Antibiotics and steroids for exacerbations of COPD in primary care: compliance with Dutch guidelines

Patient population

Four primary healthcare centres took part in the study. These centres participate in the GP-based continuous morbidity registration network of the Department of General Practice, Academic Medical Centre — University of Amsterdam. This network registers chronic diseases among other data in electronic medical records (EMRs). For the period March 2001 until March 2003, we retrospectively analysed EMRs of all registered patients at these four centres.

Data collection

Records of all patients registered with a diagnosis of COPD were retrieved from EMRs with the use of the international classification of primary care code (ICPC) R95.11 Consultations for these patients during the 2-year study period were analysed by printing and subsequently reading and coding the EMRs. Two criteria were used to assess an exacerbation of COPD:

• when ‘exacerbation’ was used by the GP as stated in the EMR; and

• when increased symptoms of COPD (cough, shortness of breath, altered sputum conditions, increased sputum production, wheezing, or illness) were explicitly mentioned in the EMR.

When patients visited their GP for an exacerbation more than once in a period of up to 3 weeks, the whole episode was considered one exacerbation. An exacerbation after full recovery from the previous exacerbation was considered a new episode, according to the GP's description in the EMR, or when the time between two consultations at the GP practice was longer than 3 weeks. Patients who were suspected to have a diagnosis of pneumonia, based on symptoms and physical examination, were excluded.

The number of patients with exacerbations, number of exacerbations per patient, daily medication, and treatment prescribed for the exacerbation were registered. Exacerbations treated by pulmonologists after having been referred by a GP were counted separately.

Statistical analysis

SPSS (version) 11.5.1 was used for statistical analysis. Differences between patients groups were analysed using χ² test. Statistical significance was set at P<0.05.

Characteristics of primary healthcare centres and patients

During the 2-year study period, 35 589 patients were registered at the four primary healthcare centres. Of these patients, 458 were registered with a diagnosis of COPD (1.3%). Median age of patients with COPD was 70 years (interquartile range = 61–77 years). Daily medication for COPD was taken by 380 (84%) patients: 83 (18%) inhaled β₂ agonists and/or anticholinergics only and 297 (66%) inhaled corticosteroids, with or without β₂ agonists and/or anticholinergics. One hundred patients (22%) used N-acetylcysteine, 22 of them without inhaled corticosteroids.

A total of 572 exacerbations in 2 years were identified in 458 patients (0.62/patient/year). In 215 patients (47%) no exacerbation in 2 years occurred, 162 (35%) had one or two exacerbations, and 81 (18%) exacerbated more often (Table 1). In patients with exacerbations, there was a median of two exacerbations over 2 years (range = 1–13)

Treatment of exacerbations

Of the 572 exacerbations, 36 (6%) were treated by pulmonologists. In the remaining 536 exacerbations, GPs prescribed a short course of oral steroids in 161 cases (30%), antibiotics in 157 (29%), and steroids were combined with an antibiotic in 121 (23%). In 97 cases (18%) no steroid course or antibiotic was prescribed, but compliance to daily medication use was checked and/or this medication was adapted (Table 1).

Treatment was influenced by patient characteristics. Compared with patients with one or two exacerbations, patients with three or more exacerbations were prescribed significantly more steroid courses and steroids combined with antibiotics, and less antibiotics without steroids or no antibiotics or steroids (P≤0.001 in all cases). In patients younger than 70 years, antibiotics without steroids were prescribed more often than in patients over 70 years of age (P = 0.02) (Table 2).

Steroid courses were usually prescribed in a dose of 30 mg per os daily for 7 or 10 days. The most frequently prescribed antibiotics were amoxycillin/clavulanate (co-amoxiclav) for 7 or 10 days (34%), doxycycline for 8 days (28%), amoxycillin for 7 days (18%), and azithromycin for 3 days (12%).

Summary of main findings

In this retrospective analysis of the medical records of patients with COPD in general practice, almost half of the patients had no acute exacerbation in the 2-year study period, one-third had one or two exacerbations, and one in five patients with COPD exacerbated more often. In instances where patients had an exacerbation of COPD, GPs often did not prescribe treatment in accordance with the Dutch College of General Practitioners guidelines for COPD.10 Antibiotics were often prescribed (52% of the exacerbations), and in almost 30% an antibiotic was the only medication prescribed.

GPs in the four primary healthcare centres treated half of the exacerbations with steroid courses. Evidence from the literature indicates that short courses of steroids are beneficial in exacerbations of COPD,1^–3 and most international guidelines advise short courses of oral steroids for patients at home.12^,13 However, only the GOLD guidelines (Global Initiative for Chronic Obstructive Lung Disease) include FEV1 below 50%.14 The Dutch College of General Practitioners guidelines for COPD are even more restrictive by advising short courses of oral steroids only in more severe exacerbations. In this sense, GPs appear to follow the Dutch guidelines.10

GPs in this study prescribed antibiotics in 23% of cases combined with steroids and in 29% of the cases without steroids. The Dutch guidelines for COPD is restrictive regarding when to prescribe antibiotics for exacerbations of COPD. They are more restrictive than the international standards, which mention altered sputum characteristics as an indication for antibiotics. It is unlikely that slightly more than half of all exacerbations in the current study fulfil the criteria of the Dutch guidelines for prescribing antibiotics. The prescription of antibiotics for exacerbations of COPD in daily practice will be even higher, as exacerbations with signs of pneumonia were excluded from the study. The possibility that GPs may sometimes label an exacerbation as a pneumonia to justify prescribing antibiotics cannot be excluded.

Guidelines are there to assist GPs to treat patients using an evidence-based approach and are not a protocol that always has to be followed in a strict way. As in general practice assessing the severity of an exacerbation may be difficult in some cases, GPs may prescribe antibiotics to be on the safe side. It is unclear why antibiotics were prescribed much more frequently than indicated in the Dutch guidelines. The Dutch guidelines may be too stringent to be followed in daily practice; for example, GPs may suspect that withholding antibiotics may lead to hospitalisation. Alternatively, deviation from the guidelines may not be clinically justified in many cases. It is unclear whether guidelines or prescribing patterns should be changed, and further study is required.

Prescriptions in patients with one or two exacerbations differed significantly from prescriptions for patients with three or more exacerbations. Short courses of oral steroids were prescribed more frequently in patients with three or more exacerbations, and antibiotics only were prescribed less frequently. Treatment in patients with three or more exacerbations was more in accordance with the Dutch guidelines than treatment in patients having one or two exacerbations.

Limitations of the study

Analysing medical records has its limitations. Because this evaluation was conducted retrospectively, information about specific symptoms during exacerbations of COPD could not be obtained. Although all patients were registered using the ICPC code for COPD, and the diagnoses of COPD made by GPs were confirmed by spirometric testing for almost all patients, it cannot be ruled out that the study might have included some patients with asthma.

Comparison with existing literature

Research published in 1998 showed that GPs prescribe antibiotics more frequently and steroid courses less frequently than indicated by the Dutch guidelines.19 Although the Dutch guidelines have been revised several times since then, little has changed in the treatment of exacerbations of COPD in primary health care. An evaluation of the most important barriers to change is indicated.20

Implications for clinical practice

In the general population there is an association between antibiotic consumption and resistance to pathogens.15^–18 Unnecessary use of antibiotics should be avoided to prevent antimicrobial resistance of pathogens. Therefore, the current authors suggest that concerted efforts should be made to fully implement the Dutch guidelines for COPD.