Respiratory Medicine (1992) 86, 503-505

Audit of the use of erythromycin in the treatment of community-acquired lower respiratory infections A. E. JOLLEY, A. J. DAVIES AND D. T. MCLEOD

Departments of Microbiology and Respirator), Medicine, Sandwell District General Hospital. West Midlands, U.K.

The British Thoracic Society (BTS) guidelines for the treatment of community-acquired pneumonia recommend initial therapy with a betalactam antibiotic, with the addition oferythromycin if there are features o f an atypical pneumonia. To see if these guidelines were being followed, a prospective study was undertaken of all adult patients admitted to hospital over a 3-month period who were given erythromycin for a community-acquired lower respiratory tract infection. Erythromycin was given to 62 patients who could be fully assessed. Continued prescription of erythromycin was justified in 10 ( 1 6 % ) - t w o patients with penicillin allergy, two with M. catarrhalis infection and one patient with legionnaires disease. Five patients had infections severe enough on admission to warrant combined therapy in line with the BTS recommendations. Five patients had erythromycin stopped on day 2. Erythromycin was prescribed on admission and continued unnecessarily in 47/62 patients, showing that the BTS recommendations are not being followed correctly.

Introduction Following a survey carried out by the British Thoracic Society (BTS) and the Public Health Laboratory Service (PHLS) about the aetiology, mortality and prognostic factors in adults with a community-acquired pneumonia (CAP), the BTS recommended that in the moderately or severely ill patient, treatment should be started to cover Streptococcus pneumoniae. When mycoplasma infection is suspected or in the case of a severe pneumonia, erythromycin or tetracycline should be added (I). Audits of antibiotic prescribing in our hospital had shown a four-fold increase in the amount of erythromycin prescribed from 1986 to 1990 (unpubl. res.). We decided to audit the use of erythromycin in patients with community-acquired lower respiratory tract infections at our District General Hospital to see how the BTS guidelines were being followed.

Methods All adults admitted to the hospital medical wards between December 1990 and March 1991 with clinically diagnosed community-acquired respiratory infections, and who were prescribed erythromycin on admission, were included in the audit. Patients were visited regularly on the ward. A diagnosis of either pneumonia or acute bronchitis was made in each patient on the basis of clinical and radiological findings. An attempt to Received 30 October 1991 and accepted in revised form 4 June 1992. 0954-6111/92/060503 + 03 $08.00/0

find a microbiological cause for infection was made by culturing blood and/or sputum in all patients on admission, and also by collecting paired sera to look for viral and atypical causes of infection including legionella and mycoplasma. Patients were assessed on admission, after 48 h in hospital (when laboratory results and the patients' clinical progress should make it clear whether an atypical pneumonia is present or not) and before discharge.

Results Sixty-six patients were recruited in all, but four were subsequently excluded as no serological studies were performed. Sixty-two patients were included in the final audit, 37 male and 25 female with a mean age 63.4 years (range 19-94). Forty-two patients were given intravenous erythromycin (mean 2-2 days, range 1-5) usually followed by oral erythromycin for a mean of 4-9 days, range 2-12; 20 patients were given oral erythromycin only for a mean of 5-3 days, range 2-12; and five patients were given erythromycin for 2 days only. Pneumonia was diagnosed in 42 patients on the basis ofradiological evidence and clinical signs. Chest radiographs were not obtained on admission in three cases. Blood cultures were collected in 37 (88%) cases, and sputum in 35 (83%). Bacterial pathogens were isolated in 21 (50%) cases, the most common isolate being S. pneumoniae, recovered from eight patients (six from © 1992 Baillidre Tindall

504

A. E. Jolley et al.

Table 1 Organisms identified

Pneumonia

Number (Total = 42)

Acute bronchitis (n = 20)

Strep Pneumoniae

8

1

H. influenzae M. catarrhalis Staph. aureus

3 0 3

4 2 0

'Coliforms' gp C B-haem. streptococcus

0

4

1

0

L. pneumophila

1

0

Influenzae B RSV

4 I

I 1

M. tuberculosis

Negative

0

1

21

6

blood culture), as shown in Table 1. In 21 patients (50%) no bacterial cause for their pneumonia could be found, but two patients had serological evidence of Influenza B infection. The three patients from whom Staph. aureus was isolated were felt to be colonized rather than infected with the organism. There were eight deaths, including three patients who had pneumococcal septicaemia. Two other patients with proven infection died despite appropriate antibiotics (one had Legionnaires disease and the other a beta-haemolytic streptococcal septicaemia). The three other patients who died had no pathogens isolated but were ill enough on admission to be given ampicillin plus erythromycin. Antibiotics had been prescribed before admission in nine of the 42 patients, and in five of these no bacterial pathogens were isolated. Acute bronchitis was diagnosed in 20 patients. Chest radiographs were not obtained on admission in seven cases. Blood cultures were collected in 14 (70%) cases, sputum in 19 (95%), and bacterial isolates were obtained in eight (40%) cases. The most common isolate was Haemophilus influenzae (four isolates). One patient had tuberculosis. No bacterial pathogens were isolated in 12 (60%) patients, but one had serological evidence of RSV infection. There were three deaths, two in patients with underlying malignancies and one in a patient who recovered from a chest infection but then died from an unrelated condition. Antibiotics had been previously prescribed in nine patients, and in five of these no bacterial pathogens were isolated. Erythromycin was prescribed alone in seven patients, two of whom had documented allergies to penicillin. Forty-five patients received a combination of ampicillin with erythromycin, and ten patients received a variety of other fl-lactam antibiotics in com-

bination with erythromycin. Continued prescription of erythromycin was necessary in ten patients in all; one with legionella infection, two with Moraxella catarrhalis infections and two with allergy to penicillins. Five other patients with pneumonia were seriously ill on admission, with three dying within 24 h. Three of these patients had no pathogens isolated, while two others had Influenza B infections. Erythromycin was prescribed and continued unnecessarily in 47 (75%) patients. Eighteen patients had been given antibiotics before admission. Those given ampiciilin or amoxycillin previously were given cefuroxime, those given other antibiotics [trimethoprim (n=4), ciprofloxacin (n=3), erythromycin (n=3), cotrimoxazole ( n = i ) ] were given ampicillin.

Discussion

This survey has shown that erythromycin was prescribed and continued unneccessarily in 75% of our patients, having increased in use four-fold over the previous 4 years. Only five patients had their erythromycin stopped at day 2 when laboratory results together with the patients' clinical progress should make it clear whether an atypical pneumonia is present. This overprescribing arises partly from a misunderstanding of the BTS recommendations and partly from the anxiety of junior medical staff to ensure they treat all possible causes of pneumonia. The standard of investigations sent and antibiotic prescribing was otherwise high in this study, making the over prescribing of erythromycin stand out. The majority of patients were initially given intravenous erythromycin making the practice expensive as well as exposing patients to antibiotics unneccessarily. There was only one patient with an atypical infection; he was admitted to the Intensive Care Unit with features of an atypical pneumonia which were not seen in the other patients. The BTS/PHLS survey of community-acquired pneumonia in 1982/3 showed that age, absence of chest pain, absence of vomiting, previous treatment with digoxin, tachypnoea, diastolic hypertension, confusion, leucopenia, leucocytosis and raised blood urea levels were significantly associated with death (1). Patients with some or all of these risk factors should be given a fl-lactam plus erythromycin, but less severely ill patients with pneumonia and all those with bronchitis should be treated initially with a fl-lactam such as ampicillin. Sputum culture will confirm the presence of fl-lactamase producing H. influenzae or M. catarrhalis within 48 h.

Erythromyc& & the treatment o f community-acquired lower respiratory infections With an expected increase in mycoplasma infections over the next 18 months, it has been suggested that erythromycin be used alone or in combination with penicillin as initial treatment for CAP (2). However, given the findings mentioned above and the fact that erythromycin-resistant pneumococcal bacteraemia has been reported (3-5), erythromycin should not be used alone in anyone presenting acutely to hospital with a CAP in case pneumococcal infection is inadequately treated. Penicillin-resistant pneumococci have been reported in many countries including Britain (6), so sensitivity testing should be performed in all cases of pneumococcal infection. In addition, erythromycin has limited activity against H. influenzae, the most common cause ofexacerbations of chronic bronchitis. It is right to consider the atypical agents as a possible diagnosis in CAP, especially during an epidemic, but initial treatment of lower respiratory tract infection should be with a fl-lactam antibiotic, with a

505

macrolide used only if the patient is moderately or severely ill with pneumonia. References

I. British Thoracic Society and the Public Health Laboratory Service. Community-acquired pneumonia in adults in British hospitals in 1982-1983: a survey of aetiology, mortality, prognostic factors and outcome. Quart J Med 1987; 62: 195-250. 2. Editorial. Mycoplasma pneumoniae. Lancet 1991; i: 651. 3. Eykyn SJ. Pneumococcaemia caused by erythromycin resistant Streptococcus pneumoniae Type 14. Lancet 1988; ii: 1086. 4. Verhaegen J, Goubau P, Verbit L, Glupczynski J, Blogie M, Yourassowsky E. Erythromycin resistant Streptococcus pneumoniae. Lancet 1988; ii: 1432-1433. 5. Warren RE, Haines D, Walpole E, Coles MAT. Erythromycin resistant Streptococcuspneurnoniae. Lancet 1988; ii: 1433. 6. Allen KD. Penicillin resistant pneumococci. J Hosp Infection 1991; 1: 3-14.

Audit of the use of erythromycin in the treatment of community-acquired lower respiratory infections.

The British Thoracic Society (BTS) guidelines for the treatment of community-acquired pneumonia recommend initial therapy with a betalactam antibiotic...
204KB Sizes 0 Downloads 0 Views