ORIGINAL CONTRIBUTION chronic obstructive pulmonary disease
Treatment of D e c o m p e n s a t e d Chronic Obstructive P u l m o n a r y Disease in the Emergency D e p a r t m e n t -- Correlation 3 e t w e e n Clinical Features and Prognosis Study objective: Patients with decompensated chronic obstructive pulmonary disease (COPD) are at high risk of relapse after treatment in an emergency department. The purpose of this study was to determine ff the risk of relapse correlates with the clinical features of the disease. Patients: Three hundred fifty-two patients with documented COPD who were treated for dyspnea in the ED of the Albuquerque Veterans Administration Medical Center over a three-year period. Methods: We reviewed the clinical features and pulmonary function tests of the patients, who were considered to have COPD if the baseline prebronchodilator one-second forced expiratory volume (FEV1) was less than 80% predicted, and less than 80% of the forced vital capacity and inhaled bronchodilators failed to increase the FEV1 to levels of more than 80% predicted. Visits for pneumonia, pneumothorax, pleural effusion, or pulmonary emboli were excluded. A relapse was defined as an unscheduled revisit to the ED within 14 days of initial treatment. Data were entered into a microcomputer data base and analyzed by a commercial statistical package. Results: Of 877 visits in which the patient was treated and released from the ED, 281 (32.0%) resulted in relapse and were considered unsuccessful. Compared with successful visits, unsuccessful visits were characterized by a shorter duration of dyspnea (P = .002), a lower entry FEV1 (P = .027), a lower discharge FEVj (P = .040), a greater number of treatments with nebulized bronchodilators (P = .009), more frequent use of parenteral adrenergic drugs (P = .006), and less frequent use of oral prednisone on discharge (P = .016). Patients with one or more relapse visits during the study period (relapsers) differed from nonrelapsers in several respects. Relapsers had a greater bronchodilator response on baseline FEV1 than nonrelapsers (P = .047). Nevertheless, relapsers required more bronchodilator treatments in the ED (P < .001); were treated more frequently with parenteral adrenergic drugs (P < .001), IV glucocorticoids (P < .001), and oral prednisone (P < .001); and recovered less of their baseline FEV~ (P < .014). Conclusion: Bronchodilator response on baseline pulmonary function testing appears to identify patients with COPD who have a poor prognosis after emergency treatment. Their poor response to intensive bronchodilatar treatment suggests that loss of bronchodilator response m a y be involved in the pathogenesis of respiratory decompensation. [Murata GH, Gorby MS, Chick T~, Halperin AK: Treatment of decompensated chronic obstructive pulmonary disease in the emergency department - Correlation between clinical features and prognosis. Ann Emerg Med February 1991;20:125-129.]
Glen H Murata, MD Michael S Gorby, MD Thomas W Chick, MD Alan K Halperin, MD Albuquerque, New Mexico From the Ambulatory Care and Pulmonary Services, Veterans Administration Medical Center; and the Department of Medicine, University of New Mexico, Albuquerque. Received for publication November 16, 1989. Revision received April 18, 1990. Accepted for publication June 21, 1990. Presented at the 12th Annual Meeting of the Society of General Internal Medicine in Washington, DC, April 1989. Address for reprints: Glen H Murata, MD, Ambulatory Care Service (11AC), Veterans Administration Medical Center, 2100 Ridgecrest Drive SE, Albuquerque, New Mexico 87108.
INTRODUCTION
Chronic obstructive pulmonary disease (COPD) is an incurable, debilitating disorder punctuated by episodes in which respiratory function deteriorates to life-threatening levels. Although several forms of treatment are available for these episodes, 1-4 results are often less than satisfactory. In 1987, diagnosis-related group 88 (which includes COPD) accounted for more than 28,000 discharges from Veterans Administration medical centers in the United States. Nearly 5% of these hospitalizations resulted in
20:2February 1991
Annals of Emergency Medicine
125/31
COPD M u r a t a et al
death, whereas 7% of cases required r e a d m i s s i o n w i t h i n 14 days of disc h a r g e . 5 In a p r e v i o u s r e p o r t , w e showed that 28% of the patients seen in our emergency d e p a r t m e n t for dec o m p e n s a t e d COPD required retreatm e n t w i t h i n 14 days of their initial visit. 6 We also found that 24% of the patients a d m i t t e d to our m e d i c a l service for r e s p i r a t o r y d e c o m p e n s a t i o n had been seen in the ED for the same problem within the preceding two weeks, A l t h o u g h d e c o m p e n s a t e d COPD is a formidable p r o b l e m in a m b u l a t o r y medicine, w e are unaware of any previous reports that e x a m i n e the relat i o n s h i p b e t w e e n t h e c l i n i c a l features of the disease and the p a t i e n t ' s r e s p o n s e to o u t p a t i e n t t r e a t m e n t . The purpose of this study was to det e r m i n e if there were differences bet w e e n v i s i t s t h a t e n d e d in r e l a p s e and those that did not. This analysis was based on a retrospective review of our ED experience over a period of three years. MATERIALS AND METHODS The Albuquerque Veterans A d m i n i s t r a t i o n M e d i c a l C e n t e r is a 365bed, acute-care hospital serving veterans in N e w Mexico, w e s t Texas, and s o u t h e r n C o l o r a d o . T h e ED is staffed by faculty and house officers from the U n i v e r s i t y of N e w Mexico. A l m o s t all of the patients who present to the m e d i c a l center w i t h dec o m p e n s a t e d C O P D are referred to the ED for evaluation, therapy, and/ or admission. T r e a t m e n t in the ED is not standardized but consists of lowflow oxygen therapy, nebulized bronchodilators, IV a m i n o p h y l l i n e , parenteral adrenergic drugs, corticosteriods, and antibiotics in varying combinations. Patients who respond to t r e a t m e n t are advised to return to the ED if s y m p t o m s recur. T h e y are also g i v e n f o l l o w - u p app o i n t m e n t s w i t h a w a l k - i n clinic, the general medical clinic, or the p u l m o nary clinic. If t h e y require additional t r e a t m e n t at t h e t i m e of follow-up, they are referred back to the ED. All ED visits are recorded in a log book that includes the p a t i e n t ' s chief c o m p l a i n t , diagnosis, t r e a t m e n t s received, and disposition. This log was screened for all v i s i t s m a d e by patients w i t h the clinical diagnosis of COPD, emphysema, asthma, or bronchitis during a period of three years. Visits were included in this study if 32/126
the p a t i e n t c o m p l a i n e d of dyspnea, cough, and/or excessive s p u t u m production and if the p a t i e n t had at least one set of baseline p u l m o n a r y funct i o n t e s t s (PFTs) p e r f o r m e d w i t h i n t h r e e y e a r s of e n t r y t h a t d e m o n strated irreversible, obstructive lung disease. Visits were e x c l u d e d if t h e p a t i e n t was t h o u g h t to h a v e pneumonia, pneumothorax, pulmonary embolism, or pleural effusion. Additional cases were excluded if review of t h e m e d i c a l r e c o r d s h o w e d t h a t the patient suffered from occupational lung disease or p u l m o n a r y fibrosis. Visits by patients w i t h COPD and left heart failure were included if they were treated w i t h bronchodilators in the ED. A p a t i e n t was considered to have obstructive lung disease on baseline PFTs if at least one set showed a onesecond forced expiratory volume (FEVI) less t h a n 80% p r e d i c t e d and less t h a n 80% of the forced vital cap a c i t y (FVC). O b s t r u c t i o n was cons i d e r e d i r r e v e r s i b l e if t h e FEV 1 rem a i n e d less t h a n 80% predicted after b r o n c h o d i l a t o r s on a l l P F T s p e r f o r m e d w i t h i n three years of entry. Response to bronchodilators was expressed as the proportion of the defect in e x p i r a t o r y flow c o r r e c t e d by inhaled albuterol (response index [RI]). RI was derived from the following equation: RI - 100 x (postbronchodilator FEV1)
-
(prebronchodilator I:EVI)
(predicted FEV~) - (prebronchodilator FEV1}
W h e n m u l t i p l e PFTs were performed d u r i n g t h e eligible period, o n l y t h e highest value for each p a r a m e t e r was used in our analysis. For ED visits in w h i c h s p i r o m e t r y was performed, percent recovery (PR) was calculated. PR was the extent to w h i c h the patient recovered baseline p r e b r o n c h o d i l a t o r FEV~ at t h e t i m e of release from the ED and was calculated from the following equation: PR = 100 x (discharge ~EV1) / (baseline FEV1)
M o s t of the patients in this study had more than one ED visit in which e x p i r a t o r y v o l u m e s were m e a s u r e d . For t h i s reason, m e a n e n t r y FEV1, m e a n discharge FEV u and m e a n recovery were calculated. Mean entry FEV 1 was t h e average FEV 1 for all visits in w h i c h FEV 1 was measured before treatment, whereas m e a n discharge FEV 1 was average discharge FEV 1. Mean recovery was the ratio of mean discharge FEV 1 to the best preA n n a l s of Emergency Medicine
TABLE 1. Baseline pulmonary function tests for 269 patients with obstructive lung disease Before After Bronchodilator Bronchodilator FEV1 (L) % Predicted FVC (L) % Predicted FEV1/FVC
1.38 _+ 0.58
1.49 ~ 0.59
43.8 ± 17.4
46.8 + 17.6
2.62 _+ 0.84
2.88 + 0.84
60,0 + 18.6
65.9 _+ 18.6
53.0 -+ 15.3
52.4 ÷ 16.2
bronchodilator FEV 1 on baseline PFTs. The frequency of use for each therap e u t i c i n t e r v e n t i o n was also calculated for each patient. A l l a v a i l a b l e p a t i e n t m e d i c a l records were reviewed. Information was extracted regarding ED visits and PFTs and recorded on forms designed for c o m p u t e r entry. D a t a were entered into a customized, microcomputer data base and analyzed by a commercial statistical package (Systat®). A previous study from our institution s h o w e d t h a t p a t i e n t s w i t h obstructive lung disease were at excessive risk for recurrent symptoms for at least 14 days. 6 For this reason, return v i s i t s o c c u r r i n g w i t h i n t h i s period were considered relapses. For historical, diagnostic, and t r e a t m e n t variables in the ED, visits ending in relapse were compared with visits that did not. For demographic characteristics, PFT variables, and frequency of therapeutic interventions, patients w i t h one or m o r e relapse visits (relapsers) were c o m p a r e d w i t h those who did not relapse at any time during the study period (nonrelapsers). D i f f e r e n c e s in t h e p r o p o r t i o n o~ n o m i n a l v a r i a b l e s b e t w e e n groups were tested by x 2 analysis. The distributions of continuous variables were e x a m i n e d by f r e q u e n c y histograms. Group differences for normally distributed variables were analyzed by the unpaired Student's t test, whereas all others were analyzed by the MannW h i t n e y U test. M u l t i p l e linear regression Was used to define the relationship b e t w e e n the n u m b e r of relapse visits for each patient and his demographic characteristics, PFTs, and c o m p o s i t e d i a g n o s t i c and treatment factors. D i a g n o s t i c factors included m e a n e n t r y FEVI, m e a n discharge FEV u and mean recovery. Treatment factors included the m e a n number of nebulized bronchodilator treatments 20:2 February 1991
COPD Murata et al
TABLE 2. Comparison of successful and unsuccessful ED visits
Visits Ending in Relapse (281)
Visits Not Ending in Relapse (596)
3,40 ± 5.24
4.63 ± 6.40
Historical Features Duration of dyspnea (days)
.002
ED diagnosis of "asthma" (%)
12.1
15.3
NS
Change in sputum (%)
34.9
35.7
NS
Patienton oral prednisone at entry (%)
26.7
25.0
NS
97.9 ± 1.1 100.2 ± 16.3
97.8 ± 1.2 97.9 ± 16.5
NS NS
26.3 ± 6,4
25.9 ± 6.3
NS
Admission FEV1 (mL) (489)
682 ± 303
766 _+ 419
.027
DischargeFEV1 (mL) (279)
899 ± 415
1,033 ± 554
,040
Arterial blood gases on room air (545) pH Po2 (ram Hg) Pco2 (mm Hg) 02 saturation (%)
7.39 59,1 37.9 87.7
7.39 59.8 38.4 87.4
NS NS NS NS
Broodcounts (378) Hematocrit White count ( x 1,000)
45.7 _+ 4.2 8.9 ± 2.8
46.6 ± 4.6 9.2 ± 3.2
NS NS
Physical Findings Temperature(F) Pulse Respirations
Laboratory Tests
+_ 0.06 ± 12.0 ± 7.5 ± 7.6
± -+ ± ±
0.06 12.3 9.2 8.1
Treatment No, of nebulizer treatments
1.22 ± 0.76
1.08 ± 0.82
.009
Parenteral sympathomimetics used in ED (%)
39.9
30,5
.006
IV corticosteroids given in ED (%)
17.4
19.6
NS
IV aminophylline used in ED (%)
19.9
15.3
NS
Oral prednisone on discharge (%)
18.1
25.5
.016
Antibiotics on discharge (%)
29.5
28.4
NS
8.5
6.4
NS
68.9 ± 25.2
78.8 ± 32.1
.010
New broechodilator added on discharge (%)
Response Percent recovery of baseline FEV1 (279)
~lues a~ expressed as mean ± SD. Pe~enlages ~fer to the p~portion of visits in which the characteristic was found.
and the frequency with which the patient was treated with IV aminophylline, IV corticosteroids, parenteral adrenergic drugs, antibiotics, and oral prednisone. RESULTS
During a period of three years, 352 20:2 February 1991
patients with irreversible, obstructive lung disease were treated for dyspnea, cough, and/or excessive sputum production in the ED of the Albuquerque Veterans Administration Medical Center. Eighty-three patients (23.6%) were admitted to the hospital on every visit. The remainAnnals of Emergency Medicine
ing 269 were treated and released from the ED on one or more occasions and were the study population of this report. The mean (+ SD) age of the latter group was 65.8 _+ 8.7 years, and all but three were men. Eighty-seven (32.3%) were on home oxygen therapy at entry, and 125 (46.5%) were given oral prednisone at least once during the study period. Baseline PFTs for the patient sample are shown (Table 1). The mean RI was 11.5 _+ 12.9%, and all subjects had persistent baseline defects in FEV1 after inhaled bronchodilators. The 269 subjects in this study made 1,157 visits to the ED during the three-year study period. Two hundred eighty visits (24.2%) resulted in admission to the hospital and were excluded from further analysis. Of the 877 visits in which the patient was treated and released from the ED, 281 (32.0%) ended in relapse and were considered unsuccessful. The remaining visits were classified as successful. In 25 visits (2.9%), the patients were considered to have left ventricular failure in addition to decompensated obstructive lung disease.
The clinical features of unsuccessful and successful visits are compared (Table 2). Duration of dyspnea was the only historical feature that had prognostic value. The risk of relapse for a duration less than 4.2 days (36.6%) was s i g n i f i c a n t l y greater than for a duration of 4.2 days or more (25.3%; P = .010). Few patients had all of the laboratory tests performed. However, for each of the tests shown, performance of the test had no effect on probability of relapse. Entry FEV 1 and discharge FEV 1 were significantly higher for successful than for unsuccessful visits. When visits were stratified by entry FEVu there was no significant difference in the relapse rate between visits in which entry FEV 1 exceeded the mean for all visits (740 mL) and those in which entry FEV 1 was less. Stratification by discharge FEV t also failed to identify visits at risk for subsequent relapse. There were several treatment differences between unsuccessful and successful visits. The number of nebulizer treatments for unsuccessful visits was significantly higher than for successful visits. Parenteral adrener~ic drugs were also used more often for unsuccessful visits, whereas 127/33
COPD Murata et al
oral prednisone was given less frequently at the time of discharge. Despite more intensive therapy, PR was less for unsuccessful than for successful visits. No differences were found in the use of IV corticosteroids, IV or oral theophylline, antibiotics, or other bronchodilators. The relapse rate for visits in which PR was more than 75% was similar to the rate for visits in w h i c h PR was less than 75%. Of the 269 patients in this study, 101 (37.5%) had one or more relapse visits in three years; 45.6% of the visits made by this group ended in relapse. Relapsers differed from nonrelapsers with respect to several characteristics (Table 3). On baseline PFTs, the RI was significantly higher for relapsers than for nonrelapsers. No differences were noted with respect to FEV1, FVC, or the FEVI:FVC ratio either b e f o r e or after bronchodilators. Relapse patients also presented with a shorter duration of symptoms, were given more nebulizer treatments, and were treated more frequently with parenteral adrenergic drugs, IV corticosteroids, and oral prednisone. For 189 patients (70.3%), it was possible to calculate mean entry FEV I. Relapsers had a significantly lower value of mean entry FEV 1 than nonrelapsers in this subset. For 133 patients (49.4%), it was also possible to calculate mean discharge FEVf and mean recovery. Despite more intensive treatment, relapse patients recovered less of their baseline FEV 1 at the time of discharge. Stepwise multiple linear regression was used to examine the relationship between number of relapse visits and several baseline PFT and treatment variables. RI, number of nebulizer treatments, use of parenteral adrenergic drugs, and oral prednisone on discharge were shown to be independent variables in this model. Although highly significant (P < .001), this model explained very little of the variance in relapse number (multiple R 2, 0.202). Addition of mean entry FEVu mean discharge FEVu and mean recovery did not improve the fit for the subset of patients in which these variables were available.
DISCUSSION COPD is a major cause of death, disability, and e c o n o m i c hardship among elderly patients in the United 34~28
TABLE 3. Comparison of relapsers and nonrelapsers Relapsers (101)
Nonrelapsers (168)
P
65.9 + 8,1
65.7 _ 9.1
NS
37.6
29.2
NS
4.57 ± 5.85
6.39 ± 7.27
.025
Prebronchodilator FEV1 (L) % Predicted FVC (L) % Predicted FEVl/FVC
1.40 45.3 2.69 61.9 51.1
± ± 4± ±
0.61 18.0 0.85 19.7 13.8
1.37 43.0 2.58 58.9 54.1
± ± ± ± ±
0.57 17.0 0.83 17.9 16.1
NS NS NS NS NS
Postbronchodilator FEV1 (L) % Predicted FVC (L) % Predicted FEVJFVC
1.53 48,0 2.89 66.4 51.2
± ± ± ± 4-
0.62 18.8 0.87 20.1 14.3
1.47 46.0 2.87 65.6 53.2
_+ 0.57 ± 16.9 ± 0.83 ± 17.6 ± 17,2
NS NS NS NS NS
Age (yr) Home oxygen use (%) Mean duration of dyspnea (days)
Baseline PFI's
Response index (249)
13.6 +_ 15.5
10.3 ± 10.8
.047
Mean entry FEV1 (mL) (189)
669 ± 296
861 ± 478
.00I
Mean discharge FEV~ (mL) (133)
903 ± 424
1,054 ± 541
Mean recovery (%) (133)
70.9 ± 27.9
84.1 ± 32,9
0.14
Mean no. of nebulizer treatments
1.21 ± 0.63
0.88 -+ 0.76
< .001
Parenteral adrenergic drugs (%) IV aminophyliine (%)
32.3 ± 30.9 15.7 ± 21.5
18.8 ± 34.9 14.5 ± 32.4
< .001 NS
NS
Treatment
IV corticosteroids (%)
18.3 ± 20.0
10.6 4- 26.3
< .001
Prednisone on discharge (%)
21.0 ± 21.7
13,8 ± 30.3
< .001
Antibiotics on discharge (%)
29.0 ± 28.9
32.0 4- 43.4
NS
Percenlages refer to the mean -+ SD frequency with which each inlervention was used.
States. Because it cannot be cured, preservation of the patient's functional status as an outpatient may be the most cost-effective strategy for this disease. Such a strategy would include the prevention of respiratory decompensation and effective outpatient management once decompensation occurs. Unfortunately, there is little evidence that outpatient therapy is effective for all COPD patients. At our institution, one fourth of the patients admitted for decompensated COPD were treated and released from our ED within the preceding two weeks. This observation suggested that modest improvements in the o u t p a t i e n t m a n a g e m e n t of COPD might produce significant reductions in the costs and hardship attributed to this disease. In a previous study, we found that 28% of ED visits made by patients Annals of Emergency Medicine
with the clinical diagnosis of asthrn~ or COPD ended in relapse. 6 Thil problem was confined largely to~ small group of patients who had clin ical features of both a s t h m a an~ COPD. It was unclear if relapse oc curred because of greater severity c illness at the time of presentation o: differences in the manner in whict these patients were treated. The put pose of this study was to examin~ the relationship among the clinicl features of COPD, its outpatient m a n a g e m e n t , and the short-ter~ prognosis for a smaller group of pa tients with well-documented, irre versible airway obstruction. We found that unsuccessful visits were characterized by a short dura tion of dyspnea, a higher number 01 nebulizer treatments, the use of pa enteral adrenergic drugs, and fail~ to give oral prednisone on discharge. i:
20:2 February 19~!
COPD Murata et al
For visits in which spirometry was performed, e n t r y FEV1, discharge FEV1, and PR were s i g n i f i c a n t l y smaller for unsuccessful than successful visits. However, stratification of ED visits with respect to the latter variables failed to produce differences in relapse rate among the different categories of risk. Relapsers had a higher bronchodilator response on baseline PFTs than nonrelapsers and were treated more vigorously with nebulized bronchodilators, parenteral adrenergic drugs, and IV and 0ral glucocorticoids in the ED. Our conclusions differ from a large body of literature that suggests that bronchodilator response on baseline PFTs cannot be used to predict the response to bronchodilator treatment in COPD. 7 Several reasons for the validity of these criticisms include the following: day-to-day variation in FEV1 is large; response is usually expressed as percent improvement over baseline flow, and small changes in patients with severe reduction in FEV1 might therefore be interpreted as a response; FEV~ may not be the best measure of obstruction; there is poor correlation between FEV 1 and dyspnea or exercise i n t o l e r a n c e ; bronchodilators may be more important in preventing deterioration than in improving baseline e x p i r a t o r y flow; and the benefits of bronchodilat0rs may be due to effects other than improved air flow. We designed this study to circumvent many of these problems. When several PFTs were available for a given patient, we selected the faighest value of FEV1 for analysis. Response to bronchodilators was expressed as the percentage of the defect in FEV1 corrected, not as a percentage change over baseline. The end point for this study was the patient's perception of his own clinical status, not a change in FEV1. Our experience was large enough for clinically important differences to be apparent. Nevertheless, our conclusions about the value of the baseline bronchodilator response as a p r o g n o s t i c i n d i c a t o r should be interpreted with caution and must be validated in a prospective manner. Pulmonary function criteria for the classification of ambulatory patients with obstructive pulmonary disease have not been established. We chose liberal criteria for this study because we expected patients treated in an 20:2February 1991
outpatient setting to have lesser degrees of airflow obstruction than hospitalized patients reported in other studies. The use of a more restrictive d e f i n i t i o n of airflow o b s t r u c t i o n would have resulted in the classification of many chronically symptomatic patients as "normal." Local custom is to measure pulmonary function when the patient has reached a maximal degree of recovery after an episode of respiratory decompensation; there were no time requirements for the definition of this baseline period. Such criteria would have been arbitrary and would not have accounted for the markedly unstable respiratory function in some of our patients. Several made 15 to 20 ED visits per year and would never have attained a baseline status defined by stable lung function over a period of several weeks. We did not wish to use selection criteria that would have removed the most problematic patients from the study. It is unclear why patients with a bronchodilator response had such a poor prognosis. The correlation between the relapse number and a variety of PFT and treatment variables was poor. This finding suggests that bronchodilator response is a marker for another process that is causally related to outcome. Part of the explanation is suggested by our comparisons of relapsers and nonrelapsers. Relapsers had a higher bronchodilator response at baseline but required much more intensive therapy in the ED for respiratory decompensation. Among those who had entry and discharge FEVlS recorded, relapsers recovered much less of their baseline FEV 1 in the ED despite more frequent use of parenteral adrenergic drugs, IV c o r t i c o s t e r o i d s , and a greater number of treatments with nebulized bronchodilators. It is possible that respiratory decompensation occurs in some COPD patients because of a change in airway sensitivity to bronchodilators. Studies have shown that exposure of the f~-adrenergic receptor to excessive stimulation results in tachyphylaxis 8 and that corticosteriod therapy reverses this defect.9, lo We believe that loss of airway sensitivity to bronchodilators should be studied prospectively as a mechanism for respiratory decompensation. We found that oral prednisone was the only treatment factor associated Annals of Emergency Medicine
with a good prognosis. This finding did not result from a patient selection bias because prednisone was used more frequently in patients with one or more relapses. It is also unlikely that prednisone was used for milder episodes of respiratory dec o m p e n s a t i o n . N e v e r t h e l e s s , the value of prednisone in the outpatient m a n a g e m e n t of d e c o m p e n s a t e d COPD should be confirmed by a randomized clinical trial. CONCLUSION Relapse is common for COPD patients who have a short duration of symptoms, who are treated with parenteral adrenergic drugs in the ED, or who are not given oral prednisone on discharge. Some patients who have a bronchodilator response on baseline PFTs lose their response during respiratory decompensation. This phenomenon might explain why bronchodilator response on baseline PFTs is so poorly correlated with the clinical response to bronchodilator therapy and w h y outpatient treatment programs are often unsuccessful. A prospective study should be conducted in other settings to confirm our findings.
REFERENCES 1. Nicotra MB, Rivera M, Awe RJ: Antibiotic therapy of acute exacerbations of chronic bronchitis. Ann Intern Med 1982;97:18-21.
2. Anthonisen NR, Manfreda J, Warren CPW, et aI: Antibiotic therapy in exacerbations of chronic obstructive pulmonary disease. Ann Intern Med 1987;106:196-204. 3. Rice KL, Leatherman JW, Duane PG, et al: Aminophylline for acute exacerbations of chronic obstructive pulmonary disease: A controlled trial. Ann Intern Med 1987; 107:305-309. 4. Albert RK, Martin TR, Lewis SW: Controlled clinical trial of methylprednisolone in patients with chronic bronchitis and acute respiratory insufficiency. An n I~tern Med 1980;92:753-758. 5. Patient Treatment File: National Statistics for Fiscal Year 1987. Washington, DC, Veterans Administration
Central Office. 6. Murata GH, Gorby MS, Chick TW, et al: Use of emergency medical services by patients with decompensated obstructive lung disease. A n n Emerg Me d 1989;18:501-506. 7. Pratter MR, Irwin RS: Predicting response to bronchodilator therapy in chronic obstructive pulmonary disease. Arch Intern Med 1988;148:1909-1910. 8. Stiles GL, Caron MG, Lefkowitz RJ: JS-Adrenergic receptors: Biochemical mechanisms of physiological regulation. Physiol Rev 1984;64:661-743. 9. Hui KKP, Conolly ME, Tashkin DP: Reversal of human lymphocyte [3-adrenoceptor desensitization by glucocorticoids. Clin Pharmacol Ther 1982,;32:566-571. 10. Brodde OE, Brinkmann M, Schemuth R, et al: Terbutaline-induced desensitization of human lymphocyte [32-adrenoceptors: Accelerated restoration of lS-adrenoceptor responsiveness by prednisone and ketotifen. J Clin Invest 1985;76:1096-1101.
129/35
Treatment of D e c o m p e n s a t e d Chronic Obstructive P u l m o n a r y Disease in the Emergency D e p a r t m e n t -- Correlation 3 e t w e e n Clinical Features and Prognosis Study objective: Patients with decompensated chronic obstructive pulmonary disease (COPD) are at high risk of relapse after treatment in an emergency department. The purpose of this study was to determine ff the risk of relapse correlates with the clinical features of the disease. Patients: Three hundred fifty-two patients with documented COPD who were treated for dyspnea in the ED of the Albuquerque Veterans Administration Medical Center over a three-year period. Methods: We reviewed the clinical features and pulmonary function tests of the patients, who were considered to have COPD if the baseline prebronchodilator one-second forced expiratory volume (FEV1) was less than 80% predicted, and less than 80% of the forced vital capacity and inhaled bronchodilators failed to increase the FEV1 to levels of more than 80% predicted. Visits for pneumonia, pneumothorax, pleural effusion, or pulmonary emboli were excluded. A relapse was defined as an unscheduled revisit to the ED within 14 days of initial treatment. Data were entered into a microcomputer data base and analyzed by a commercial statistical package. Results: Of 877 visits in which the patient was treated and released from the ED, 281 (32.0%) resulted in relapse and were considered unsuccessful. Compared with successful visits, unsuccessful visits were characterized by a shorter duration of dyspnea (P = .002), a lower entry FEV1 (P = .027), a lower discharge FEVj (P = .040), a greater number of treatments with nebulized bronchodilators (P = .009), more frequent use of parenteral adrenergic drugs (P = .006), and less frequent use of oral prednisone on discharge (P = .016). Patients with one or more relapse visits during the study period (relapsers) differed from nonrelapsers in several respects. Relapsers had a greater bronchodilator response on baseline FEV1 than nonrelapsers (P = .047). Nevertheless, relapsers required more bronchodilator treatments in the ED (P < .001); were treated more frequently with parenteral adrenergic drugs (P < .001), IV glucocorticoids (P < .001), and oral prednisone (P < .001); and recovered less of their baseline FEV~ (P < .014). Conclusion: Bronchodilator response on baseline pulmonary function testing appears to identify patients with COPD who have a poor prognosis after emergency treatment. Their poor response to intensive bronchodilatar treatment suggests that loss of bronchodilator response m a y be involved in the pathogenesis of respiratory decompensation. [Murata GH, Gorby MS, Chick T~, Halperin AK: Treatment of decompensated chronic obstructive pulmonary disease in the emergency department - Correlation between clinical features and prognosis. Ann Emerg Med February 1991;20:125-129.]
Glen H Murata, MD Michael S Gorby, MD Thomas W Chick, MD Alan K Halperin, MD Albuquerque, New Mexico From the Ambulatory Care and Pulmonary Services, Veterans Administration Medical Center; and the Department of Medicine, University of New Mexico, Albuquerque. Received for publication November 16, 1989. Revision received April 18, 1990. Accepted for publication June 21, 1990. Presented at the 12th Annual Meeting of the Society of General Internal Medicine in Washington, DC, April 1989. Address for reprints: Glen H Murata, MD, Ambulatory Care Service (11AC), Veterans Administration Medical Center, 2100 Ridgecrest Drive SE, Albuquerque, New Mexico 87108.
INTRODUCTION
Chronic obstructive pulmonary disease (COPD) is an incurable, debilitating disorder punctuated by episodes in which respiratory function deteriorates to life-threatening levels. Although several forms of treatment are available for these episodes, 1-4 results are often less than satisfactory. In 1987, diagnosis-related group 88 (which includes COPD) accounted for more than 28,000 discharges from Veterans Administration medical centers in the United States. Nearly 5% of these hospitalizations resulted in
20:2February 1991
Annals of Emergency Medicine
125/31
COPD M u r a t a et al
death, whereas 7% of cases required r e a d m i s s i o n w i t h i n 14 days of disc h a r g e . 5 In a p r e v i o u s r e p o r t , w e showed that 28% of the patients seen in our emergency d e p a r t m e n t for dec o m p e n s a t e d COPD required retreatm e n t w i t h i n 14 days of their initial visit. 6 We also found that 24% of the patients a d m i t t e d to our m e d i c a l service for r e s p i r a t o r y d e c o m p e n s a t i o n had been seen in the ED for the same problem within the preceding two weeks, A l t h o u g h d e c o m p e n s a t e d COPD is a formidable p r o b l e m in a m b u l a t o r y medicine, w e are unaware of any previous reports that e x a m i n e the relat i o n s h i p b e t w e e n t h e c l i n i c a l features of the disease and the p a t i e n t ' s r e s p o n s e to o u t p a t i e n t t r e a t m e n t . The purpose of this study was to det e r m i n e if there were differences bet w e e n v i s i t s t h a t e n d e d in r e l a p s e and those that did not. This analysis was based on a retrospective review of our ED experience over a period of three years. MATERIALS AND METHODS The Albuquerque Veterans A d m i n i s t r a t i o n M e d i c a l C e n t e r is a 365bed, acute-care hospital serving veterans in N e w Mexico, w e s t Texas, and s o u t h e r n C o l o r a d o . T h e ED is staffed by faculty and house officers from the U n i v e r s i t y of N e w Mexico. A l m o s t all of the patients who present to the m e d i c a l center w i t h dec o m p e n s a t e d C O P D are referred to the ED for evaluation, therapy, and/ or admission. T r e a t m e n t in the ED is not standardized but consists of lowflow oxygen therapy, nebulized bronchodilators, IV a m i n o p h y l l i n e , parenteral adrenergic drugs, corticosteriods, and antibiotics in varying combinations. Patients who respond to t r e a t m e n t are advised to return to the ED if s y m p t o m s recur. T h e y are also g i v e n f o l l o w - u p app o i n t m e n t s w i t h a w a l k - i n clinic, the general medical clinic, or the p u l m o nary clinic. If t h e y require additional t r e a t m e n t at t h e t i m e of follow-up, they are referred back to the ED. All ED visits are recorded in a log book that includes the p a t i e n t ' s chief c o m p l a i n t , diagnosis, t r e a t m e n t s received, and disposition. This log was screened for all v i s i t s m a d e by patients w i t h the clinical diagnosis of COPD, emphysema, asthma, or bronchitis during a period of three years. Visits were included in this study if 32/126
the p a t i e n t c o m p l a i n e d of dyspnea, cough, and/or excessive s p u t u m production and if the p a t i e n t had at least one set of baseline p u l m o n a r y funct i o n t e s t s (PFTs) p e r f o r m e d w i t h i n t h r e e y e a r s of e n t r y t h a t d e m o n strated irreversible, obstructive lung disease. Visits were e x c l u d e d if t h e p a t i e n t was t h o u g h t to h a v e pneumonia, pneumothorax, pulmonary embolism, or pleural effusion. Additional cases were excluded if review of t h e m e d i c a l r e c o r d s h o w e d t h a t the patient suffered from occupational lung disease or p u l m o n a r y fibrosis. Visits by patients w i t h COPD and left heart failure were included if they were treated w i t h bronchodilators in the ED. A p a t i e n t was considered to have obstructive lung disease on baseline PFTs if at least one set showed a onesecond forced expiratory volume (FEVI) less t h a n 80% p r e d i c t e d and less t h a n 80% of the forced vital cap a c i t y (FVC). O b s t r u c t i o n was cons i d e r e d i r r e v e r s i b l e if t h e FEV 1 rem a i n e d less t h a n 80% predicted after b r o n c h o d i l a t o r s on a l l P F T s p e r f o r m e d w i t h i n three years of entry. Response to bronchodilators was expressed as the proportion of the defect in e x p i r a t o r y flow c o r r e c t e d by inhaled albuterol (response index [RI]). RI was derived from the following equation: RI - 100 x (postbronchodilator FEV1)
-
(prebronchodilator I:EVI)
(predicted FEV~) - (prebronchodilator FEV1}
W h e n m u l t i p l e PFTs were performed d u r i n g t h e eligible period, o n l y t h e highest value for each p a r a m e t e r was used in our analysis. For ED visits in w h i c h s p i r o m e t r y was performed, percent recovery (PR) was calculated. PR was the extent to w h i c h the patient recovered baseline p r e b r o n c h o d i l a t o r FEV~ at t h e t i m e of release from the ED and was calculated from the following equation: PR = 100 x (discharge ~EV1) / (baseline FEV1)
M o s t of the patients in this study had more than one ED visit in which e x p i r a t o r y v o l u m e s were m e a s u r e d . For t h i s reason, m e a n e n t r y FEV1, m e a n discharge FEV u and m e a n recovery were calculated. Mean entry FEV 1 was t h e average FEV 1 for all visits in w h i c h FEV 1 was measured before treatment, whereas m e a n discharge FEV 1 was average discharge FEV 1. Mean recovery was the ratio of mean discharge FEV 1 to the best preA n n a l s of Emergency Medicine
TABLE 1. Baseline pulmonary function tests for 269 patients with obstructive lung disease Before After Bronchodilator Bronchodilator FEV1 (L) % Predicted FVC (L) % Predicted FEV1/FVC
1.38 _+ 0.58
1.49 ~ 0.59
43.8 ± 17.4
46.8 + 17.6
2.62 _+ 0.84
2.88 + 0.84
60,0 + 18.6
65.9 _+ 18.6
53.0 -+ 15.3
52.4 ÷ 16.2
bronchodilator FEV 1 on baseline PFTs. The frequency of use for each therap e u t i c i n t e r v e n t i o n was also calculated for each patient. A l l a v a i l a b l e p a t i e n t m e d i c a l records were reviewed. Information was extracted regarding ED visits and PFTs and recorded on forms designed for c o m p u t e r entry. D a t a were entered into a customized, microcomputer data base and analyzed by a commercial statistical package (Systat®). A previous study from our institution s h o w e d t h a t p a t i e n t s w i t h obstructive lung disease were at excessive risk for recurrent symptoms for at least 14 days. 6 For this reason, return v i s i t s o c c u r r i n g w i t h i n t h i s period were considered relapses. For historical, diagnostic, and t r e a t m e n t variables in the ED, visits ending in relapse were compared with visits that did not. For demographic characteristics, PFT variables, and frequency of therapeutic interventions, patients w i t h one or m o r e relapse visits (relapsers) were c o m p a r e d w i t h those who did not relapse at any time during the study period (nonrelapsers). D i f f e r e n c e s in t h e p r o p o r t i o n o~ n o m i n a l v a r i a b l e s b e t w e e n groups were tested by x 2 analysis. The distributions of continuous variables were e x a m i n e d by f r e q u e n c y histograms. Group differences for normally distributed variables were analyzed by the unpaired Student's t test, whereas all others were analyzed by the MannW h i t n e y U test. M u l t i p l e linear regression Was used to define the relationship b e t w e e n the n u m b e r of relapse visits for each patient and his demographic characteristics, PFTs, and c o m p o s i t e d i a g n o s t i c and treatment factors. D i a g n o s t i c factors included m e a n e n t r y FEVI, m e a n discharge FEV u and mean recovery. Treatment factors included the m e a n number of nebulized bronchodilator treatments 20:2 February 1991
COPD Murata et al
TABLE 2. Comparison of successful and unsuccessful ED visits
Visits Ending in Relapse (281)
Visits Not Ending in Relapse (596)
3,40 ± 5.24
4.63 ± 6.40
Historical Features Duration of dyspnea (days)
.002
ED diagnosis of "asthma" (%)
12.1
15.3
NS
Change in sputum (%)
34.9
35.7
NS
Patienton oral prednisone at entry (%)
26.7
25.0
NS
97.9 ± 1.1 100.2 ± 16.3
97.8 ± 1.2 97.9 ± 16.5
NS NS
26.3 ± 6,4
25.9 ± 6.3
NS
Admission FEV1 (mL) (489)
682 ± 303
766 _+ 419
.027
DischargeFEV1 (mL) (279)
899 ± 415
1,033 ± 554
,040
Arterial blood gases on room air (545) pH Po2 (ram Hg) Pco2 (mm Hg) 02 saturation (%)
7.39 59,1 37.9 87.7
7.39 59.8 38.4 87.4
NS NS NS NS
Broodcounts (378) Hematocrit White count ( x 1,000)
45.7 _+ 4.2 8.9 ± 2.8
46.6 ± 4.6 9.2 ± 3.2
NS NS
Physical Findings Temperature(F) Pulse Respirations
Laboratory Tests
+_ 0.06 ± 12.0 ± 7.5 ± 7.6
± -+ ± ±
0.06 12.3 9.2 8.1
Treatment No, of nebulizer treatments
1.22 ± 0.76
1.08 ± 0.82
.009
Parenteral sympathomimetics used in ED (%)
39.9
30,5
.006
IV corticosteroids given in ED (%)
17.4
19.6
NS
IV aminophylline used in ED (%)
19.9
15.3
NS
Oral prednisone on discharge (%)
18.1
25.5
.016
Antibiotics on discharge (%)
29.5
28.4
NS
8.5
6.4
NS
68.9 ± 25.2
78.8 ± 32.1
.010
New broechodilator added on discharge (%)
Response Percent recovery of baseline FEV1 (279)
~lues a~ expressed as mean ± SD. Pe~enlages ~fer to the p~portion of visits in which the characteristic was found.
and the frequency with which the patient was treated with IV aminophylline, IV corticosteroids, parenteral adrenergic drugs, antibiotics, and oral prednisone. RESULTS
During a period of three years, 352 20:2 February 1991
patients with irreversible, obstructive lung disease were treated for dyspnea, cough, and/or excessive sputum production in the ED of the Albuquerque Veterans Administration Medical Center. Eighty-three patients (23.6%) were admitted to the hospital on every visit. The remainAnnals of Emergency Medicine
ing 269 were treated and released from the ED on one or more occasions and were the study population of this report. The mean (+ SD) age of the latter group was 65.8 _+ 8.7 years, and all but three were men. Eighty-seven (32.3%) were on home oxygen therapy at entry, and 125 (46.5%) were given oral prednisone at least once during the study period. Baseline PFTs for the patient sample are shown (Table 1). The mean RI was 11.5 _+ 12.9%, and all subjects had persistent baseline defects in FEV1 after inhaled bronchodilators. The 269 subjects in this study made 1,157 visits to the ED during the three-year study period. Two hundred eighty visits (24.2%) resulted in admission to the hospital and were excluded from further analysis. Of the 877 visits in which the patient was treated and released from the ED, 281 (32.0%) ended in relapse and were considered unsuccessful. The remaining visits were classified as successful. In 25 visits (2.9%), the patients were considered to have left ventricular failure in addition to decompensated obstructive lung disease.
The clinical features of unsuccessful and successful visits are compared (Table 2). Duration of dyspnea was the only historical feature that had prognostic value. The risk of relapse for a duration less than 4.2 days (36.6%) was s i g n i f i c a n t l y greater than for a duration of 4.2 days or more (25.3%; P = .010). Few patients had all of the laboratory tests performed. However, for each of the tests shown, performance of the test had no effect on probability of relapse. Entry FEV 1 and discharge FEV 1 were significantly higher for successful than for unsuccessful visits. When visits were stratified by entry FEVu there was no significant difference in the relapse rate between visits in which entry FEV 1 exceeded the mean for all visits (740 mL) and those in which entry FEV 1 was less. Stratification by discharge FEV t also failed to identify visits at risk for subsequent relapse. There were several treatment differences between unsuccessful and successful visits. The number of nebulizer treatments for unsuccessful visits was significantly higher than for successful visits. Parenteral adrener~ic drugs were also used more often for unsuccessful visits, whereas 127/33
COPD Murata et al
oral prednisone was given less frequently at the time of discharge. Despite more intensive therapy, PR was less for unsuccessful than for successful visits. No differences were found in the use of IV corticosteroids, IV or oral theophylline, antibiotics, or other bronchodilators. The relapse rate for visits in which PR was more than 75% was similar to the rate for visits in w h i c h PR was less than 75%. Of the 269 patients in this study, 101 (37.5%) had one or more relapse visits in three years; 45.6% of the visits made by this group ended in relapse. Relapsers differed from nonrelapsers with respect to several characteristics (Table 3). On baseline PFTs, the RI was significantly higher for relapsers than for nonrelapsers. No differences were noted with respect to FEV1, FVC, or the FEVI:FVC ratio either b e f o r e or after bronchodilators. Relapse patients also presented with a shorter duration of symptoms, were given more nebulizer treatments, and were treated more frequently with parenteral adrenergic drugs, IV corticosteroids, and oral prednisone. For 189 patients (70.3%), it was possible to calculate mean entry FEV I. Relapsers had a significantly lower value of mean entry FEV 1 than nonrelapsers in this subset. For 133 patients (49.4%), it was also possible to calculate mean discharge FEVf and mean recovery. Despite more intensive treatment, relapse patients recovered less of their baseline FEV 1 at the time of discharge. Stepwise multiple linear regression was used to examine the relationship between number of relapse visits and several baseline PFT and treatment variables. RI, number of nebulizer treatments, use of parenteral adrenergic drugs, and oral prednisone on discharge were shown to be independent variables in this model. Although highly significant (P < .001), this model explained very little of the variance in relapse number (multiple R 2, 0.202). Addition of mean entry FEVu mean discharge FEVu and mean recovery did not improve the fit for the subset of patients in which these variables were available.
DISCUSSION COPD is a major cause of death, disability, and e c o n o m i c hardship among elderly patients in the United 34~28
TABLE 3. Comparison of relapsers and nonrelapsers Relapsers (101)
Nonrelapsers (168)
P
65.9 + 8,1
65.7 _ 9.1
NS
37.6
29.2
NS
4.57 ± 5.85
6.39 ± 7.27
.025
Prebronchodilator FEV1 (L) % Predicted FVC (L) % Predicted FEVl/FVC
1.40 45.3 2.69 61.9 51.1
± ± 4± ±
0.61 18.0 0.85 19.7 13.8
1.37 43.0 2.58 58.9 54.1
± ± ± ± ±
0.57 17.0 0.83 17.9 16.1
NS NS NS NS NS
Postbronchodilator FEV1 (L) % Predicted FVC (L) % Predicted FEVJFVC
1.53 48,0 2.89 66.4 51.2
± ± ± ± 4-
0.62 18.8 0.87 20.1 14.3
1.47 46.0 2.87 65.6 53.2
_+ 0.57 ± 16.9 ± 0.83 ± 17.6 ± 17,2
NS NS NS NS NS
Age (yr) Home oxygen use (%) Mean duration of dyspnea (days)
Baseline PFI's
Response index (249)
13.6 +_ 15.5
10.3 ± 10.8
.047
Mean entry FEV1 (mL) (189)
669 ± 296
861 ± 478
.00I
Mean discharge FEV~ (mL) (133)
903 ± 424
1,054 ± 541
Mean recovery (%) (133)
70.9 ± 27.9
84.1 ± 32,9
0.14
Mean no. of nebulizer treatments
1.21 ± 0.63
0.88 -+ 0.76
< .001
Parenteral adrenergic drugs (%) IV aminophyliine (%)
32.3 ± 30.9 15.7 ± 21.5
18.8 ± 34.9 14.5 ± 32.4
< .001 NS
NS
Treatment
IV corticosteroids (%)
18.3 ± 20.0
10.6 4- 26.3
< .001
Prednisone on discharge (%)
21.0 ± 21.7
13,8 ± 30.3
< .001
Antibiotics on discharge (%)
29.0 ± 28.9
32.0 4- 43.4
NS
Percenlages refer to the mean -+ SD frequency with which each inlervention was used.
States. Because it cannot be cured, preservation of the patient's functional status as an outpatient may be the most cost-effective strategy for this disease. Such a strategy would include the prevention of respiratory decompensation and effective outpatient management once decompensation occurs. Unfortunately, there is little evidence that outpatient therapy is effective for all COPD patients. At our institution, one fourth of the patients admitted for decompensated COPD were treated and released from our ED within the preceding two weeks. This observation suggested that modest improvements in the o u t p a t i e n t m a n a g e m e n t of COPD might produce significant reductions in the costs and hardship attributed to this disease. In a previous study, we found that 28% of ED visits made by patients Annals of Emergency Medicine
with the clinical diagnosis of asthrn~ or COPD ended in relapse. 6 Thil problem was confined largely to~ small group of patients who had clin ical features of both a s t h m a an~ COPD. It was unclear if relapse oc curred because of greater severity c illness at the time of presentation o: differences in the manner in whict these patients were treated. The put pose of this study was to examin~ the relationship among the clinicl features of COPD, its outpatient m a n a g e m e n t , and the short-ter~ prognosis for a smaller group of pa tients with well-documented, irre versible airway obstruction. We found that unsuccessful visits were characterized by a short dura tion of dyspnea, a higher number 01 nebulizer treatments, the use of pa enteral adrenergic drugs, and fail~ to give oral prednisone on discharge. i:
20:2 February 19~!
COPD Murata et al
For visits in which spirometry was performed, e n t r y FEV1, discharge FEV1, and PR were s i g n i f i c a n t l y smaller for unsuccessful than successful visits. However, stratification of ED visits with respect to the latter variables failed to produce differences in relapse rate among the different categories of risk. Relapsers had a higher bronchodilator response on baseline PFTs than nonrelapsers and were treated more vigorously with nebulized bronchodilators, parenteral adrenergic drugs, and IV and 0ral glucocorticoids in the ED. Our conclusions differ from a large body of literature that suggests that bronchodilator response on baseline PFTs cannot be used to predict the response to bronchodilator treatment in COPD. 7 Several reasons for the validity of these criticisms include the following: day-to-day variation in FEV1 is large; response is usually expressed as percent improvement over baseline flow, and small changes in patients with severe reduction in FEV1 might therefore be interpreted as a response; FEV~ may not be the best measure of obstruction; there is poor correlation between FEV 1 and dyspnea or exercise i n t o l e r a n c e ; bronchodilators may be more important in preventing deterioration than in improving baseline e x p i r a t o r y flow; and the benefits of bronchodilat0rs may be due to effects other than improved air flow. We designed this study to circumvent many of these problems. When several PFTs were available for a given patient, we selected the faighest value of FEV1 for analysis. Response to bronchodilators was expressed as the percentage of the defect in FEV1 corrected, not as a percentage change over baseline. The end point for this study was the patient's perception of his own clinical status, not a change in FEV1. Our experience was large enough for clinically important differences to be apparent. Nevertheless, our conclusions about the value of the baseline bronchodilator response as a p r o g n o s t i c i n d i c a t o r should be interpreted with caution and must be validated in a prospective manner. Pulmonary function criteria for the classification of ambulatory patients with obstructive pulmonary disease have not been established. We chose liberal criteria for this study because we expected patients treated in an 20:2February 1991
outpatient setting to have lesser degrees of airflow obstruction than hospitalized patients reported in other studies. The use of a more restrictive d e f i n i t i o n of airflow o b s t r u c t i o n would have resulted in the classification of many chronically symptomatic patients as "normal." Local custom is to measure pulmonary function when the patient has reached a maximal degree of recovery after an episode of respiratory decompensation; there were no time requirements for the definition of this baseline period. Such criteria would have been arbitrary and would not have accounted for the markedly unstable respiratory function in some of our patients. Several made 15 to 20 ED visits per year and would never have attained a baseline status defined by stable lung function over a period of several weeks. We did not wish to use selection criteria that would have removed the most problematic patients from the study. It is unclear why patients with a bronchodilator response had such a poor prognosis. The correlation between the relapse number and a variety of PFT and treatment variables was poor. This finding suggests that bronchodilator response is a marker for another process that is causally related to outcome. Part of the explanation is suggested by our comparisons of relapsers and nonrelapsers. Relapsers had a higher bronchodilator response at baseline but required much more intensive therapy in the ED for respiratory decompensation. Among those who had entry and discharge FEVlS recorded, relapsers recovered much less of their baseline FEV 1 in the ED despite more frequent use of parenteral adrenergic drugs, IV c o r t i c o s t e r o i d s , and a greater number of treatments with nebulized bronchodilators. It is possible that respiratory decompensation occurs in some COPD patients because of a change in airway sensitivity to bronchodilators. Studies have shown that exposure of the f~-adrenergic receptor to excessive stimulation results in tachyphylaxis 8 and that corticosteriod therapy reverses this defect.9, lo We believe that loss of airway sensitivity to bronchodilators should be studied prospectively as a mechanism for respiratory decompensation. We found that oral prednisone was the only treatment factor associated Annals of Emergency Medicine
with a good prognosis. This finding did not result from a patient selection bias because prednisone was used more frequently in patients with one or more relapses. It is also unlikely that prednisone was used for milder episodes of respiratory dec o m p e n s a t i o n . N e v e r t h e l e s s , the value of prednisone in the outpatient m a n a g e m e n t of d e c o m p e n s a t e d COPD should be confirmed by a randomized clinical trial. CONCLUSION Relapse is common for COPD patients who have a short duration of symptoms, who are treated with parenteral adrenergic drugs in the ED, or who are not given oral prednisone on discharge. Some patients who have a bronchodilator response on baseline PFTs lose their response during respiratory decompensation. This phenomenon might explain why bronchodilator response on baseline PFTs is so poorly correlated with the clinical response to bronchodilator therapy and w h y outpatient treatment programs are often unsuccessful. A prospective study should be conducted in other settings to confirm our findings.
REFERENCES 1. Nicotra MB, Rivera M, Awe RJ: Antibiotic therapy of acute exacerbations of chronic bronchitis. Ann Intern Med 1982;97:18-21.
2. Anthonisen NR, Manfreda J, Warren CPW, et aI: Antibiotic therapy in exacerbations of chronic obstructive pulmonary disease. Ann Intern Med 1987;106:196-204. 3. Rice KL, Leatherman JW, Duane PG, et al: Aminophylline for acute exacerbations of chronic obstructive pulmonary disease: A controlled trial. Ann Intern Med 1987; 107:305-309. 4. Albert RK, Martin TR, Lewis SW: Controlled clinical trial of methylprednisolone in patients with chronic bronchitis and acute respiratory insufficiency. An n I~tern Med 1980;92:753-758. 5. Patient Treatment File: National Statistics for Fiscal Year 1987. Washington, DC, Veterans Administration
Central Office. 6. Murata GH, Gorby MS, Chick TW, et al: Use of emergency medical services by patients with decompensated obstructive lung disease. A n n Emerg Me d 1989;18:501-506. 7. Pratter MR, Irwin RS: Predicting response to bronchodilator therapy in chronic obstructive pulmonary disease. Arch Intern Med 1988;148:1909-1910. 8. Stiles GL, Caron MG, Lefkowitz RJ: JS-Adrenergic receptors: Biochemical mechanisms of physiological regulation. Physiol Rev 1984;64:661-743. 9. Hui KKP, Conolly ME, Tashkin DP: Reversal of human lymphocyte [3-adrenoceptor desensitization by glucocorticoids. Clin Pharmacol Ther 1982,;32:566-571. 10. Brodde OE, Brinkmann M, Schemuth R, et al: Terbutaline-induced desensitization of human lymphocyte [32-adrenoceptors: Accelerated restoration of lS-adrenoceptor responsiveness by prednisone and ketotifen. J Clin Invest 1985;76:1096-1101.
129/35
