ORIGINAL CONTRIBUTION asthma, PEFR and; pulmonary emergencies, asthma

Failure of Peak Expiratory Flow Rate to Predict Hospital Admission in Acute Asthma

Our study evaluated the ability of early peak expiratory flow rates (PEFR) to be predictive of hospital admission while double blinding the measure- ments and using a standardized treatment protocol and specific admission criteria. The measurements were recorded prior to and after initial therapy. Eighty-six acute asthmatic episodes in 51 patients were analyzed. Nine (10.5%) of the episodes resulted in hospitalization; 77 (89.5%) resulted in discharge. No correlation with admission was found for the initial PEFR, a second PEFR following treatment, or the differences between these two values. Our study indicates that early PEFRs alone are not useful predictors of the need for hospital admission. Although a significant number of pa- tients in our study population presented with severe airflow obstruction, they responded sufficiently to treatment to permit discharge. [Martin TG, Elenbaas RM, Pingleton SH: Failure of peak expiratory flow rate to predict hospital admission in acute asthma. Ann Emerg Med 11:466-470, September 1982.]

INTRODUCTION The management of acute asthma is a common and frequently frustrating

clinical problem. While the drug therapy of acute asthma is usually straight- forward, the disposition of patients (ie, discharge or admission) is made diffi- cult by a lack of finite criteria on which to base a decision. Such decisions usually depend on the clinical examination, arterial blood gas data, and an assessment of patient response to therapy.

Recent reports j4 have suggested that simple pulmonary function tests are useful in making patient dispositions. If one could predict which patients would be unresponsive to traditional therapy early in treatment, they could be hospitalized and their management expedited. Measurement of ei ther forced expiratory volume in one second (FEV1) or peak expiratory flow rate (PEFR) before and after treatment have been reported 14 to be useful in iden- tifying those who eventually require hospitalization.

The purpose of our study was to evaluate the ability of early measure- ments of PEFRs to predict the necessity of admission in patients with acute asthma. We have attempted to correlate measurement of PEFR prior to any therapy and the change in PEFR following epinephrine with eventual hospi- tal admission.

Thomas G. Martin, MD* Robert M. Elenbaas, PharmD t**§ Susan H. Pingleton, MD ~§ Kansas City, Missouri

From the Department of Emergency Medicine, St. Vincent's Hospital and Medical Center;* the Departments of Emergency Health Services t and Pulmonary Medicine, ~ Truman Medical Center; and the Schools of Pharmacy** and Medicine, ~ University of Missouri, Kansas City, Missouri.

Supported by the Department of Medicine and the Department of Emergency Health Services, Truman Medical Center.

Presented at the University Association for Emergency Medicine Annual Meeting in San Antonio, Texas, April 198t.

Address for reprints: Robert M. Elenbaas, PharmD, Department of Emergency Health Services, Truman Medical Center, 2301 Holmes Street, Kansas City, Missouri 64108.

METHODS Patients of either sex between the ages of 15 and 39 years who met the

American Thoracic Society criteria for as thma s and presented to the emergency department of Truman Medical Center with an acute episode were evaluated prospectively. Patients with cardiac or other pulmonary dis- eases (eg, pneumonia, pneumothorax) were specifically excluded. A PEFR measurement was obtained prior to any therapy using a Wright Peak Flow Meter (Armstrong Industries, Inc). The PEFR was measured by an emergency physician or nurse with the patient in the sitting position. The best of three attempts was recorded and used in subsequent data analysis. By affixing a removable cover to the face of the peak flow meter, test results were blinded from the patient and treating physician (Figure 1).

11:9 September 1982 Annals of Emergency Medicine 466/21

PEFR AND ASTHMA Martin, Elenbaas & Pingleton

Fig. 1. Wright Peak Flow Meter with removable cover added to enable blinding of data.

Patients were then given 0.3 mg epi- nephrine subcutaneously and treated according to a protocol (Table 1). While the aminophyl l ine mainte- nance dose is probably excessive by current standards, 68 it was consistent with recommendations 9 available at the time the study was designed. PEFR was measured again 20 minutes following the epinephrine dose. The Wright Peak Flow Meter used is inca- pable of quantifying a flow rate less than 60 L/minute; therefore, flow rates below quantification were re- corded as 60 L/minute and this value was used in subsequent calculations. Absolute PEFR was normalized be- tween patients for sex, age, and height from data supplied with the instru- ment by its manufacturer, 1° and was expressed as percent predicted PEFR.

The change in PEFR (A PEFR) was calculated by subtracting percent pre- dicted before treatment from that ob- tained 20 minutes after epinephrine administration.

The prespecified hospital admission criteria included the following: 1) pre- treatment PaCO2 > 45 mm Hg, PaO2 < 50 mm Hg, or pH < 7.20; or PaCO2 rising more than 5 mm Hg per hour accompanied by lack of improvement in clinical manifestations of asthma; 2) dyspnea and wheezing not marked- ly diminished upon completion of the treatment protocol; 3) complications such as pneumonia, pneumothorax, or unexplained pulmonary infiltrates on chest film; or 4) the patient's third pre- sentation for asthma within a seven- day period. Patients were discharged from the emergency department at any time during treatment when their dyspnea was absent and wheezes were markedly diminished or absent.

Informed consent was obtained from each patient prior to inclusion in the study. The study protocol was approved by the Clinical Research Re- view Board of Truman Medical Center and the Institutional Review Board of the University of Missouri - - Kansas City. Statistical comparisons were ac- complished with the Mann-Whitney U test. The level of significance was set at P < .05.

RESULTS Fifty-five patients representing 99

emergency department visits were

22/467

TABLE 1. Standarized treatment protocol for acute asthma

Time

0 min

20 min

40 min

60 min (1 hr)

150 min (2.5 hr)

240 rain (4 hr)

360 rain (6 hr)

Treatment

PEFR; epinephrine 0.3 mg sub-Q.

PEFR; epinephrine 0.3 mg sub-Q; begin aminophylline if respiratory rate > 40/min, PaO2 < 50 mm Hg, or PaCO2 > 45 mm Hg.

Epinephrine 0.3 mg sub-Q.

Aminophylline 6 mg/kg over 30 min, followed by 1.0 mg/kg/hr (dosage adjusted if on outpatient theophylline); isoetharine solution 0.5 cc via oxygen-powered nebulizer.

Aminophylline 3 mg/kg over 30 min, followed by 1.3 mg/kg/hr (or regimen based on measure of serum theophylline concentration).

Isoetharine solution 0.5 cc via nebulizer.

Consult for admission; initiate IV methylpredniso- lone 40 mg q6h.

TABLE 2. PEFR data in all patients

% predicted PEFR before treatment*

% predicted PEFR 20 min after epinephrine*

A PEFR*

*Mean _+ standard deviation.

Patient Visits Yielding Admission

(N = 9)

18 -+- 5%

23 _+ 8%

5_+6%

Patient Visits Yielding Discharge

(N = 77)

27 -+ 16%

43 -+ 21%

15 -+ 13%

P

< .05

< .01

< .05

Annals of Emergency Medicine 11:9 September 1982

3> O..

B d) Q_

O ::3-

~S O_

3> O..

B

O_

c] 03 d3

~B O_

3> O_ 3 m.

E)_

O9

::3-

~S G) O_

% Predicted PEFR Before Treatment

10 20 30

' i' ' el b OtO e#

": - . : . . -L. • O O 4 B ~ . • • ~ • 4 • o l ! - ' ] "

40 50 60 70 80 90 100 110 I I I I I I I |

• i l l • ~ • O 0 • ~ 0 • O G I P • OP 0 0 •

% Predicted PEFR After Treatment

10

P •

2O

I o • •

30 40 50 60 70 80 90 100 110 I I i I I I I I I

• • • dP

I • o : ~ - : - ~ i ~ .o • . . • l, p . . : . : . - - , . . .

I ° 4P

10 20

't' • 0 0 P• ~ •

: ° - . I

PEFR After Epinephrine (% Predicted)

30 40 50 60 70 I I I I i

80 90 100 110 I I I I

~:t .o ; ° t " . " • , •

2

studied during August and September 1979. Eleven visits were excluded be- cause care' did not adhere to the treat- ment protocol (Table 1). Two addition- al visits were excluded because the patients were admitted to the hospital simply because the visit constituted the third emergency department pre- sentat ion wi th in seven days. This arbitrary criterion was enforced re- gardless of the patient 's clinical re- sponse to treatment. Therefore, 86 visits for acute asthma among 51 pa- tients form the basis of this report.

11:9 September 1982

Twenty-one patients were men, 31 were women. Mean age was 22 years.

Data describing the patient sample are given (Table 2 and Figure 2). Nine of 86 patient visits (10.5%) resulted in hospital admission; 77 (89.5%) re- sulted in discharge.

Data describing PEFR in the admis- sion and discharge groups are shown (Tables 2 and 3). Further descriptions of percent predicted PEFR prior to and after treatment are also shown (Fig- ures 2, 3, and 4). Even though signifi- cant differences exist between the

Annals of Emergency Medicine

Fig. 2. Early PEFR data. Vertical line indicates the mean for a particular group. Note the extensive overlap.

admission and discharge groups in these parameters, there is too much overlap between groups for the differ- ence to be clinically useful. For exam- ple, while all patients admitted had a percent predicted PEFR prior to ther- apy < 30% and a & PEFR < 20%, so did the majority of individuals dis- charged; however, none of the patients with an initial percent predicted PEFR

468/23

PEFR AND ASTHMA Martin, Elenbaas & Pingleton

> 30% or A PEFR > 20% was hospi- talized.

Banner 1 had suggested that one must use measurements of pulmonary functions before and after initial treat- ment to be able to predict hospital admission. We correlated both param- eters with admission and discharge, but because of overlap, no admission criteria could be chosen that would be both sensitive and specific (Figures 2 and 3).

Comparison of those individuals discharged who returned within 48 hours because of recurrent airflow ob- struction with those who did not re- turn is shown (Table 3). No difference was noted in the early PEFR param- eters. This is in agreement with Lul- la's findings. H

D I S C U S S I O N

Efficient management of the patient with acute asthma includes a decision about whether to admit or to attempt treatment in the emergency depart- m e n t w i t h the hope of avoid ing admission. It would be advantageous to identify the patient in need of hos- pi tal izat ion early in his course in order to avoid long periods of treat- ment prior to admission and less effec- tive use of emergency department re- sources and personnel. Conversely, identification of the patient who will respond satisfactorily to a relatively brief, but intensive, course of manage- ment may avoid unnecessary hospital admissions. Attempts have been made to develop a method to predict the need for hospital admission in acute asthma based on early measurement of simple pulmonary function tests.

Banner et al 3 have reported on the use of a PEFR to predict hospital admission in 67 episodes of acute asthma. They found t h a t n o patients successfully treated (ie, discharged without return within 48 hours) had both an initial percent predicted PEFR

16% arid a A PEFR after a single epinephrine injection ~< 16%. Five of seven patients hospitalized had mea- surements below both these values. They suggested that patients wi th PEFRs below both these values be ad- mitted promptly. The PEFRs were not blinded from the physician and may have biased disposition. If such bias

Fig. 3. PEFR before and after treat- ment in acute asthma.

Fig. 4. PEFR before and change (A) af- ter treatment in acute asthma.

TABLE 3. Early PEFR data in discharged patients

Patients Discharged Without Relapse

Within 2 Days (N = 63)

% predicted PEFR before treatment*

% predicted PEFR 20 min after epinephrine*

A PEFR*

*Mean - standard deviation.

28 _+ 18%

44 _+ 22%

16 _+ 13%

Patients Discharged With Relapse Within 2 Days

( N = 14)

23 +- 6%

38 -+ 16%

15 -+ 13%

P

> .05

> .05

> 05

3

,,o! 100~-

< r I'v" 60~-

O

~ 50 n.- a_wEL 40 t

O " O

~ 20 m

• Discharged • Admitted

•Q

. .

~• ~ • , , o , • ° 08° • •

• • •

f I I T 0 2r0 3r0 40 510 60 710 ~n0

% Predicted PEFR After Epinephrine

__/

100 II0

E

O

4

~lo F

nooj:

60

4o i

5o

2o i '°I I

- i 0 I

-5

• Discharged • Admitted

• e & ,~ • • • •

• e L • ~ • • •

" 6 A • ~ • • A o • c A • • o~, • • •

l l 1 I I l I I I I J.

0 5 10 i5 20 25 50 35 40 45 50 A % Predicted PEFR

24/469 Annals of Emergency Medicine 11:9 September 1982

did occur, a spuriously high correla- tion would be expected between dis- position and the PEFR.

If Banner's admission criteria were applied to our data (Figures 3 and 4), on ly four of the n ine a d m i s s i o n s would have been detected. Thirteen patients who met these criteria were discharged, and only two re turned within 48 hours, leaving 11 possible needless admissions.

In a study similar in design to Ban- ner's, Nowak et al 2 evaluated the abil- ity of FEV1 measurements to identify the patient requiring admission in 85 episodes of acute asthma. Of patients wi th an initial FEV1 ~< 0.6 L, 80% either were admit ted or had subse- quent respiratory problems following discharge. For sake of comparison, one might approximate an FEV1 of 0.6 L to be equivalent to a percent predicted PEFR of 20% in the average adult. Re- view of our data revealed that only 12 of 37 patients (32%) with a percent predicted PEFR ~< 20% required ad- mission or returned within 48 hours. This admission criterion was neither sensitive nor specific as an indicator of poor outcome in our patients.

Our results indicate that early PEFR measurements alone were not able to identify the acute asthmatic patient who would eventually require admis- sion. We could not correlate PEFR pa- rameters taken before and after initial therapy wi th pat ient ou tcome in a way which could provide a sensitive and specific predictor of the need for admission. For example, if an admis- sion criterion with 100% sensitivity were desired in order to identify all admissions (pre- and post- t reatment percent predicted PEFR < 25% and < 40%, respectively) a specificity of only 23% wou ld resul t (ie, 30 pa t ien ts eventual ly discharged in our s tudy would have been considered for ad- mission by these criteria).

Our data further suggest that predic- tion of likelihood of relapse within 48 hours cannot be done with early PEFR (Table 3). This is in agreement with the findings of Lulla et al, ll but dis- agrees with the results of Nowak 2'3 and Kelsen. 4 Thei r pos t - t r ea tmen t measurements were taken at the time

of discharge; ours were made after ini- tial treatment. The tendency for re- lapse thus may depend on the degree of pulmonary function improvement after the entire treatment period, but not after the first few minutes of treat- ment.

Our data suggest that it may be possible to identify the individual who will respond satisfactorily to an inten- sive, but relatively brief, emergency department management scheme like that used here. No patient with an ini- tial percent predicted PEFR > 30% or A PEFR > 20% was hospitalized (Fig- ure 2). In addition the practice, com- mon in some institutions, of routinely admitting any patient who fails to re- spond s a t i s f a c t o r i l y to an in i t i a l course of inhaled or parenteral sym- pathomimetic may actually result in the premature and unnecessary hos- pitalization of many individuals.

CONCLUSION Measurement of PEFR cannot be

used as an early screen or predictor of the need for hospital admiss ion in adult patients with acute asthma. It appears that a substantial number of individuals present ing wi th severe airflow obstruct ion (ie, percent pre- dicted PEFR < 20% to 30%) will re- spond satisfactorily enough to allow discharge and home management. Un- like other investigators, we cannot recommend a level of initial peak flow or change in pulmonary function after early therapy below which hospital- ization should be considered routine. H o w e v e r PEFR m e a s u r e m e n t m a y identify the individual who does not need admission. These patients are those with pretreatment percent pre- dicted PEFR > 30% or A PEFR > 20% twenty minutes after a single subcuta- neous injection of 0.3 mg epinephrine.

Direct measurement of pulmonary function may also be useful because it allows quantification of the degree of airflow obstruction and objective dem- onstration of a patient's improvement or deterioration in response to therapy. This facilitates transfer of the patient f rom one physician to another with min imal discont inui ty of care. Pul-

m o n a r y func t ion test ing done just prior to discharge may detect those pa- tients likely to relapse because of m- adequate treatment, 4 although Lulla's findings 11 question this. Our recom- mendations are based on data drawn from a sample of young people who met strict criteria for the definition of asthma and did not have pneumonia or o ther compl ica t ions tha t migh t adversely influence pulmonary gas ex- change. Our results should not be ex- trapolated to older individuals with chronic obstructive lung disease or to pat ients wi th pneumonia , pneumo- thorax, or other acute or chronic pul- monary disease.

REFERENCES 1. Banner AS, Shah RS, Addington WW: Rapid prediction of need for hospitalization in acute asthma, lAMA 235:1337-1338, 1976.

2. Nowak RM: Spirometric evaluation of acute bronchial asthma. JACEP 8:9-12, 1979.

3. Nowak RM, Pensler MI, Sarkar DD, et al: Comparison of peak expiratory flow and FEV1 admission criteria for acute bronchial asthma. Ann Emerg Med 11:64-69, 1982.

4. Kelsen SG, Kelsen DP, Fleegler BF, et al: Emergency room assessment and treat- ment of patients with acute asthma - - Adequacy of conventional approach. Am J Med 60:52-59, 1976.

5. American Thoracic Society: Chronic bronchitis, asthma, and pulmonary emphy- sema. Am Rev Respir Dis 85:762-768, 1962.

6. Monitoring serum theophylline levels. Clin Pharmacokinetics 3:294-297, 1978.

7. I.V. dosage guidelines for theophylline products. FDA Bulletin.

8. Kordash TR: Theophylline levels in asthmatic patients given continuous in- travenous aminophylline. J Allergy Clin Immunol 57:248, 1976.

9. Piafsky KM, Olgilvie RI: Dosage of theophylline in bronchial asthma. N Engl J Med 292:1218-1222, 1975.

10: Gregg I, Nunn AJ: Peak expiratory flow in normal subjects. Br Med I 3:282-284, 1973. 1 l. Lulla S, Newcomb W: Emergency management of asthma in children. J Pediatr 97:346-350, 1980.

11:9 September 1982 Annals of Emergency Medicine 470/25