Aminophylline

The objective of this study was to determine if IV aminophylline reduces the risk of relapse after treatment of decompensated COPD in an ED. Forty-six visits in which IV aminophylline was given (T visits) were compared with an equal number of visits in which it was withheld (N visits) with respect to pretreatment serum theophylline level, number of treatments with nebulized bronchodilators and use of parenteral [beta]-adrenergic drugs, IV corticosteroids and prednisone. The difference in 48-h relapse rates for T and N visits was examined by McNemar's test. No differences were found between T and N visits with respect to vital signs, pretreatment FEV[sub.1.], arterial blood gas values, hematocrit level or blood leukocyte count. The 48-h relapse rate for T visits (22.2 percent) was significantly higher than for N visits (6.7 percent; p=0.035). Aminophylline does not appear to be beneficial for outpatients with decompensated COPD and may be harmful.

IV=intravenous; COPD=chronic obstructive pulmonary disease; ED=emergency department; [FEV.sub.1]=forced expiratory volume in 1 s; PFTs=pulmonary function tests; VC=vital capacity; FVC=forced vital capacity; pH=negative logarithm of hydrogen ion activity; [PO.sub.2]=partial pressure of oxygen; [Pco.sub.2]=partial pressure of carbon dioxide

The role of theophylline in the management of COPD is controversial. [1-4] Certain clinical trials have shown that theophylline is not beneficial for patients with stable COPD. [5-7] However, others have shown that treatment reduces symptoms, increases expiratory flow rates and improves diaphragmatic function. [8] Less information is available about the value of IV aminophylline for patients with acute exacerbations of COPD. Rice et al [9] conducted a randomized clinical trial of aminophylline on 30 hospitalized patients with respiratory decompensation. Treatment had no effect upon the rate of improvement in symptoms or [FEV.sub.1]. No information is available about the value of IV aminophylline for outpatients with decompensated COPD.

The purpose of thi study was to determine if IV aminophylline reduces the relapse rate after treatment of decompensated COPD in an ED. A retrospective analysis was done on all visits over a four-year period in which aminophylline was given to patients with serum theophylline levels less than 10 mg/L. Visits in which aminophylline was given were compared with visits in which it was withheld. Pair-wise matching was used to control for other treatment factors.

METHODS

Procedures and Patient Selection

The Albuquerque VA Medical Center is a 300-bed acute-care facility serving veterans in New Mexico and west Texas. The ED is staffed by a large number of faculty and house officers from the University of New Mexico. Patients who come to the medical center with decompensated COPD are referred to the ED for evaluation and therapy. The treatment for COPD is not standardized at this institution. All decisions regarding medications are made by the attending physician on duty and are based on his assessment of the patient's condition. Those who respond to therapy are instructed to return to the ED if symptoms recur. During this survey, no attempt was made to influence the physician's decision to use aminophylline.

Information was extracted from the medical record for all visits for respiratory complaints for a period of four years. Data were recorded on standardized forms and entered into a microcomputer. The visits reported in this study were found by searching these computer files. A visit was considered for this study if: (1) the patient was given the diagnosis of COPD, bronchitis or emphysema; (2) the visit was for increased shortness of breath; and (3) the patient had at least one set of baseline PFTs that showed chronic airflow obstruction.

Pulmonary function tests were reviewed if they were performed within three years of the patient's first visit during the observation period. A patient was considered to have chronic airflow obstruction if (1) the [FEV.sub.1] was less than 75 percent predicted and (2) the [FEV.sub.1./VC] was less than 75 percent on all baseline PFTs, both before and after use of inhaled bronchodilators. Bronchodilator response was expressed as the percentage of improvement in [FEV.sub.1] over baseline. When multiple PFTs were performed within three years of entry, only the highest values for [FEV.sub.1], VC and bronchodilator response were used.

Visits were excluded if the patient had pneumonia, pneumothorax, pleural effusion, pulmonary embolism, left heart failure, pulmonary fibrosis, occupational lung disease or clinical features of asthma. A patient was considered to have asthma [10] if he had one or more of the following: (1) onset of symptoms before the age of 40 years; (2) a history of eczema, atopic dermatitis, multiple allergies or aspirin sensitivity; or (3) minimal symptoms between episodes of respiratory decompensation.

Visits in which aminophylline was given (T or treatment visits) were compared with matched visits in which it was withheld (N or nontreatment visits). For each T visit, one N visit was selected as a control. A pair of visits was considered a match if the following conditions were met: (1) the difference in the pretreatment serum theophylline level was less than 3 mg/L; (2) the number of treatments with inhaled bronchodilators differed by one or less; and (3) parental [beta]-adrenergic drugs, IV corticosteroids and prednisone were either given or withheld for both members of the pair.

Controls were chosen so that there was no difference in the number of visits in which an antibiotic or an increased dose of oral theophylline was given at the time of discharge. Matches were performed by one of the authors (G.H.M.) blinded with respect to outcome. To avoid weighting the data by including patients with multiple visits, no patient was allowed to contribute more than three visits to any treatment category or more than four to the entire study.

All patients treated for decompensated COPD over the four-year period were categorized according to the frequency of respiratory decompensation and the likelihood of relapse. A patient was considered "resource-intensive" if he made eight or more visits for respiratory distress. The 51 patients in this category made over half of all ED visits. A patient was considered "high risk" if he relapsed within 48 h on 25 percent or more of the visits in which aminophylline was withheld. The 34 patients in this category were responsible for 70 of the 99 relapses during the study period.

Statistical Methods

Data were analyzed by a commercial statistical package (Systat). Continuous variables are expressed as mean [+ or -] SD. Analysis of variance was used to evaluate differences in continuous variables among patient categories. Differences in the proportion of nominal variables between T and N visits were examined by McNemar's test. Differences in continuous variables were analyzed either by the paired Student's t test or by the Wilcoxon signed ranks test. Confidence intervals for the differences in relapse rates for T and N visits were calculated by the method of Snedecor and Cochran. [11]

RESULTS

Over a period of four years, there were 63 visits in which aminophylline was given to a patient with decompensated COPD and a serum theophylline level of less than 10 mg/L was recorded. Forty-six of these visits (T visits) were compared with an equal number of matched visits in which aminophylline was withheld (N visits). The remaining visits were not used because no matches could be found (12 visits) or because they were made by patients who had already contributed four visits to the study (five visits).

Sixty patients were responsible for the 92 visits in this study. All but one of the subjects were men. Fourteen patients contributed visits to both treatment categories (group 1). Twenty-two patients accounted for the remaining T visits (group 2), while 24 were responsible for the remaining N visits (group 3). The clinical features of these groups are shown in Table 1. Group 1 patients were more likely to be receiving oxygen at home (chi-square = 7.24; p = 0.027). However, the groups were similar with respect to age, number of ED visits over the four-year period or the number of relapses at 48 h. No differences were found for [FEV.sub.1], [FEV.sub.1]/VC ratio or bronchodilator response on baseline PFTs.

The T and N visits are compared in Table 2. Duration of symptoms, entry vital signs, arterial blood gas levels, hematocrit value and leukocyte count were the same for the two types of visits. The [FEV.sub.1] was measured upon entry to the ED for 26 T visits and for 28 N visits. The pretreatment [FEV.sub.1] for T visits (0.60 [+ or -] 0.21 L) was similar to the [FEV.sub.1] for N visits (0.67 [+ or -] 0.22 L; p = NS). No differences were found in the proportion of T and N visits that were made by resource-intensive patients, high-risk patients, oxygen-dependent patients or patients on a maintenance regimen of prednisone at the time of presentation. One-fourth of the visits in each category were preceded by another visit within 14 days.

The pretreatment theophylline level for T visits (4.7 [+ or -] 3.2 ml/L) was similar to the level for N visits (4.8 [+ or -] 3.0 mg/L). The mean dose of aminophylline given during T visits was 252 [+ or -] 96 mg. The length of stay in the ED was significantly longer for T than for N visits.

One of the patients given aminophylline developed pneumonia within 24 h. This visit and its match were excluded from further analysis. At 48 h, the relapse rate for T visits (22.2 percent) was significantly higher than the rate for N visits (6.7 percent; p = 0.035). The difference in the 48-h relapse rate was 15.5 percent (95 percent confidence interval: 0.7 to 30.3 percent). Relapses

[TABULAR DATA OMITTED]

[TABULAR DATA OMITTED]

occurred ten times after aminophylline treatment and involved nine patients. All of these patients complained of recurrent dyspnea and were retreated with inhaled bronchodilators. Serum theophylline levels were significantly higher at the time of relapse (12.4 [+ or -] 7.0 mg/L) than they were before treatment (5.3 [+ or -] 2.8 mg/L; p = 0.006). Levels were less than 10 mg/L in three of ten cases and more than 20 mg/L in one. All six patients requiring admission to the hospital within 48 h had been treated with aminophylline.

DISCUSSION

The methylxanthines have many biologic properties that could be of benefit to patients with COPD. Their ability to dilate smaller airways has been known for over 50 years. [12] Other actions include improved mucociliary clearance, [13] stimulation of respiratory drive, [14] improved right ventricular performance [15] and increased diaphragmatic contractility and endurance. [8] Despite these properties, several randomized clinical trials in patients with stable COPD [5-7] and in hospitalized patients with respiratory distress [9] have not shown that theophylline is beneficial. No information is available about the role of IV aminophylline in the outpatient management of decompensated COPD.

The purpose of this study was to determine if IV aminophylline reduces the relapse rate for patients with decompensated COPD treated in an ED. This study was restricted to patients who were taking oral theophylline and had a subtherapeutic level at the time of presentation. Visits in which aminophylline was given were compared with visits in which aminophylline was withheld. Pair-wise matching was used to control for pretreatment theophylline level, number of treatments with nebulized bronchodilators and the use of parenteral [beta]-adrenergic drugs, intravenous corticosteroids and prednisone. Controls were selected so that there was no difference in the number of patients given an antibiotic or an increased dose of oral theophylline at the time of discharge. We found that patients treated with IV aminophylline had a significantly higher relapse rate at 48 h. Examination of confidence intervals showed that there was less than a 5 percent probability that aminophylline would prevent more than one relapse in every 100 cases.

These findings could have resulted from differences in the prognosis of T and N visits at the time of treatment. We therefore examined a large number of factors that could affect the relapse rate. A small number of patients contributed visits to both treatment groups. For the remaining subjects, no differences were found between treated and untreated patients with respect to age, number of visits or relapses over four years, baseline [FEV.sub.1] or baseline bronchodilator response. Initial vital signs, pretreatment FEV [1], arterial blood gas values and hematologic values were similar for T and N visits. There were no differences in the proportion of visits made by patients who were oxygen-dependent or receiving prednisone at the time of entry. Patients who were frequent users of emergency services or who had a history of relapse also were equally represented. One fourth of the visits in each group occurred within two weeks of a prior visist, suggesting that aminiphylline was not used more frequently in the setting of a recent treatment failure.

Only two features distinguished T from N visits. The length of stay in the ED was 50 min longer for T visits. This observation was expected, since aminophylline was administered by a slow infusion. Matching eliminated any other treatment differences except for oxygen therapy and IV fluids. Since a longer length of stay increases the need for more oxygen and fluid therapy, this difference should have biased the outcome in favor of T visits. The T visits also occurred earlier in the study period than N visits. However, the mean interval between each T visit and its control was only 4.8 months, and it is unlikely that long-term trends could have resulted in a better prognosis for N visits. We have been unable to demonstrate that the relapse rate for decompensated COPD has changed at our institution in the last four years. Thus, the lack of benefit for aminophylline could not be attributed to major differences between the two types of visits.

Nevertheless, it is difficult to remove selection biases from a retrospective study of COPD. One reason is that cause of relapse for patients with respiratory decompensation is unknown. Although T and N visits were similar, T visits could have had a poorer prognosis because of risk factors not examined by this study. If that were the case, the benefits of aminophylline would have been obscured. Since this bias can be reduced by random assignment to treatment categories, we feel that a prospective clinical trial should be done to confirm our results.

There are several reasons why theopylline may not be beneficial for patients with decompensated COPD. Theophylline has little effect in patients receiving maximal doses of [beta]-adrenergic drugs. Barclay et al [16] studies the effect of bronchodilators on FVC in ten patients with bronchitis. Subjects were treated with salbutamol until a plateau of response was achieved. Infusion of theophylline at that point resulted in only slight increases in FVC in four patients. In addition, the relationship between serum theophyline levels and improvement in respiratory function is log-linear. [17] Changes in lung function therefore become small as the serum level of theophylline approaches the therapeutic range. Aminophylline would have been more effective if it was given to patients with levels near 0 ml/L, but there were too few of these patients in this study to determine the magnitude of this effect.

The higher relapse rate for T visits may have been due to stimulation of respiratory drive in patients with a limited capacity to respond. Vereen et al [18] have shown that theophylline increases ventilation, but not exercise capacity, in patients with stable COPD. Our treated patients may have had a greater sensation of dyspnea for any given level of exercise. Other studies have shown that theophylline inhibits the function of polymorphonuclear leukocytes and alveolar macrophages. [19-21] Clearance of bacterial pathogens may have been inhibited in the aminophylline-treated group, resulting in more protracted symptoms.

The results of this study are similar to those reported by Rice et al, [9] who conducted a randomized clinical trial of aminophylline on 30 hospitalized patients with decompensated COPD. Their patients were treated with a standardized protocol consisting of supplemental oxygen, nebulized bronchodilators, VI aminophylline was infused to maintain a serum theophylline level between 72 and 83 [mu]mol/L. No differences were noted between treated and control patients with respect to improvement in [FEV.sub.1], FVC or dyspnea. Because our study was retrospective, we do not have serum theophylline levels for all patients at 24 and 48 h. However, levels less than 10 mg/l were found in only three of ten treated patients at the time of relapse.

This study was designed to evaluate the effect of aminophylline on the use of hospital facilities by patients with COPD. The 48-h relapse rate was chosen as the outcome because it was practical and has an impact on hospital costs and staffing requirements. We did not measure pulmonary function of patients who did not return to the ED. Whether aminiphylline affects airway and diaphragmatic function in outpatients with respiratory distress can only be answered by a prospective clinical study.

In summary, the use of IV aminophylline in patients with decompensated COPD does not appear to be beneficial and possibly is harmful. These results do not appear to be the result of selection or treatment biases. Randomized clinical trials should be done before its use can be justified in the outpatient setting.

REFERENCES

[1] Rossing TH. Methylxanthines in 1989. Ann Intern Med 1989; 110:502-04

[2] Drazen JM, Gerard C. Reversing the irreversible. N Engl J Med 1989; 320:1555-56

[3] Rogers RM, Owens GR, Pennock BE. The pendulum swings again: toward a rational use of theophylline. Chest 1985; 87:280-82

[4] Dull WL, Alexander MR. Theophylline in stable chronic airflow obstruction. Arch Intern Med 1984; 144:2399-2401

[5] Eaton ML, MacDonald FM, Church TR, Niewoehner DE. Effects of theophylline on breathlessness and exercise tolerance in patients with chronic airflow obstruction. Chest 1982; 82:538-42

[6] Alexander MR, Dull WL, Kasik JE. Treatment of chronic obstructive pulmonary disease with orally administered theophylline. JAMA 1980; 244:2286-90

[7] Eaton ML, Green BA, Church TR, McGowan T, Neiwoehner DE. Efficacy of theophylline in "irreversible" airflow obstruction. Ann Intern Med 1982; 92:758-61

[8] Murciano D, Auclair M, Pariente R, Aubier M. A randomized, controlled trial of theophylline in patients with severe chronic obstructive pulmonary disease. N Engl J Med 1989; 320:1521-25

[9] Rice KL, Leatherman JW, Duane PG, Snyder LS, Harmon KR, Abel J, et al. Aminophylline for acute exacerbations of chronic obstructive pulmonary disease. Ann Intern Med 1987; 107:305-09

[10] American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dir 1987; 136:225-44

[11] Snedecor GW, Cochran WG, eds. Statistical methods. Ames: The Iowa State University Press, 1980:121-24

[12] Persson CGA. On the medical history of xanthines and other remedies for asthma: a tribute to HH Salter. Thorax 1985; 40:881-86

[13] Wanner A. Effects of methylxanthines on airway mucociliary function. Am J Med 1985; 79(suppl 6A):16-21

[14] Lakshminarayam S, Sahn SA, Weil JV. Effect of aminophylline on ventilatory responses in normal man. Am Rev Respir Dis 1978; 117:33-38

[15] Matthay RA, Berger HJ, Davies R, Loke J, Gottshalk A, Zaret BL. Improvement in cardiac performance by oral long-acting theophylline in chronic obstructive pulmonary disease. Am Heart J 1982; 104:1022-26

[16] Barclay J, Whiting B, Addis GJ. The influence of theophylline on maximal response to salbutamol in severe chronic obstructive pulmonary disease. Eur J Clin Pharmacol 1982; 22:389-93

[17] Mitenko PA, Ogilvie RI. Rational intravenous doses of theophylline. N Engl J Med 1973; 289:600-03

[18] Vereen LE, Kinasewitz GT, George RB. Effect of aminophylline on exercise performance in patients with irreversible airway obstruction. Arch Intern Med 1986; 146:1349-51

[19] Nielson CP, Crowley JJ, Morgan ME, Vestal RE. Polymorphonuclear leukocyte inhibition by therapeutic concentrations of theophylline is mediated by cyclic-3',5'-adenosine monophosphate. Am Rev Respir Dis 1988; 137:25-30

[20] Nelson S, Summer WR, Jakab GJ. Aminophylline-induced suppression of pulmonary antibacterial defenses. Am Rev Respir Dis 1985; 131:923-27

[21] O'Neil SJ, Sitar DS, Klass DJ, Taraska VA, Kepron W, Mitenko PA. The pulmonary disposition of theophylline and its influence on human alveolar macrophage bactericidal function. Am Rev Respir Dis 1986; 134:1225-28

COPYRIGHT 1990 American College of Chest Physicians
COPYRIGHT 2004 Gale Group