Download Effect of ED Management on ICU Use in Acute Pulmonary Edema

Effect of ED Management on ICU Use in Acute
Pulmonary Edema
ALFRED SACCHE-I-TI, MD,*I" EDWARD RAMOSKA, MD,*I$
MARY ELLEN MOAKES, BA,I PEG McDERMOTT, RN,*
VERN MOYER, BSI
Acute pulmonary edema (APE) is a common Emergency Department (ED)
presentation requiring admission to an intensive care unit (ICU). This
study was undertaken to examine the effect of ED management on the
need for ICU admission in patients with APE. ED records of APE patients
were abstracted for patient age, prehospital and ED pharmacological
treatment, diagnoses, airway interventions, and ICU length of stay (LOS).
Statistical analysis was through multiple regression, logistic regression,
chi-square, and ANOVA. One hundred eighty-one patients composed the
study group. Pharmacological treatment included nitroglycerin (NTG),
147 patients (81%); morphine sulfate (MS), 88 (49%); loop diuretics (LD),
133 (73%); and captopril sublingual (CSL), 47 (26%). Use of CSL and MS
were associated with opposing needs for ICU admission. MS use was
associated with increased ICU admissions (odds ratio, 3.08; P = .002),
whereas CSL use was associated with decreased ICU admissions (odds
ratio, 0.29; P = .002). Morphine sulfate use also demonstrated an increased need for endotracheal intubation (ETI) (odds ratio, 5.04; P = .001),
whereas CSL demonstrated a decreased need for ETI (odds ratio, 0.16;
P = .008). Ninety-three patients required some form of respiratory support. Forty received noninvasive pressure support ventilation (NPSV)
from a bilevel positive airway pressure system (BiPAP), and 60 received
endotracheal intubation. Some patients received more than 1 form of
respiratory support; all other patients received supplemental oxygen
only. The ICU-LOS associated with different airway interventions were
supplemental oxygen, 0.72 days; BiPAP, 1.48 days; and ETI, 3.70 days
(P < .001). Specific ED pharmacological interventions are associated
with a decreased need for ICU admission and endotracheal intubation
in acute pulmonary edema patients, whereas use of noninvasive pressure support ventilation correlates with a reduction in the ICU length of
stay for patients who do require critical care admission. (Am J Emerg
Meal 1999;17:571-674. Copyright © 1999 by W.B. Saunders Company)
Acute pulmonary edema (APE) is a common Emergency
Department (ED) presentation requiring intensive care unit
(ICU) admission. I-3 Multiple studies have focused on the
effects of individual treatments on patients with this problem
by examining the responses of isolated physiological parameters to specific interventions. 4-9 None of these reports
From *Our Lady of Lourdes Medical Center Department of Emergency Medicine; 1-Coordinated Health Services Research & MIS
Division; and :J:Methodist Hospital Department of Emergency Medicine.
Manuscript received June 19, 1999, returned July 29, 1998;
revision received August 24, 1998, accepted November 1, 1998.
Presented at the American College of Emergency Physicians
Research Forum October 19, 1997, San Francisco, CA.
Address reprint requests to Dr Sacchetti, Department of Emergency Medicine, Our Lady of Lourdes Medical Center, 1600 Haddon
Ave, Camden, NJ 08103.
Key Words:Acute pulmonary edema, iCU utilization, noninvasive
pressure support ventilation, BiPAP, captopril, morphine, endotracheal intubation.
Copyright © 1999 by W.B. Saunders Company
0735-6757/99/1706-0015510.0(3/0
addressed specifically the impact of such interventions on
patient outcome and resource utilization. This study attempts to establish a correlation between ED management
and admission to the ICU.
METHODS
The site for this study was Our Lady of Lourdes Medical Center
in Camden, New Jersey, a tertiary care university-affiliated community teaching hospital. Records of ED patients with a diagnosis of
either acute pulmonary edema or congestive heart failure with
respiratory failure presenting to the ED between December 1, 1992
and October 31, 1996 were examined. Inclusion criteria required
clinical evidence of acute pulmonary edema and severe respiratory
distress as evidenced by treating personnel's descriptions of
marked dyspnea, diaphoresis, use of accessory muscles, or notation
of impression of acute respiratory compromise with no associated
primary lung pathology and no other evident cause for the patient's
respiratory distress. Vital signs were not employed as part of the
inclusion criteria because of their lack of correlation with illness
severity. Some of the most critically ill patients had near normal
vital signs as they passed from tachypnea and hypertension to
agonal respirations and hypotension. In addition, many of the most
severely ill patients had therapeutic interventions initiated before
obtaining vital signs or placement of any monitoring equipment.
Accepted ED charts were abstracted for patient age, diagnoses,
prehospital and ED pharmacological interventions, possibility of
myocardial infarction (MI), and airway management. Hospital
records were reviewed for ICU length of stay (LOS) and discharge
diagnoses. Patients not admitted to the ICU were assigned an
ICU-LOS of 0. Airway management was divided into those
patients receiving no ventilatory support, those receiving noninvasire pressure support ventilation (NPSV), and those undergoing
endotracheal intubation (ETI). NPSV was provided through a nasal
mask with a bilevel positive airway pressure system (BiPAP)
(Respironics-SPD, Murraysville, Pa). In patients receiving more
than 1 form of therapy, airway assignments were based on the
highest level of support required.
Pharmacological agents studied included use of loop diuretics
either furosemide or bumetanide, morphine sulfate (MS), nitroglycerin (NTG) either sublingual, intravenous, or transcutaneous, and
captopril sublingual (CSL).
Statistical analysis was performed, using logistic regression,
multiple regression, ANOVA, chi-square, and Student's t-test with
the JMP statistical software package from SAS Institute (Cary,
NC).
RESULTS
A total of 2,466 patient records with a diagnosis of
congestive heart failure or pulmonary edema were screened
with 181 identified for analysis. The mean patient age was
69.7 years ( + / - 1 . 0 years), with a median of 73 years.
Forty-seven (26%) patients were dialysis-dependent chronic
571
572
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 17, Number 6 • October 1999
TABLE1. Summaryof Prehospital and Emergency Department
Pharmacological Therapy
TABLE3. Summary of Prehospital and Emergency Department
Airway Management
Therapy
Number
Percent
Type of Airway
Number
Percent
Nitroglycerin
NTG sublingual
NTG intravenous
NTG topical
NTG total*
Captopril sublingual
Morphine sulfate IV (prehospital and ED)
Morphine sulfate (ED only)
Loop diuretics
82
27
109
147
47
88
61
133
45.30
14.90
60.20
81.20
25.90
48.60
33.70
73.00
Supplimental oxygen only
Respiratory failure
Noninvassive pressure support ventilation
Failed and intubated
Successful
Weaned off NPSV in ED
Endotracheal intubation
ETI prehospital
ETI in ED
Extubated in ED
88
93
40
6
34
20
60
27
33
3
49
51
22
3
19
11
33
15
18
2
*Represents total number of patients given nitroglycerin (NTG) by
any route. Patients receiving NTG by multiple routes counted only
once.
renal failure patients, and 28 (15%) patients had an additional ED diagnosis of rule-out MI. Four patients had a
discharge diagnosis including myocardial infarction, all of
which were subendocardial.
ED pharmacological management is summarized in Table
1. Table 2 lists the regression analysis for the need for
intensive care admission, ETI, and ICU LOS for each agent.
One hundred two patients (56%) were admitted to the ICU,
58 (32%) patients were admitted to a floor or stepdown unit,
and 21 (12%) patients were discharged. Use of CSL and MS
were associated with opposing needs for ICU admission.
MS use was associated with increased ICU admissions (odds
ratio, 3.09; P = .002), whereas CSL use was associated with
decreased ICU admissions (odds ratio, 0.29; P = .002).
Morphine sulfate also increased the need for endotracheal
intubation (ETI) (odds ratio, 5.04; P = .001) while CSL
decreased the need for ETI (odds ratio, 0.16; P = .008).
Patients receiving both MS and CSL demonstrated characteristics similar to the CSL group, with only 14% requiring ETI
and 42% requiring ICU admission (P < .001). Mean ICU
length of stay for all study patients was 1.84 days ( + / - 0.21
days) with a median of 1.0 days. Patients receiving sublingual captopril required a significantly shorter ICU-LOS,
0.96 days, compared with those not receiving CSL, 2.16
days (P = .026). No other pharmacological intervention
significantly affected ICU-LOS.
The uses of different airway interventions are summarized
in Table 3. Ninety-three (51%) patients required some form
of respiratory support. Forty (22%) received NPSV from a
BiPAP, and 60 (33%) underwent ETI, 27 (45%) prehospital
and 33 (55%) in the ED. Some patients received more than 1
ABBREVIATIONS:NPSV, noninvasive pressure support ventilation; ETI,
endotracheal intubation; ED, emergency department.
form of respiratory support. The ICU-LOS for airway
interventions were supplemental oxygen, 0.72 days; BiPAP,
1.48 days; and ETI, 3.70 days (P < .001). There was also no
statistical difference in ICU-LOS for those patients intubated in the field and those intubated in the ED. ICU
admission was required for 27% of the supplemental oxygen
group, 61% of the NPSV group, and 97% of the ETI patients
(P < .001).
Six (15%) of the BiPAP patients failed NPSV and
required ETI. Twenty (50%) of the BiPAP patients were
successfully weaned from the BiPAP system in the ED.
Three (5%) of the intubated patients were extubated in the
ED. Four (10%) of the BiPAP patients were discharged from
the ED after weaning. There were no significant differences
in hospital discharge diagnoses of MI in the BiPAP group
compared with the other 2 groups.
A total of 47 patients were hemodialysis-dependent
chronic renal failure (CRF) patients. Sixteen (51%) CRF
patients were discharged without admission after interventional dialysis, compared with only 5 (4%) of the non-CRF
patients (P < .001). ICU admission was required for 38% of
CRF patients compared with 63% of non-CRF (P < .004).
Intubation rates were not significantly different between the
2 groups, with 26% of CRF patients and 36% non-CRF
patients requiring ETI (P = .2).
DISCUSSION
Intensive care unit utilization by ED patients has been
reviewed for a number of cardiac-related diseases, but not
for acute pulmonary edema. 1°-13 In addition, no studies on
TABLE2. Regression Analysis for Multiple Pharmacological Treatments, Prehospital and Emergency Department
Variable
ICU Admit
Odds Ratio
P
95% CI
ETI Odds
Ratio
P
95% CI
MI
Age
Captopril
NTG
MS
Loop diuretic
2.7
0.14
0.29
1.78
3.09
1.06
,064
,048
.002
.23
.002
,89
0.99-8.35
0.02-0.94
0.13-0.63
0.69-4.67
1.54-6.30
0.46-2.39
1.2
0.06
0.16
0.72
5.04
0.9
.62
.03
.008
.54
<.001
.81
0.49-3.27
0.01-0.72
0.05-0.44
0.24-2.08
2.31-11.76
0.36-2.27
NOTE:Age was stratified as a continuous variable; NTG use was included as positive if administered to the patient in any form.
ABBREVIATIONS:LOS, length of stay (days); NTG, nitroglycerin; MS, morphine sulfate.
ICU
LOS
2.71
0.96
1,90
2,17
2.09
P
.135
.281
.026
.95
.46
.266
SACCHETTI ET AL • ICU USE IN PULMONARY EDEMA
the efficacy of individual treatments for APE have examined
their effect on ICU admissions? -8,14,15This study shows an
association between specific ED treatments and ICU admission in this population.
Use of CSL was associated with a significant decrease in
both ETI and ICU admissions. CSL's efficacy was probably
due to captopril's multiple physiological actions. Captopril
decreases peripheral vascular resistance, reduces left ventricular (LV) afterload, and improves LV diastolic dysfunction by
dampening the adrenergic crisis associated with severe
dyspnea and CHF. 15-22 Prior studies looking specifically at
captopril also have shown a reduction in ETI in patients
receiving the drug, an improvement in myocardial oxygen
consumption, and improved outcome in acute MI patients.16As,19
Morphine sulfate's use in acute pulmonary edema is
difficult to justify based on the data in this and other
studies. 23 Its use resulted in higher intubation rates, probably
through respiratory depression and consequently higher ICU
admission rates.
NTG use itself did not show a significant impact on
patient outcome. Because some form of this drug was used
in such a high percentage of patients (81%), it may have
been difficult to detect a significant difference in the data
analysis. Subgroup analysis did show that patients receiving
NTG in more than 1 form, and thus a greater amount of
NTG, were less likely to require ETI (P < .03). This is
consistent with other reports showing the advantages of
NTG in APE patients. 24
The use of loop diuretics also did not show any correlation
with ICU admission rates or need for intubation. This
absence of any effect from these agents probably reflects the
fact that blood flow to the kidneys is reduced during the
hyperadrenergic state of pulmonary edema. 25,26It is not until
the respiratory crisis is resolved that blood flow to the
kidneys returns and the loop diuretic's beneficial actions
begin. This might explain why the CRF patients in whom
diuresis was delayed for hours had as good or better
outcomes than those patients with intact renal functions who
were pharmacologically diuresed. Furosemide is known to
increase both pulmonary and peripheral vascular resistance,
although these effects are probably clinically not significant. 26The lack of immediate efficacy of the loop diuretics is
also supported by other studies on the acute care of APE
patients .24,26
Airway management in pulmonary edema patients directly impacted on ICU use. As expected, patients undergoing ETI required the most ICU resources, whereas those
receiving no ventilatory support used the least resources.
Patients with respiratory failure managed with noninvasive
pressure support ventilation required significantly fewer
ICU admissions than intubated patients (P = .001) but more
than those patients requiring no respiratory support
(P = .001).
The most obvious explanation for the increased ICU use
for intubated patient is that these patients were the sickest.
However, a comparison with those patients intubated by
paramedics does not necessarily support this premise.
Because paramedics do not have access to the BiPAP
system, they must intubate all respiratory failure patients. A
subset of these patients should comprise the supposed less
573
sick respiratory failure patients who would have received
NPSV in the ED. These patients should then have an
ICU-LOS similar to that of the ED NPSV patients, not ETI
patients. The combination of these 2 patient groups should
produce a shorter ICU-LOS for prehospital ETI patients than
ED-ETI patients. This difference is not present in this study
population. The ICU-LOS for ED intubations was 3.6 days,
compared with 3.7 days for prehospital intubations (P = NS).
Previous studies have shown NPSV to be very effective in
the management of CHF patients.27-34
Six (15%) patients failed NPSV. This failure rate is higher
than that of previous reports for NPSV in pulmonary edema
patients. 3°,34 This drop in success rate may indicate our
department's trend toward attempting NPSV on patients
with more severe respiratory failure than in our earlier
report.
This study indicates that there may be important correlations between ED management and ICU use in APE patients,
but it cannot demonstrate a causal effect. The retrospective
nature of the study and the inability to objectively compare
the degree of respiratory distress between patients prevents
us from proving that MS leads to increased intubation rates
and not that patients in need of intubation receive MS. In
designing our selection criteria, we attempted to isolate only
those patients with severe symptoms and tried to identify a
population that was equally ill and at risk for respiratory
failure.
We believe even with these limitations that we have
identified some trends in the treatment of APE patients that
warrant exploration in prospective studies.
CONCLUSION
There exists an association between specific ED treatments and ICU admission rates. Interventions with the
greatest positive correlation include CSL and use of NPSV,
whereas that with the greatest negative correlation is use
of MS.
REFERENCES
1. Gazes PC: Treatment of acute pulmonary edema. Heart Dis
Stroke 1994;3:205-209
2. Fromm RE, Varon J, Gibbs LR: Congestive heart failure and
pulmonary edema for the emergency physician. J Emerg Med
1995;13:71-87
3. Burkart F: Heart failure syndrome: Rationale for current drug
treatment. Eur Heart J 1995;16:2-3 (suppl F)
4. Kraus PE: Acute preload effects of furosemide. Chest 1990;98:
124
5. Francis GS, Siegel RM, Goldsmith, et al: Acute vasoconstrictor
response to intravenous furosemide in patients with chronic congestive heart failure. Ann Intern Med 1985;103:1-6
6. Dupuis J: Nitrates in congestive heart failure. Cardiovasc Drug
Ther 1994;8:501-507
7. Melandri G, Semprini F, Brazi A, et al: Comparative haemodynamic effects of transdermal vs. intravenous nitroglycerin in acute
myocardial infarction with elevated pulmonary artery pressure. Eur
Heart J 1990; 11:649-655
8. Nashed AH, Allegra JR: Intravenous nitroglycerin boluses in
treating patients with cardiogenic pulmonary edema. Am J Emerg
Med 1995;13:612-613
9. Barnett J, Zink KM, Touchon RC: Sublingual captopril in the
treatment of acute heart failure. Curr Ther Res 1991;49:274-281
10. Tatum JL, Jesse RL, Kontos MC, et al: Comprehensive
strategy for the evaluation and triage of the chest pain patient. Ann
Emerg Med 1997;29:116-125
574
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 17, Number 6 • October 1999
11. Hedges JR, Gibler WB, Young GP, et al: Multicenter study of
creatine kinase-MB use: Effect on chest pain clinical decision
making. Acad Emerg Med 1996;3:7-15
12. Hoekstra JW, Hedges JR, Gibler WB, et al: Emergency
department CK-MB: A predictor of ischemic complications. National
cooperative CK-MB project group. Acad Emerg Med 1994;1:17-27
13. Mulcahy B, Coates WC, Henneman PL, et al: New-onset atrial
fibrillation: When is admission medically justified? Acad Emerg Med
1996;3:114-119
14. Hamilton JR, Carter WA, Gallagher J: Sublingual captopril in
acute pulmonary edema. Acad Emerg Med 1996;3:205-212
15. Haude M, Steffen W, Erbel R, et al: Sublingual administration
of captopril versus nitroglycerin in patients with severe congestive
heart failure. Int J Cardio11990;27:351-359
16. Sacchetti AD, McCabe J, Torres M, et al: ED management of
acute congestive heart failure in renal dialysis patients. Am J Emerg
Med 1993;11:644-647
17. Langes K, Siebels J, Kuck KH: Efficacy and safety of intravenous captopril in congestive heart failure. CurrTher Res 1993;53:167176
18. Leeman M, Degaute JP: Invasive hemodynamic evaluation of
sublingual captopril and nifedipine in patients with arterial hypertension after abdominal aortic surgery. Crit Care Med 1995;23:843-847
19. Sogaard P, Nogaard A, Gotzsche CO, et al: Therapeutic
effects of captopril on ischemia and dysfunction of the left ventricle
after q-wave and non-Q-wave myocardial infarction. Am Heart J
1994;127:1-7 (suppl 1)
20. Anning PB, Grocott-Mason RM, Lewis MJ, et al: Enhancement
of left ventricular relaxation in the isolated heart by an angiotensin-converting enzyme inhibitor. Circulation 1995;92:2660-2665
(suppl 9)
21. Kingma JH, van Gilst WH, Peels CH, et al: Acute intervention
with captopril during thrombolysis in patients with first anterior
myocardial infarction. Eur Heart J 1994;15:898-907
22. Pipilis A, Flather M, Collins R, et al: Hemodynamic effects of
captopril and isosorbide mononitrate started early in acute myocardial infarction: A randomized placebo-controlled study. J Am Coil
Cardiol 1993;22:73-79
23. Hoffman JR, Reynolds S: Comparison of nitroglycerin, morphine and furosemide in treatment of presumed pre-hespital pulmonary edema. Chest 1987;92:586-593
24. Cotter G, Metzkor E, Kaluski E, et al: Randomized trial of
high-dose isosorbide dinitrate plus low-dose furosemide versus
high-dose furosemide plus low dose disosorbide dinitrate in severe
pulmonary edema. Lancet 1998;351:389-393
25. Fett DL, Cavero PG, Burnett JC Jr: Low-dose atrial natriuretic
factor and furosemide in experimental acute congestive heart failure.
J Am Soc Nephrol 1993;4:162-167
26. Sacchetti AD, Harris RH: Acute cardiogenic pulmonary edema.
Postgrad Med 1998;103:145-166
27. Meduri GU, Conoscenti CC, Menashe P, et al: Noninvasive
face mask ventilation in patients with acute respiratory failure. Chest
1989;95:865-870
28. Pennock BE, Crawshaw L, Kaplan PD: Noninvasive nasal
mask ventilation for acute respiratory failure, Chest 1994;105:441444
29. Baratz DM, Westbrook PR, Shah FK, et al: Effect of nasal
continuous positive airway pressure on cardiac output and oxygen
delivery in patients with congestive heart failure. Chest 1992;102:13971401
30. Sacchetti AD, Harris RH, Paston C, et al: Bilevel positive
airway pressure system use in acute congestive heart failure:
Preliminary case series. Acad Emerg Med 1995;2:714-718
31. Bersten AD, Holt AW, Vedig AE, et al: Treatment of severe
cardiogenic pulmonary edema with continuous positive airway pressure delivered by face mask. N Engl J Med 1991;325:1825-1830
32. Bradley TD, Halloway RM, McLaughlin PR, et al: Cardiac
output response to continuous positive airway pressure in congestive
heart failure. Am Rev Respir Dis 1992;145:377-382
33. Baratz DM, Westbrook PR, Shah PK, et al: Effect of nasal
continuous positive airway pressure on cardiac output and oxygen
delivery in patients with congestive heart failure. Chest 1992;102:13971401
34. Pollack CV, Torres MT, Alexander L: Feasibility study of the
use of bilevel positive airway pressure for respiratory support in the
emergency department. Ann Emerg Med 1996;27:189-192