Journal

of Hospital

Infection

Investigation endocarditis

(1991)

19, 107-l

14

of nosocomial prosthetic due to antibiotic-resistant

Staphylococcus R. Menzies, Microbiology

epidermidis

D. MacCulloch

Department,

and

Green Lane Hospital, 3, New Zealand

Accepted

valve

for publication

B. Cornere Greenlane

26 j%ly

West,

Auckland

1991

Summary: A reservoir of antibiotic-resistant Staphylococcus epidermidis strains in our cardiac surgery unit appeared to be the source of organisms responsible for three cases of early prosthetic valve endocarditis. Staphylococcus epidermidis isolates recovered from the skin of 13 patients before and after surgery were compared. All were typed by plasmid profile, antimicrobial susceptibility and slime production. The three strains from early prosthetic valve endocarditis resembled the antibiotic-resistant nosocomial strains recovered from the skin of eight patients following surgery and the environment of the operating theatres. These strains expressed resistance to oxacillin, gentamicin, kanamycin and tobramycin and most produced slime, whereas those isolated from the skin of patients at the time of admission were predominantly susceptible to antibiotics and few produced slime. Keywords: Nosocomial cus epidermidis; slime

infection; production.

prosthetic

valve

endocarditis;

Stuphylococ-

Introduction Staphylococcus epidermidis is an important cause of prosthetic heart valve infection.lm3 It is ubiquitous on human skin which is the apparent source of early postoperative infections. l-3 Some evidence indicates that strains of S. epidermidis associated with early prosthetic valve endocarditis may be acquired during hospitalization and that these strains are different from those colonizing patients’ skin at the time of admission to hospital.3B4 Strains present at the time of admission are usually susceptible to antibiotics whereas those cultured from hospital patients following surgery are frequently multiply-resistant.3s4 We investigated three cases of prosthetic valve endocarditis due to multiply antibiotic-resistant S. epidermidis. Each infection occurred within 2 months of surgery at intervals of c. 6 months. Because antimicrobial Correspondence West, Auckland 0195-670t/91/100107+08

to: Rosalie Menzies, 3, New Zealand.

Microbiology

Department,

SOS.OO,O

Green

Lane

Hospital,

0 1991 The Hospital

107

Greenlane

Infection

Society

,

108

R. Menzies

et al.

susceptibility and plasmid profiles indicated that each isolate was a different strain, and a different surgical team was responsible for each heart valve replacement, it seemed unlikely that a common source of infection such as a carrier was involved. Our aim was to determine whether the source of these strains was the patients’ own flora or a reservoir of organisms in the cardiac surgery unit. The method of investigation was a survey of S. epidermidis strains cultured from skin swabs taken from 13 patients on admission to hospital and 7 days after cardiac by-pass surgery. At the same time, the operating theatre environment was sampled for the presence of antibiotic-resistant S. epidermidis during surgical procedures. We felt that a wider investigation could not be justified in the absence of an outbreak of infection. Materials

and

methods

Patients for cardiac surgery were admitted to the general ward of the cardiac surgery unit no earlier than the day before surgery. After surgery they were transferred to an Intensive Care Unit for at least 24 h before transfer to a special care room in the general ward. At discharge they were resident in the area of the general ward which acted as a reception area for incoming patients. An average of 910 cardiac by-pass operations are carried out each year and 220 of these are heart valve replacements, with prosthetic valves. Since 1977 patients for cardiac by-pass surgery have received antibiotic prophylaxis with cephradine (E. R. Squibb & Sons), 1 g iv. on anaesthetic induction, 1 g i.v. at the start of by-pass surgery and 1 g i.v. on return to the Intensive Care Unit. This was followed by 500 mg q 6 h p.o. or i.v. for 4 days. The night before surgery patients were shaved and had a total body wash with 4% chlorhexidine gluconate. Preoperatively the body wash was repeated and 2.5% tincture of iodine was applied to the incision site. In theatre the iodine tincture application was repeated before surgery, and following surgery the wound was covered with a sterile dry dressing. Between 1987 and 1989 three patients developed early prosthetic valve endocarditis. All patients recovered; two after replacement of the infected prosthetic material and antibiotic treatment, and the third after treatment with antibiotics only. Their details are summarized in Table I.

Patient skin survey The anterior nares and skin over the sternum of 13 patients admitted to hospital between 18 January 1989 and 29 May 1989 were swabbed on admission to hospital and 7 days after cardiac by-pass surgery; heart valve replacement (six patients), coronary artery grafting (six patients) and repair of aneurysm (one patient). The first patient available each week was swabbed. All patients were from the elective surgery waiting list, 18 to 65 years of age and had not received antibiotics or been in hospital within the preceding 4 weeks. The nose was swabbed with a dry swab and the skin over

P.V.E.

due

the sternum with a swab moistened Culture swab transport system).

109

to S. epidermidis

with

Amies

transport

medium

(Difco

Operating theatres During the patient survey the two operating theatres used for cardiac surgery were sampled for the presence of antibiotic-resistant S. epidermidis strains during surgical procedures. Twenty agar plates, Columbia agar (Gibco) with 5% sheep blood, were placed to detect staphylococci in areas of staff movement. Each plate remained in position for the duration of one surgical procedure, average time 3 h.

Culture methods Each swab of the anterior nares and skin over the sternum was cultured on Columbia agar with 5% sheep blood and mannitol salt agar (Difco), and incubated aerobically at 35°C for 48 h. Staphylococcal colonies with representative morphology from each swab were identified and up to 20 presumptive S. epidermidis isolates were typed by susceptibility to 16 antimicrobial agents, plasmid profile and slime production. Agar plates from operating theatres were incubated at 35°C for 48 h and four staphylococcal colonies from each were identified. All presumptive S. epidermidis isolates were tested for susceptibility to oxacillin, and oxacillin-resistant isolates were typed by susceptibility to 16 antimicrobial agents, plasmid profile and slime production.

Identification

of isolates and strain differentiation

Presumptive identification as S. epidermidis was made by Gram’s stain, catalase production and failure to produce acid from trehalose and mannitol.5,6 Identification of probable nosocomial strains and isolates from prosthetic valve endocarditis as S. epidermidis was confirmed by API Staph (France). Strains were typed by antimicrobial susceptibility pattern, plasmid profile and slime production. 2,3,7 Susceptibility to 16 antimicrobial agents was determined by the disc diffusion method on Mueller-Hinton agar (Gibco) as recommended by the National Committee * for Clinical Laboratory Standards.’ Antibiotics tested were penicillin, oxacillin, gentamicin, kanamycin, tobramycin, amikacin, streptomycin, erythromycin, clindamycin, chloramphenicol, tetracycline, rifampicin, vancomycin, trimethoprim/sulphamethoxazole, neomycin and fusidic acid. Slime production was detected by a tube method.’ Plasmid DNA was isolated and separated by a rapid method adapted for S. epidermidis.‘O Isolates were grown on Columbia agar with 5% sheep blood instead of in L broth and bacterial cells were suspended in 10 ml of sterile distilled water to an opacity approximately equal to a number 5 McFarland standard. Lysostaphin (Sigma) was used at double the concentration suggested for

Staphylococcus aweus.

110

R. Menzies

et al.

Results

Characteristics of strains from prosthetic valve endocarditis These strains all had different plasmid and antimicrobial susceptibility profiles. Antimicrobial resistances are shown in Table I. Resistance to penicillin, oxacillin, gentamicin, kanamycin and tobramycin and slime production were common to all strains. Distribution of strains from patient skin swabs and operating theatres Antimicrobial susceptibility, slime production and plasmid profile separated the 215 S. epidermidis isolates obtained from skin swabs into 92 different strains; 49 strains were recovered on admission only, 39 after surgery only and four both on admission and after surgery. The number of strains isolated from each patient at any one time varied from one to eight (average four). Colonization of the anterior nares and skin over the sternum by the same strain was more prevalent after surgery (eight patients) than before surgery (two patients). Identical S. epidermidis strains from postoperative skin swabs, not found on admission swabs, from more than one patient or from one patient and an operating theatre were considered to be nosocomial in origin. Four nosocomial strains were isolated from postoperative swabs from eight

Table

Sex

I. Summary

Age (years)

of the clinical

history and microbiological prosthetic valve endocarditis

Operation

Postoperative

findings

infection

Onset

Treatment

51

21.9.87 Mitral valve replaced with prosthetic valve

2 months

Vancomycin + rifampicin Replacement of infected valve

Male

50

8 days

Male

27

1.7.88 Aortic valve + ascending aorta replaced with prosthetic valve + dacron graft 4.4.89 Porcine mitral valve replaced with prosthetic valve

Vancomycin + gentamicin + rifampicin Replacement of infected valve + graft Vancomycin + rifampicin + flucloxacillin SClindamycin + rifampicin + fusidic acid Discharged on ciprofloxacin

* A, Amikacin; C, chloramphenicol; Cc, clindamycin; oxacillin; P, penicillin; Sxt, trimethoprim/sulphamethoxazole; $ Changed to this regimen because of neutropenia.

G, gentamicin; To,

the three

patients

with

Microbiological results

Male

7 days

for

S. epidermidis from blood cultures + prosthetic valve. Resistant to P,Ox,G,K,To,N,Sxt* S. epidermidis from blood cultures + prosthetic valve + graft. Resistant to P,Ox,G,K,To,Cc,C,Sxt* S. epidermidis from blood cultures. Resistant to P,Ox,G,K,To,A,N,C,Sxt*

K, kanamycin; tobramycin.

N, neomycin;

Ox,

P.V.E. due

to

111

S. epidermidis

patients and one operating theatre. One strain was isolated from three patients and two strains each from one pair of patients. None of these patients were resident in the same area of the hospital at the same time. Operating theatres yielded one strain which was identical to an isolate recovered from a patient discharged from hospital 3 months previously. Antimicrobial susceptibility of strains from patient survey swabs and operating theatres Antimicrobial susceptibility tests of S. epidermidis strains isolated on admission and after surgery showed a very highly significant difference in antimicrobial susceptibility (Table II), P

Investigation of nosocomial prosthetic valve endocarditis due to antibiotic-resistant Staphylococcus epidermidis.

A reservoir of antibiotic-resistant Staphylococcus epidermidis strains in our cardiac surgery unit appeared to be the source of organisms responsible ...
548KB Sizes 0 Downloads 0 Views