Early Use of Adjunctive Therapies for Pediatric Acute Respiratory Distress Syndrome: A PARDIE Study

Abstract

Rationale: Few data exist to guide early adjunctive therapy use in pediatric acute respiratory distress syndrome (PARDS).Objectives: To describe contemporary use of adjunctive therapies for early PARDS as a framework for future investigations.Methods: This was a preplanned substudy of a prospective, international, cross-sectional observational study of children with PARDS from 100 centers over 10 study weeks.Measurements and Main Results: We investigated six adjunctive therapies for PARDS: continuous neuromuscular blockade, corticosteroids, inhaled nitric oxide (iNO), prone positioning, high-frequency oscillatory ventilation (HFOV), and extracorporeal membrane oxygenation. Almost half (45%) of children with PARDS received at least one therapy. Variability was noted in the median starting oxygenation index of each therapy; corticosteroids started at the lowest oxygenation index (13.0; interquartile range, 7.6-22.0) and HFOV at the highest (25.7; interquartile range, 16.7-37.3). Continuous neuromuscular blockade was the most common, used in 31%, followed by iNO (13%), corticosteroids (10%), prone positioning (10%), HFOV (9%), and extracorporeal membrane oxygenation (3%). Steroids, iNO, and HFOV were associated with comorbidities. Prone positioning and HFOV were more common in middle-income countries and less frequently used in North America. The use of multiple ancillary therapies increased over the first 3 days of PARDS, but there was not an easily identifiable pattern of combination or order of use.Conclusions: The contemporary description of prevalence, combinations of therapies, and oxygenation threshold for which the therapies are applied is important for design of future studies. Region of the world, income, and comorbidities influence adjunctive therapy use and are important variables to include in PARDS investigations.

Keywords: acute respiratory distress syndrome; extracorporeal membrane oxygenation; neuromuscular blocking agents; nitric oxide; prone position.

Figure 1.

Individual adjunctive therapy use in pediatric acute respiratory distress syndrome (PARDS) over time. Each panel illustrates the percentage of patients receiving a specific therapy on the initial day of PARDS through Day 3, with a separation of those who were newly started on that day. Total number of patients for each day was as follows: Initial, n = 624; Day 1, n = 613; Day 2, n = 583; and Day 3, n = 560. Those who were started on the therapy on the indicated day are shown in black. Those already receiving the therapy, defined as receiving the therapy the day prior, are shown in gray. (A) Continuous neuromuscular blockade (cNMB), (B) corticosteroid use for PARDS, (C) inhaled nitric oxide (iNO), (D) prone positioning, (E) high-frequency oscillatory ventilation (HFOV), and (F) extracorporeal membrane oxygenation (ECMO).

Figure 2.

Percentage of patients who received adjunctive therapies for pediatric acute respiratory distress syndrome, ordered by worst median oxygenation index on the day adjunctive therapy was started. Values displayed are percentage of the entire cohort. Worst oxygenation index (OI; median and interquartile range) on the day the adjunctive therapy was started is shown. If Pa_O2 was not available, an oxygen saturation index was used and converted to OI equivalent. cNMB = continuous neuromuscular blockade; ECMO = extracorporeal membrane oxygenation; HFOV = high-frequency oscillatory ventilation; iNO = inhaled nitric oxide; Prone = prone positioning.

Figure 3.

Heat map of adjunctive therapy use stratified by region of the world and country income determined by the 2016 World Bank geo-economic groups. (A) Heat map of adjunctive therapies stratified by region of the world. Values displayed are percentages of patients from each region who were treated with the indicated adjunctive therapy. Darker color indicates a higher percentage of use in that region. There were 413 patients originating from North America, 91 from Europe, 82 from Central and South America, and 38 from Australia, Asia, and the Middle East. Prone positioning and high-frequency oscillatory ventilation (HFOV) were both less commonly applied in North America, both with P < 0.001. P values were determined using chi-square with application of the Bonferroni correction. ***P < 0.001. There was not a significant difference among geographic regions in use of continuous neuromuscular blockade (cNMB), inhaled nitric oxide (iNO), corticosteroids specifically for pediatric acute respiratory distress syndrome, or extracorporeal membrane oxygenation (ECMO). (B) Heat map of adjunctive therapies stratified by country income. Values displayed are percentage of patients in high- or middle-income countries. Darker color indicates a higher percentage of use. There were 540 patients from high-income countries and 84 patients from middle-income countries (n = 84). High- and middle-income groups were compared using the chi-square test. Prone positioning (P < 0.001) and HFOV (P = 0.01) were more commonly applied in middle-income countries. **P = 0.01 and ***P < 0.001. There was not a significant difference among income groups in use of cNMB, iNO, corticosteroids specifically for pediatric acute respiratory distress syndrome, or ECMO.