Neurally adjusted ventilatory assist versus pressure support ventilation: a randomized controlled feasibility trial performed in patients at risk of prolonged mechanical ventilation

Table 4 Sedation outcomes

Variable	NAVA	PSV	Effect estimates NAVA-PSV MD (95% CI)	P value
Propofol dose, g (median, IQR), n^a	1.7 (1.1 to 2.6), 32	1.8 (1.0 to 2.8), 33	0.0 (− 0.6 to 0.5)	0.933
Midazolam dose, mg (median, IQR), n^a	50.3 (29.3 to 84.9), 14	54.0 (26.0 to 95.3), 15	− 3.3 (− 41.5 to 25.5)	0.706
Fentanyl equivalents dose, mg (median, IQR), n^a,b	2.1 (1.3 to 3.4), 39	2.2 (1.6 to 3.5), 36	− 0.1 (− 0.7 to 0.5)	0.706
Bolus doses, n (median, IQR)^a	2.4 (0.7 to 4.2)	2.7 (1.6 to 4.3)	0.0 (− 1.3 to 1.0)	0.902
RASS while invasively ventilated (median, IQR)	− 2.1 (− 3 to − 0.9)	− 2.4 (− 3.6 to − 1.5)	0.4 (− 0.2 to 1.0)	0.139
RASS while in assigned mode (median, IQR), n^c	− 0.5 (− 1.6 to − 0.1), 33	− 1.4 (− 2.5 to − 0.8), 35	0.8 (0.1 to 1.4)	0.020
RASS in CMV modes (median, IQR), n^d	− 3.0 (− 3.7 to − 2.5), 35	− 3.5 (− 4.0 to − 2.8), 37	0.4 (0.0 to 0.9)	0.059

Doses are mean total administered dose per infusion day from randomization to study end. Bolus doses are mean number administered per infusion day. RASS assessed and recorded hourly by clinical staff and mean scores were calculated for each participant; group medians of individual participant mean scores were analysed
IQR interquartile range, RASS Richmond Agitation and Sedation Scale, NMB neuromuscular blockers, MD median difference, calculated using the Hodges Lehmann estimating method [15]
^aExcludes participant 20. Source data missing
^bFentanyl equivalent conversion factors: 1 mg fentanyl = 0.5 mg, remifentanyl = 100 mg, morphine = 50 mg diamorphine
^cSix NAVA arm and 2 PSV arm participants did not receive the assigned mode. RASS scores were missing in one PSV arm participant
^dTwo NAVA arm participants did not receive CMV

ISSN: 1364-8535