Discharged on Supplemental Oxygen from an Emergency Department in Patients with Bronchiolitis. S Halstead, G Roosevelt, S Deakyne, L Bajaj Pediatrics Vol.

Discharged on Supplemental Oxygen from an Emergency Department in Patients with

Bronchiolitis.

S Halstead, G Roosevelt, S Deakyne, L Bajaj

Pediatrics Vol 129 (3), Mar 2012. 604 - 611

presented by

Eleanor McCowen

17th Jan 2013

Plan of talk.

• Background and overview of study

• Following the CASP format:– Screening questions– Are the results of the study valid?– What are the results?– Will the results help me locally?

• Other relevent papers

CASP = Critical Appraisal Skills Programme

Background

• Bronchiolitis most common reason for US hospital admission in < 12mo

• Admission rates for Bronchiolitis increasing

(>150 000 p.a. in USA) though outcomes unchanged.• Considerable variability in management policies• Small changes in O2 sats cut-off values large

differences in admission rates• O2 therapy often persists after other parameters

normalised.• Several previous studies: What O2 cut-off is safe for

discharge?

Background cont’d

• Denver Colorado tertiary paediatric ED• In 2005 they introduced a home O2 protocol

- for uncomplicated bronchiolitis with hypoxia• This is their evaluation of whether the protocol reduced

admission rates safely

Denver Colorado (1600m elevation)

Methods• Denver PED(Paediatric Emergency Dept) + 4 satellite

PEDs• Retrospective notes review • Four years: 2005 - 2009• Included: All children 1- 18 mo diagnosed with

Bronchiolitis between Nov + April• Excluded: if pre-existing oxygen therapy• 4,194 illnesses studied • Also costed ave hospital stay, and Home O2 alternative

Methods (cont’d)

They recorded basic and demographic data, and which patients had:

• Initial Admission/ transfer• Discharge on Room Air (RA)

– And how many readmitted

• Discharge on Home Oxygen– And how many readmitted

Of home O2 readmissions, recorded:• Reason for readmission• Adverse airway outcomes (ICU, NIV, intubation)

Home O2Protocol

Patient must be:• 3 – 18 mo age (min 48/40 corrected age)• No chronic cardiopulmonary disease• Observed for >8 hrs with 2 hrly obs + continuous SaO2.• Have SaO2 ≥90% sleeping, feeding and awake• Be in ≤ 0.5 L/min oxygen• Feeding adequately• No evidence of respiratory deterioration

Home O2Protocol (cont’d)

• Categorical variables: used x2 or Fischer exact tests• Continuous variables: used Student t or Wilcoxon rank sum

tests

PCP = Primary care provider, i.e. Office Paediatrician or GP

Statistics

Then if doctor and parent comfortable with Home O2 discharge:

Discharged on home O2 after parent training24hr F/U appointment arranged with PCP* or in PED

Results

Results

649 (15%) initially discharged on O2

4194 illnesses

1162 (28%) initially admitted to hospital

2383 (57%) initially discharged on RA

RA = room air

90 (4%) admitted subsequent visit*

(95%CI: 3.1 – 4.6%)

38 (6%) admitted subsequent visit*

(95%CI: 4.3 0 7.9%)

*P = 0.03

Overall D/C on RA(Not Admitted)

D/C on RAThen Admitted

D/C on O2(Not Admitted)

D/C on O2Then Admitted

Admit

Illnesses, n (%)

4194 2293 (55) 90 (2) 611 (15) 38 (0.9) 11562 (28)

Mean age, mo (SD)

7.6 (4.8) 7.8 (4.4) a 6.0 (5.0) a 8.9 (4.4) b 7.2 (3.8) b 6.3 (5.1)

Male patient (%)

2497 (59) 1371 (60) 63 (70) 364 (60) 28 (74) 671 (58)

Medicaid (%)

2348 (56) 1340 (58) 54 (60) 319 (52) 22 (58) 613 (55)

D/C, discharged from hospital.a P < .01: D/C home on RA (not admitted) versus D/C home on RA then admitted.b P = .03: D/C home on O2 (not admitted) versus D/C home on O2 then admitted.

TABLE 1 Demographic Characteristics of Study Population

Table 2 Page 606

• For those readmitted after discharge in air (vs those who were not readmitted), initial temp, RR, HR and use of epinephrine were higher.

• For those sent home on oxygen then readmitted (vs not readmitted), no difference in clinical characteristics, including mean lowest SaO2 in room air.

TABLE 3 Reasons for Subsequent Admissionfor Patients Discharged From Hospital on OxygenReason a n = 39

Increased O2 19 (59%)

Increased work of breathing 17 (44%)

Parental concern/compliance 10 (62%)

Intravenous fluids for poor oral intake 4 (18%)

Problem with home O2 2 (10%)

a Patients may have more than 1 reason for subsequent admission.

Other results• None of those readmitted on home oxygen required

advanced airway management.• Thirty-five patients under 3 month who were discharged

on O2 included (3 readmitted).• Two readmissions for unrelated reasons (not

bronchiolitis) not included in readmission data.• Mean length for stay for home O2 patients 10 hours.

Conclusions

• “First study demonstrating that home oxygen protocol for selected patients [..]can be successful and sustainable.”

• Admission rates 40% historically, to 31% now.• More patients (6% vs 4%) were readmitted in the home O2

discharge group than the D/C in room air group.• Discuss costs (about half for home O2 vs admit) though

continuing care beyond first F/U not included

Conclusions (cont’d)

• Apnoea: important concern. Prev study (691 pts, 19 had apnoea), those who had apnoeas were: < 1 mo, ex-preterm and <48/40 CGA, or had an apnoea witnessed at home. Home O2 protocol excludes all the above.

• No readmitted patients “had apnoea requiring advanced airway management” (any had apnoea though?)

• Readmission to non-network site would not be recorded (though they think they would know about any deaths!)

• Altitude – increased familiarity with home O2

- what would their patients’ SaO2 be at sea level?

CASP Screening questions

1. Did the study address a clearly focused issue?Yes : Is home oxygen a safe way to lower admission rates

for Bronchiolitis in their hospital? They asked:• Qu 1 Were admission rates lowered? • Qu 2 Were adverse outcomes increased?

2. Did the authors use an appropriate method to answer their question?

Yes, Retrospective Comparative study acceptable.

Randomised Prospective study would have allowed them to collect more information, eg more detail on undesirable outcomes, comparison of length of stay.

Are the results of the study valid?

4. Was the exposure accurately measured to minimize bias? Yes

• ‘Exposure’ group clearly defined. • Recruitment to exposure group followed clear protocol.• The exceptions detailed (eg 1- 3 mo babies) didn’t follow

the protocol but this should not skew their data, as 1 – 3 mo babies included in the study.

5. Was the outcome accurately measured to minimize bias?

• All objective measures eg admission rates, ICU, NIV, etc.

So yes, but…• Other potential adverse effects not examined

– eg. feeding difficulties, infective complications, carer anxiety, availability and quality of community support.

6. Have the authors identified all important confounding factors?

• Readmissions to non-network facilities: Cannot tell how significant this effect is. Possible that those unhappy with initial Rx more likely to go elsewhere.

• Would want more information on who provides community care and the work load created.

7. A. Was the follow up of subjects complete and long enough?

Yes – Would not expect any delayed adverse effects.

What are the results?

8. Bottom line results?

• They reduced admission rates and none of the children sent home on oxygen came back needing airway intervention

10. Do you believe the results?

• Convincingly large numbers: 649 children sent home on home O2

Will the results help me locally?

11. Can the results be applied to the local population?

• Generalisable: Included all patients 1 -18 mo with bronchiolitis (unless already on home oxygen).

• BUT.. because of their altitude, with the same sats, our patients likely to be sicker.

…and to our local setting?• Hospital: medical and nursing assessment, training for

home O2• Community facilities: community nursing (or GPs??),

follow up appt, weaning.• Oxygen supplies, • Parents: No demographic data, but s.e. status and

parental education likely to be an important factor. How important are the cultural differences between here and America, eg in acceptance of Rx?

12. Do the results of this study fit with other available evidence?

• Yes - admission rates, ave. length of stay, percentage of children eligible for home oxygen.

What’s good about this paper?• Simple• Good numbers of patients• Interesting new take on an old problem

What’s not good?• Brushed over the community follow up a bit• Could have covered other adverse outcomes, including

history of apnoea.

Any comments?

(Some other relevant papers to follow)

Other relevant papers: 1‘Outpatient management of patients with bronchiolitis

discharged home on oxygen: a survey of general paediatricians’ (Utah)

Sandweiss DR , Kadish HA , Campbell KA

Clin Pediatr (Phila)   May 2012; Vol 51 ( 5 ) P 442-6

“Pediatricians are not routinely managing home oxygen for hypoxic bronchiolitis patients. Variable weaning process, difficulties in determining oxygen stoppage, multiple follow-up visits, and prolonged home oxygen usage highlight the need to evaluate the impact of this emerging practice.”

Other relevant papers: 2‘Impact of home oxygen therapy on hospital stay for infants

with acute bronchiolitis.’

Gauthier M , Vincent M , Morneau S , Chevalier I

Eur. J. Pediatr. Dec 2012; Vol 171 (12 ) 1839-1844

University of Montreal, Canada. Looked at sending home on O2 after 24 hours.

“7.1 % of patients, a mean of 1.8 days prior to real discharge[…] The number of patient-days of hospitalization which would have been saved would be 3.0 % of total in-patient-days[…] not significantly decrease the overall burden of hospitalization for bronchiolitis.”

Other relevant papers: 3‘Home oxygen for children with acute bronchiolitis.’

Tie SW , Hall GL , Peter S et al

Arch. Dis. Child. Aug 2009; Vol 94 (8 ) 641-3

Perth, Australia.

D/C after 24 hrs. Randomised to home oxygen or inpatient care. Reduced hospital stay by 2 days.

Discussion

Thankyou