2004 OPEN FORUM Abstracts
DO DIFFERENT CAUSATIVE AGENTS PRODUCE DIFFERENT BRONCHIOLITIS PATIENT OUTCOMES? A PILOT STUDY.
Bennion BS RRT & Julie Ballard BS RRT, Primary Children’s
Medical Center, Salt Lake City, Utah.
Introduction: Our facility is a 232-bed pediatric, tertiary, teaching facility. We have been monitoring outcomes of bronchiolitis patients (pt) for the past seven seasons (defined as Nov. 1-April 30). Co-morbidities (congenital heart disease, chronic lung disease, immunocompromised pts & neuromuscular diseases) and/or a history of prematurity (< 38 weeks) have been identified as conditions that can complicate the course of bronchiolitis. It has been our impression that different causative agents produce varying pt outcomes.
Methods: Data were retrospectively extracted from hospital databases. Inclusion criteria were: (1) pts < 2 years of age, (2) admitted to the hospital during the 02-03 or 03-04 seasons with any diagnosis of bronchiolitis, and (3) respiratory culture obtained for respiratory syncytial virus (RSV), adenovirus (AV), influenza A (IA), influenza B (IB) &/or parainfluenza (PF) obtained via nasopharyngeal suction (NPS). We sought to test our assumption about causative agents in routine bronchiolitis pts (no co-morbidities as described above or history of prematurity). Inhaled medication treatments (tx) included albuterol, xopenex &/or epinephrine. We reviewed causative agents in regard to txs/pt, initial oxygen need, nasopharyngeal suction (NPS) events per pt, pts in the intensive care unit (ICU), pts on mechanical ventilation (MV) or non-invasive ventilation (NIV), length of stay (LOS) & cost/case.
Results: 1524 pts met the initial criteria with 297 omitted for co-morbidities &/or prematurity. Another 77 pts were excluded for having had no respiratory viral panel done. Four pts were positive for two agents (1 RSV/PF, 1 RSV/AV & 2 RSV/IA), & 233 pts had negative (NEG) results for bronchiolitis causative agents for which we test. This left us with 1150 pts & 1154 viral panel studies. Although we test for haemophilus influenza, no pts were positive for this organism. Statistical analysis included ANOVA with multiple comparison post-hoc analysis for continuous data & Chi square analysis for nominal data. Results are reported in Table One.
Table One: Outcomes of Bronchiolitis Patients With Different Causative Agents
|Mean #Txs/Pt||2.3||8.2||1.7||3.3||3.1||2.9||< 0.01|
|Initial O2 (LPM)||0.64||0.50||0.69||0.18||0.44||0.59||0.68|
|Mean # NPS/Pt||7.6||18.4||6.0||6.8||8.6||6.9||< 0.01|
|# (%) ICU Pts||60 (7)||1 (11)||4 (7)||0 (0)||0 (0)||9 (4)||0.24|
|# (%) Pts on MV/NIV||86 (11)||2 (22)||2 (4)||0 (0)||2 (6)||11(5)||0.03|
|LOS Median Days||3||5||2||2||2||2||0.27|
Discussion: It was our observation
that AV bronchiolitis pts are generally sicker and require an
increased consumption of resources. Post-hoc comparison showed
significance in mean number of txs & NPS events per pt between AV
& the other agents. Chi square analysis showed significance in
pts on MV/NIV. Statistical significance was not reached for initial
O2 need, number of ICU pts, LOS or cost/case. We recognize
that there were probably not enough pts to study and too much scatter
in our results in the areas lacking statistical significance;
therefore, several corporate databases are being merged to increase
the number of pts to study where we will again test the hypothesis.