Comparative Effect of High-Frequency Nasal Cannula and Noninvasive Ventilation on the Work of Breathing and Postoperative Pulmonary Complication after Pediatric Congenital Cardiac Surgery: A Prospective Randomized Controlled Trial

Abstract

Background: Various forms of commonly used noninvasive respiratory support strategies have considerable effect on diaphragmatic contractile function which can be evaluated using sonographic diaphragm activity parameters.

Objective: To compare the magnitude of respiratory workload decreased as assessed by thickening fraction of the diaphragm and longitudinal diaphragmatic strain while using high-flow nasal cannula (HFNC) and noninvasive ventilation (NIV) modes [nasal intermittent positive pressure ventilation (NIPPV) and bilevel positive airway pressure (BiPAP)] in pediatric patients after cardiothoracic surgery.

Methodology: This prospective randomized controlled trial was performed at a tertiary care surgical intensive care unit in postcardiac surgery patients aged between 1 and 48 months, who were randomly allocated into three groups: 1) HFNC (with flows at 2 L/kg/min), 2) NIPPV via RAMS cannula in PSV mode (pressure support 8 cmH2O, PEEP 5 cmH2O), and 3) BiPAP in nCPAP mode (CPAP of 5 cmH2O). Measurements were recorded at baseline after extubation (R0) and subsequently every 12 hourly (R1, R2, R3, R4, R5) at 12, 24, 36, 48, and 60 hours respectively until therapy was discontinued.

Results: Sixty patients were included, with 20 patients each in the NIPPV group, HFNC group, and BiPAP group. Longitudinal strain at crura of diaphragm was lower in the BiPAP group as compared to HFNC group at R2-R4 [R2 (-4.27± -2.73 vs - 8.40± -6.40, P = 0.031), R3 (-5.32± -2.28 vs -8.44± -5.6, P = 0.015), and R4 (-3.8± -3.42 vs -12.4± -7.12, P = 0.040)]. PFR was higher in HFNC than NIPPV group at baseline and R1-R3[R0 (323 ± 114 vs 264 ± 80, P = 0.008), R1 (311 ± 114 vs 233 ± 66, P = 0.022), R2 (328 ± 116 vs 237 ± 4, P = 0.002), R3 (346 ± 112 vs 238 ± 54, P = 0.001)]. DTF and clinical parameters of increased work of breathing remain comparable between three groups. The rate of reintubation (within 48 hours of extubation or at ICU discharge) was 0.06% (1 in NIPPV, 1 in BiPAP, 2 in HFNC) and remain comparable between groups (P = 1.0).

Conclusion: BiPAP may provide better decrease in work of breathing compared to HFNC as reflected by lower crural diaphragmatic strain pattern. HFNC may provide better oxygenation compared to NIPPV group, as reflected by higher PFR ratio. Failure rate and safety profile are similar among different methods used.

Figure 1

(a) Schematic diagram showing probe position at right costal margin to delineate anatomical structures; (b) assessment of diaphragmatic thickening fraction using M mode

Figure 2

(a) Schematic diagram illustrates the different parts of the diaphragm, (b) diaphragmatic strain measurement using a phased-array transducer positioned just below the right costal margin around the midclavicular line at the right hemidiaphragm

Figure 3

Consort diagram depicting subject enrolment in the study

Figure 4

Box–Whisker plot showing comparison of (a) DTF; (b) respiratory rate; (c) PaCO2; (d) PF ratio at different time points. * showing significance between HFNC and BiPAP groups; ^ showing significance between HFNC and NIPPV groups. [Repeated measure ANOVA test was used for within group analysis, one-way ANOVA test for between group analysis, and multiple pairwise comparison using Bonferroni correction, P value < 0.05 considered significant. (DTF-diaphragmatic thickening fraction; PFR = PaO2/FiO2 ratio; PaCO₂- arterial carbon dioxide in mmHg; R0, R1, R2, R3, R4, and R5-readings at 0, 12, 24, 36, 48, 60 hours respectively after application of different NIV methods)]

Figure 5

Box–Whisker plot comparing: (a) STC-Longitudinal strain for crura (b) STT-strain of whole diaphragm; (c) STD-strain of dome of diaphragm; (d) STZ-Strain of zone of apposition, at different time points. * showing significance between HFNC and BiPAP groups. [Repeated measure ANOVA test was used for within group analysis, one-way ANOVA test for between group analysis, and multiple pairwise comparison using Bonferroni correction, P value < 0.05 considered significant. (R0, R1, R2, R3, R4, and R5-readings at 0, 12, 24, 36, 48, 60 hours respectively after application of different NIV methods)]