged mechanical ventilation duration compared to findings in most comparable studies. Notably, our study identified flow asynchrony as a particularly prevalent form of patient-ventilator dyssynchrony, occurring frequently alongside ineffective triggering.

Bangladesh Crit Care J September 2025; 13 (2): 70-79