Q J Med 2001; 94: 373-378
© 2001 Association of Physicians
Antimicrobial management of acute exacerbation of chronic airflow limitation
From the Aintree Chest Centre, University Hospital Aintree, Liverpool, UK
Received 9 March 2001 and in revised form 3 May 2001
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Antibiotics are frequently administered for exacerbations of chronic obstructive pulmonary disease and asthma, yet their role remains unclear. We prospectively audited the antimicrobial management of 167 patients aged >50 years hospitalized for exacerbations of chronic airflow limitation. Antibiotics were commenced on admission for 151 (90%) patients (oral 52%, intravenous 38%), including 17/23 (74%) with no evidence of fever, purulent sputum, leucocytosis or inflammatory chest X-ray changes. The mean number of different antibiotics prescribed was 1.8 (range 0–6); a wide range of antibiotics and antibiotic combinations were used. Sputum samples were sent for microbiological examination in 101 (61%) patients. Sputum culture was positive in 34, but only 11 (7% of the total) had amoxycillin-resistant organisms in their sputum. Seventeen patients (10%) developed diarrhoea while in hospital. Under logistic regression analysis, total number of antibiotics prescribed (p<0.0001) and age (p=0.0062) were the two factors associated with hospital-acquired diarrhoea. Only 34% of patients had received an influenza vaccination in the winter of the study, and 10% a pneumococcal vaccination within the last 5 years. In routine clinical practice, aggressive antibiotic therapy was frequently administered to patients admitted with chronic airflow limitation, despite limited clinical, radiological and microbial indications. Excessive use of antibiotics has important implications, including morbidity (antibiotic-associated diarrhoea), cost and the potential for increased microbial antibiotic resistance. A minority of patients with chronic airflow limitation are being vaccinated against influenza and Pneumococcus.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Exacerbations of chronic obstructive pulmonary disease (COPD) and asthma account for the great majority of hospital admissions due to respiratory disease (1600 admissions annually in our health district of 350000). The cause of exacerbations is frequently unknown but infections are often implicated.
British Thoracic Society guidelines1 have suggested that antibiotics are only useful in patients with exacerbations of COPD when at least two of the following three features are present: increased breathlessness, increased sputum volume and the development of purulent sputum. Similarly, in exacerbations of asthma, there is no evidence that routine antibiotics improve patients' outcomes.2 Despite these guidelines, we believed that in hospital clinical practice, most patients admitted with exacerbations of COPD and asthma were being prescribed antibiotics. The Department of Health recommends that all patients with chronic respiratory disease should receive an annual influenza vaccination and pneumococcal vaccination every 5 years.3 A previous audit undertaken at this hospital in 1993–944 revealed that only 48% of patients admitted with exacerbation of chronic airflow limitation had received an influenza vaccination during the year of the study.
There were two main aims of this prospective audit. First, to audit antibiotic use in routine hospital clinical practice and record any problems associated with their use, and second, to audit the number of patients who had received influenza and/or pneumococcal vaccination in accordance with Department of Health guidelines.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Patient population
All patients aged 50 years or over presenting to the respiratory or infectious disease wards with an exacerbation of chronic airflow limitation over a 3-month period from October to January were eligible for inclusion into the study. For the purposes of the study, chronic airflow limitation was defined as chronic obstructive respiratory disease requiring maintenance therapy with inhaled or nebulized bronchodilators and/or inhaled corticosteroids. An exacerbation was defined as an increase in breathlessness and an increase in at least two of the following; cough frequency, cough severity, sputum volume, sputum purulence and wheeze.
Study design
This was a prospective audit of routine clinical practice. Investigation and medical management of the patients included in the study was at the discretion of the admitting physician. The investigators visited the wards daily to identify patients. The following data were collected on a specifically designed proforma by patient interview and from nursing and medical notes: antibiotic and/or oral corticosteroid prescription in the 2 weeks prior to hospitalization; influenza vaccination that year and pneumococcal vaccination in the previous 5 years; suspected immunocompromise (i.e. chronic renal failure or liver disease, long-term systemic corticosteroids or other immunosuppressant drugs); temperature on admission; sputum appearances (i.e. colourless or purulent); chest radiograph findings; leucocyte count on admission; whether and when sputum was sent for microscopy and culture, and if so, what result was obtained; antibiotic administration during and after admission; and the development of hospital-acquired diarrhoea.
Statistical analysis
Data were entered into an Access database and subsequently analysed using SPSS (release 8.0.0) statistical software. As this was an observational study, simple percentages were calculated for most parameters. Mann-Whitney U test was used for non-parametric data comparisons. Forward stepwise logistic regression analysis was performed to assess factors potentially associated with the development of hospital-acquired diarrhoea.
Ethical approval
All patients gave written informed consent to participate in the audit which was approved by South Sefton District Ethics Committee. Only one patient refused to participate in the audit.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Patient population
Entry criteria were fulfilled for 167 patients (102 males and 65 females). The mean age of the patients was 70 years (range 50–96). Further details relating to patient characteristics on admission are summarized in Table 1
. Fifty-six patients (34%) reported that they had received an influenza vaccination in the winter of the study and 17 (10%) that they had received a pneumococcal vaccination within the last 5 years.
|
The majority of patients made an uneventful clinical recovery, but three patients were ventilated in the Intensive Care Unit, of whom one died. A total of seven patients (4%) died in hospital. The median duration of hospital stay was 10 days, with a range of 2–75 days.
Microbiology
Sputum samples were sent for microbiological investigation from 101 of the 167 patients (61%), but only 26 samples were collected at the time of admission. Bacterial pathogens were isolated by culture from 34 of the 101 sputum samples (34% of samples, and 20% of the patient population overall). A single pathogen was isolated from 28 patients, but two or more pathogens were isolated from six patients, resulting in a total of 43 isolates from the 34 patients.
Table 2 shows the sputum sample culture results. Haemophilus influenzae was the commonest isolate and was ampicillin-sensitive in 89% of isolates. Moraxella catarrhalis was less frequently isolated, but was more commonly ampicillin-resistant. No penicillin-resistant pneumococci were isolated. Staphylococcus aureus was found in seven patients: four of these isolates were methicillin-resistant S. aureus. In many cases the organisms isolated were of uncertain clinical significance, such as where a scanty growth of S. aureus or Pseudomonas spp. was obtained.
|
Blood cultures were obtained in 28 patients but were positive in only one (S. pneumoniae)
Antibiotic therapy
At the time of admission, empirical antibiotic therapy was commenced in 151 of the 167 patients (90%), and a total of 156 patients (93%) received antibiotics at some stage during their admission. Most patients received treatment with a single antibiotic on admission, but by discharge, 41 had received three or more antibiotic agents (Figure 1). A wide range of antibiotics and antibiotic combinations was used during the patients' admissions (Table 3). Antibiotics were administered intravenously to 64 patients (38%) and orally to 87 (52%) at the time of admission. Co-amoxiclav was the most commonly prescribed antibiotic on admission, but other broad-spectrum antibiotic agents were also used, even for patients who had not received previous antibiotic treatment in the community (of the 101 patients who had not received prior antibiotic therapy, 44 were commenced on co-amoxiclav at the time of admission). Of those given macrolide antibiotics on admission, most (44/62, 70%) received them in combination with other antibiotic therapy.
|
|
The association between antibiotic therapy and clinical parameters on admission is shown in Table 4
. There was no evidence that dual antibiotics and intravenous administration were particularly targeted at vulnerable patient groups such as the elderly and immunocompromised. However, Table 4 shows that the admitting doctors were prescribing more aggressive therapy in patients with radiological changes at presentation or other markers of pulmonary sepsis such as purulent sputum, fever or leucocytosis. Thus 58% of patients with chest radiograph (CXR) consolidation received intravenous antibiotics on admission compared to only 26% without CXR changes (p<0.0001). Nevertheless overall level of antibiotic use was high in all subgroups, with 17/23 patients (74%) who had no evidence of pulmonary sepsis as judged by an absence of fever, purulent sputum, leucocytosis or inflammatory CXR changes being prescribed antibiotics.
|
Hospital-acquired diarrhoea
Six patients had diarrhoea at the time of admission. Of the 161 patients who were asymptomatic on admission, 17 (10%) developed diarrhoea whilst in hospital. Clostridium difficile infection was confirmed microbiologically in only four cases, but despite this relatively low rate of detection of C. difficile, there was a clear association between increased antibiotic use and hospital-acquired diarrhoea (Figure 2).
|
Logistic regression analysis was used to identify factors associated with the development of diarrhoea in hospital (the six patients who had diarrhoea on admission were excluded). In an initial univariate analysis, the following were statistically associated (p<0.05) with hospital-acquired diarrhoea: Increasing age, duration of in-patient stay, antibiotic therapy, number of antibiotics prescribed at time of admission, total number of antibiotics prescribed, and treatment with clarithromycin, ciprofloxacin, cefotaxime or trimethoprim. Diarrhoea had a tendency to be more common in patients treated with amoxycillin, cefuroxime, co-amoxiclav or erythromycin, but these associations failed to reach statistical significance. In a multivariate forward stepwise logistic regression analysis of the factors statistically associated with hospital acquired diarrhoea by univariate analysis, only the total number of antibiotics prescribed (p<0.0001) and increasing age (p=0.0062) were included in the final model.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
This audit was designed to determine if antibiotic prescription and choice was influenced by clinical, laboratory or radiographic indicators of infection in patients with a history of chronic airflow obstruction. The main burden of chronic airflow obstruction occurs in the over-50 age group and choosing this age cut-off also excluded many patients with simple asthma.
Despite the large and still increasing number of patients seen with exacerbations of chronic airflow limitation, the role of infection remains unclear. Two long-term studies in COPD5,6 have shown that viruses are important in 20–30% of exacerbations, and this may be an underestimate due to difficulties in virus detection. The tracheobronchial tree of patients with clinically stable COPD is often colonized with potentially pathogenic organisms, and cultures of bronchial brushings have shown no difference in bacterial numbers between these patients and those with disease exacerbation.7 Therefore, culture of an organism during an exacerbation of chronic airflow limitation does not necessarily imply that the organism is responsible for the exacerbation. However, a meta-analysis of nine randomized trials8 showed an overall positive (though small) effect of antibiotics in the treatment of COPD exacerbations. The results of a double-blind placebo-controlled trial by Anthonisen et al.9 suggest antibiotics are most useful in patients with at least two of: increased breathlessness; increased sputum volume; and increased sputum purulence. These findings are reflected in the current British Thoracic Society guidelines on the management of COPD.1
Despite these guidelines, we found that antibiotics were prescribed at some time during admission in 93% of patients. A surprising 44% received intravenous antibiotics at some time during their hospitalization. There was no evidence that dual antibiotics and intravenous administration were targeted at particularly vulnerable patient groups, such as the elderly or immunocompromised. However, those with purulent sputum, fever, leucocytosis or chest radiograph changes were more likely to receive parenteral antibiotics.
Sputum was sent for microscopy and culture in >50% of patients, but only in 16% prior to commencement of hospital antibiotic therapy, and positive sputum cultures were obtained in only 20% of our patient population. This audit did not examine whether the results of sputum culture altered prescribing behaviour or clinical outcome, and more work needs to be done in this area.
Antibiotic choice should depend upon the likely organisms and local resistance patterns. In the year this audit was undertaken, the local resistance rate of S. pneumoniae to penicillin was 11%, and of H. influenzae to erythromycin was 23%. As expected in this patient population, H. influenzae was the commonest isolate. In the UK and elsewhere, this organism is showing increasing resistance to aminopenicillins, mainly due to the production of beta lactamases.10,11 In this audit, the ampicillin resistance rate of H. influenzae was relatively low at 11%. Ampicillin resistance in the other common isolates was also low. Amoxycillin therefore remains a reasonable choice in patients in whom antibiotics are clinically indicated. Despite hospital-based protocols available at the time of this audit suggesting amoxycillin as first-line therapy in this clinical situation, the use of co-amoxiclav was striking. Fifty percent of patients were commenced on this at the time of admission, and 63% received co-amoxiclav at some stage. The widespread use of this drug can not be justified on the basis of failed antibiotic treatment prior to admission or the pattern of resistance in the sputum isolates.
Diarrhoea is a common side-effect of treatment with most antibiotics. However, it is worrying that the incidence of diarrhoea due to C. difficile has been increasing in many UK hospitals in recent years. This increase has been associated in some cases with the use of cephalosporins.12,13 Ten percent of our patients developed hospital-acquired diarrhoea, although only four had proven C. difficile. Diarrhoea was strongly associated with total number of antibiotics prescribed per patient and patient age.
Those who developed diarrhoea had more lengthy hospital admissions, but the median length of stay was not increased for the nine patients who developed diarrhoea within one week of admission. This suggests that prolonged admission was a risk factor for hospital-acquired diarrhoea, rather than vice versa.
The Department of Health recommends that all patients with chronic respiratory disease, including COPD and asthma, should have annual influenza and five-yearly pneumococcal vaccinations.3 It is well-proven that influenza vaccination reduces the severity and subsequently the mortality associated with influenza infection in elderly people.14 One study15 has shown significant reductions in hospitalization for pneumonia and of death in elderly patients with chronic lung disease who had received influenza vaccination. Similar reductions were seen in those receiving pneumococcal vaccination, and the combination of vaccines showed additive benefit. Our audit was undertaken over the winter months, when patients should have been vaccinated for the season. The rate of self-reported vaccination may be an underestimate of the actual rate due to inaccurate recall, but only one third reported receiving a recent influenza vaccination and 10% an up-to-date pneumococcal vaccination. There has been no improvement in uptake of influenza vaccination since our last audit;4 at this time, pneumococcal vaccination was not recommended for these patients.
This audit suggests that antibiotics are being overused in the treatment of exacerbations of chronic airflow limitation in hospitals, in terms of total numbers of antibiotic prescriptions, choice of antibiotic and route of administration. This has implications to the patient in terms of side-effects, to the hospital, in increased pharmacy costs, and for the future, in problems with antibiotic resistance. Protocols founded on the evidence-based British Thoracic Society guidelines, but tailored to local bacterial sensitivities, should be drawn up. All hospital doctors and primary-care physicians should be made aware of these protocols and compliance with them should be audited on a regular basis. Education of patients with chronic lung disease should include information about the value of influenza and pneumococcal vaccination. Simple letters of invitation for vaccination with follow-up letters to non-attenders may improve vaccination rates.16
|
Notes |
|---|
Address correspondence to Dr L. Davies, Aintree Chest Centre, University Hospital Aintree, Longmoor Lane, Liverpool L9 7AL. e-mail: LISA.DAVIES{at}aht.nwest.nhs.uk
*Present address: Public Health Laboratory, Withington Hospital, Manchester M20 2LR
**Present address: Monklands Hospital, Monkscourt Avenue, Airdrie ML6 0JS
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. British Thoracic Society guidelines for the management of COPD. The COPD guidelines group of the standards of care committee of the BTS. Thorax1997; 52(S5):S1–28.
2. Graham VAL, Milton AF, Knowles GK, Davies RJ. Routine antibiotics in hospital management of acute asthma. Lancet1982; i:418–20.
3. Department of Health. Immunisation Against Infectious Disease. London, HMSO, 1996.
4. Fox R, French N, Davies L et al. Influenza immunisation and viral respiratory tract infections in patients with chronic airflow limitation. Resp Med1995; 89:559–61.[Web of Science][Medline]
5. Smith CB, Golden CA, Kanner RE, Renzetti AD. Association of viral and mycoplasma pneumonia infections with acute respiratory illness in patients with chronic obstructive pulmonary disease. Am Rev Resp Dis1980; 121:225–32.[Web of Science][Medline]
6. Gump DW, Phillips CA, Forsyth BR, et al. Role of infection in chronic bronchitis. Am Rev Resp Dis1976; 113:465–74.[Web of Science][Medline]
7. Pela R, Marchesani F, Agostinelli C, Staccioli D, Cecarini L, Bassotti C, Sanguinetti CM. Airways microbial flora in COPD patients with stable conditions and during exacerbations: a bronchoscopic investigation. Monaldi Arch Chest Dis1998; 53:262–7.[Medline]
8.
Saint S, Bent JS, Vittinghoff E, Grady D. Antibiotics in chronic obstructive pulmonary disease exacerbations: a meta-analysis. JAMA1995; 273:957–60.
9. Anthonisen NR, Manfreda J, Warren CPW, et al. Antibiotic therapy in acute exacerbations of chronic obstructive pulmonary disease. Ann Intern Med1987; 106:196–204.
10. Thomson CJ. Antibiotic resistance: epidemiology of resistance—respiratory tract infections. J Med Microbiol1997; 46:442–5.[Medline]
11. Cruneberg RN, Felmingham D, Harding I, Shrimpton SB, Nathwani A. The Nearchus project: antibiotic susceptibility of respiratory pathogens and clinical outcome in lower respiratory tract infections at 27 centres in the UK. Int J Antimicrob Agents1998; 10:127–33.[Web of Science][Medline]
12. Zadik PM, Moore AP. Antimicrobial associations of an outbreak of diarrhoea due to Clostridium difficile. J Hosp Infect1998; 39:189–93.[Web of Science][Medline]
13.
Ludlam H, Brown N, Sule O, Redpath C, Coni N, Owen G. an antibiotic policy associated with reduced risk of Clostridium difficile-associated diarrhoea. Age Ageing1999; 28:578–80.
14. Kilbourne ED. Inactivated influenza vaccines. In: Plotkin SA, Mortimer EA, eds. Vaccines. Philadelphia, WB Saunders, 1988:420–34.
15. Nichol KL. The additive benefits of influenza and pneumococcal vaccines during influenza seasons among elderly persons with chronic lung disease. Vaccine1999; 17(S1):S91–3.
16. Igoe G, Bedford D, Howell F, Collins S. How to improve the uptake of influenza vaccination in older persons at risk. Ir J Med Sci1999; 168:107–8.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||