Advanced bedside monitoring for patients with acute respiratory failure during non-invasive respiratory support: A systematic review and meta-analysis

BackgroundPatients with acute respiratory failure requiring non-invasive respiratory support are at high risk of deterioration. Different advanced monitoring instruments are available that can provide objective measurements. However, there is currently no evidence synthesis on these instruments. The aim of this project is to systematically synthetise data identifying the advanced monitoring instruments used and their effectiveness. MethodsWe conducted a systematic search of MEDLINE (via Pubmed), EMBASE, Web of Science, Cochrane Library and CINAHL (PROSPERO registration: CRD42024597047). We included studies with acute respiratory failure patients requiring non-invasive respiratory support where the investigators used advanced monitoring instrument during hospital stay. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. ResultsSeventy-eight studies including 3709 patients fulfilled the selection criteria. The monitoring instruments used were respiratory muscle ultrasound in 32% (n= 25/78), oesophageal manometry in 32% (n= 25/78), electrical impedance tomography in 24% (n= 19/78), electrical activity of the diaphragm (Eadi) catheter in 18% (n= 14/78) and surface EMG of parasternal muscle in 6% (n= 5/78). Thirteen studies (17%) used a multi-modal monitoring approach. Patients failing non-invasive respiratory support showed higher oesophageal pressure ({Delta}Pes) [MD 12.60 (95% CI 4.03;21.16), p=0.004], lung ultrasound score (LUS) [MD 3.93 (95% CI 1.29,6.570), p=0.003] and parasternal intercostal thickening fraction (PIC-TF%) [MD 12.58 (95% CI 8.02,17.13), p<0.001] but lower diaphragmatic thickening fraction (DTF%) [MD - 17.20 (95% CI -20.97,-13.42); p<0.001] and lower diaphragmatic excursion (DE) [MD - 0.95 (95% CI -1.08,-0.82); p<0.001. ConclusionAdvanced monitoring instruments may detect patient failing non-invasive respiratory support. Take home messageAdvanced bedside monitoring during non-invasive respiratory support can provide unique physiological insights into respiratory muscle workload and treatment response in acute respiratory failure. Our meta-analysis shows that five measurements: i) oesophageal pressure changes ({Delta}Pes), ii) lung ultrasound score (LUS), iii) parasternal intercostal thickening fraction (PIC-TF%), iv) diaphragmatic thickening fraction (DTF%) and v) diaphragmatic excursion (DE) may discriminate patients who are responders from non-responders to non-invasive respiratory support.