Epidemiology of methicillin-resistant Staphylococcus aureus in a Warsaw hospital

Journal

of Hospital

Infection

(1996)

Epidemiology Staphylococcus

34, 151-160

of methicillin-resistant aweus in a Warsaw

*, V. T. Rosdahl?,

G. Mlynarczyk

R. Skov-f

hospital

and A. Mlynarczyk”

“Department of Clinical Bacteriology, Medical Academy, Chalubiriskiego 5, 02-004 Warsaw, Poland, and +Staphylococcus Laboratory, Statens Sevuminstitut, Copenhagen, Denmark Received 23 February

1996; revised manuscvipt

accepted 22 March

1996

Summary: Methicillin-resistant Staphylococcus aureus (MRSA) isolates were collected during two eight-month periods in 1991 and 1994, respectively. In order to study the epidemiology, all 74 strains were characterized by phagetyping, antibiotic resistance pa%erns and DNA-restriction map aftercleavage with SmaI enzvme. and oulsed-field eel electrophoresis (PFGE). These investigations confirmed that MRSA in-the hospital, in 1991 and 1’994, was not due to the spread of one or two clones, but by the simultaneous occurrence of a few well characterized strains and sporadic, occurring strains of different phage-types. Some of these might have developed from the more commonly occurring strains. Isolates from 1994 were more resistant to antibiotics in vitro, than the 1991 isolates. The typing results also indicated that whilst most of the MRSA strains in 1994 were different comuared with those of 1991, some of the strains might have been present in both years. The PFGEtyping was more discriminatory and gave a higher typability than the phagetyping, especially among the multiply resistant isolates of MRSA from 1994. Among the less resistant strains the phage-typability was high and with only few exceptions, there was a good correlation between PFGE-type and phagetype. Keywords: MRSA; comial infections.

phage-typing;

pulsed-field

gel electrophoresis;

noso-

Introduction Methicillin-resistant Staphylococcus aureus (MRSA) has become an increasing problem in most of the world, except for Scandinavia.’ Their origin and spread has been extensively studied.2 In Poland approximately 50% of all S. auyeus isolates are MRSA. 3,4 Polish strains include two clusters: a heteroresistant cluster which is not multiply resistant, and a more homogeneous resistant population which is mainly multiply resistant.” A variety of epidemiological markers have been used for typing of S. auyeus, and some can be also applied to MRSA. Previously, the most

Correspondence 0195-6703/96/100151+10

to: Dr Graiyna

Mlynarczyk

$12.00/O

0 1996 The Hospital

151

Infectmn

Soaety

152

G. Mlrynarczyk

et al.

frequently used marker was bacteriophage-typing with the international set of phages,5 often in combination with experimental phages. For epidemiologic differentiation of MRSA, other phenotypic, as well as genotypic, analyses have been applied. These include the determination of antibiotic resistance patterns, biotyping, capsular polysaccharide serotyping, ribotyping, plasmid patterns determination, restriction analysis of chromosomal DNA by classical electrophoresis, pulsed-field gel electrophoresis (PFGE), genetic hybridization or polymerase chain reaction (PCR) mediated genome fingerprinting.6-‘1 PFGE is at present one of the most recommended methods. 12-15In our study we used classical methods: phagetyping and antibiotic resistance patterns, as well as results obtained after PFGE. The aim of the study was to determine, with both phenotypic and genotypic methods, if the MRSA in a Polish hospital was due to the spread of a single, or few, clones in the hospital and if different sporadic types also occurred. Methods

Bacterial isolates These were 74 MRSA isolates from patients at the clinical Hospital no. 1 of Medical Academy in Warsaw, a lOOO-bed hospital. The 74 MRSA represented one isolate from all patients infected with MRSA in the period May to December 1991 (35 strains) and April to October 1994 (39 strains). No strains from carriers, or the environment, were included in the analysis. The specimens originated from wounds, abscesses, pus, bones, blood and eyes, but not from throat, nose or skin. The specimens came from the following wards: intensive care, transplantation, surgery, orthopaedics, dermatology, endocrinology, gynaecology and neonatal. Detection of methicillin resistance Methicillin resistance was detected using the ATB OXA-test (BioMerieux), and was checked by the disc method on 7.5% NaCl Mueller-Hinton medium using Neosensitabs (Rosco Diagnostica, Taastrup, Denmark) and incubated at 30°C in air. Antimicrobial susceptibility testing The susceptibility to other antibiotics was assessed with Neosensitabs16 and the procedure and interpretation guidelines used were as recommended by the manufacturer. The following antibiotics were included: penicillin (P), methicillin (M), streptomycin (S), tetracycline (T), gentamicin (G), erythromycin (E), rifampicin (R), ciprofloxacin (C) and fusidic acid (F). Bacteriophage-typing Phage-typing of S. aweus was performed according to the method of Blair and Williams,’ using the present international set of typing phages. The

Typing

of MRSA

153

phages were employed in routine test dilution (RTD) and concentrations of 100 x RTD and 1000 x RTD. The subdivision into phage groups I, II and III was carried out according to Parker.” Furthermore, all strains of phage-type 95, and strains belonging to the 83A complex, the 52, 52A, 80, 81 complex and 94, 96 complex (group V) were recorded as separate patterns. Strains lysed by phages belonging to different lytic groups were recorded as mixed, while non-typable strains were designated NT. Two strains were regarded as belonging to different types if they differed by two or more strong reactions. PuviJication of staphylococcal DNA for PFGE A few colonies from an overnight blood agar culture were suspended in tryptic broth to obtain OD of 0.15 at 600 nm, and 5 mL of this suspension were centrifuged (5000 g for 10 mins). The pellet was suspended in 0.5 mL pH 8.0) and mixed with cold PIV buffer (1 M NaCl, 10 mM Tris-HCl, 0.5 mL of 1% low-melting point agarose (Sigma) in PIV buffer. The mixture was poured into the slots of a plastic insert mould and cooled for 10 min at 4°C. Solidified blocks were transferred to 3 mL fresh lytic solution (0.5% 10 units lysostaphin (Sigma), 0.05 pg laurylsarcosine, 100 pg lysozyme, RNAse (Boehringer Manheim) per mL of grand-buffer (6 mM Tris-HCl, 10 mM EDTA, 0.2 % deoxycholate, pH 8.0). After 2 h incubation at 37”C, blocks were placed in 3 mL of ESP buffer (0.5 M EDTA, 1% laurylosarcosine, 1 mg/mL proteinase K) and incubated overnight at 50°C. Blocks were washed in distilled water and incubated with 2 mL of TEbuffer (10 mM TRIS-HCL, 0.1 rnM EDTA) and 30 ~1 of 100 mM PMSF (phenylmethylsulphonylfluoride) for 30 min at room temperature, twice, followed by a 15 min wash in distilled water and 30 min incubation in TE buffer repeated twice. The blocks were transferred to a tube containing TE buffer and placed in a refrigerator until use. Digestion with SmaI Approximately l/10 of a block was transferred to the eppendorf tube with 250 FL of reaction-buffer of SmaI (50 mM KCl, 20 mM Tris-HCl, 5 mM MgC&, pH 7.4) and incubated for 1 h at room temperature. After removing of buffer, 50 ltL of the SmaI solution in reaction-buffer (10 units/ml) was added and samples were incubated overnight at room temperature. PFGE The blocks digested with SmaI were transferred into the slots of a 20-well 1% agarose (Life Technologies) gel that had been prepared in 0.5 TBEbuffer (5.4 g Tris-HCl, 2.75 g boric acid, 2 mL 0.5 mM EDTA, pH 8.0). The lambda-ladder standard (48.5-1618.5 kb; Bio-Labs) was used as a DNA marker. Electrophoresis was performed in 0.5 TBE-buffer at 200 V, for 23 h at 14°C on a CHEF-DR II system (BioRad Hercules, CA, USA). Initial A time was 5.0 s, final A time was 60.0 s, start ratio 1.0 and

G. Mlynarczyk

154

Table I. Phage-types

of MRSA

et al. isolated

in 1991 and 1994

No. of isolates in the year 1991

Phage-type

Abbreviation

5

81u 77/81/83A/84 42E/47/75/93 +u 6/42E/53/54/75/81/83A/84u 75177 + u 6U

6/53/54/75/8lu 83A 1000 x &4u 2 83A/93 93u 83A/93 29193~ 53/93u 1000 x 29/84/95u 29/77/81/83A/84/93u 96 95u 3c/55/71 NT Total

1994

i i

1: 1

1 :

i 0

7

1 5

:, i 1 0 :

1000 x

i 2 3 0 2 0 2 1

: :, :

35

6 39

L M NT

mode 0.01. Interpretation of the band-picture was performed according to Tenover et al.*’ Indistinguishable (0 bands difference) and closely related (two to three bands difference) patterns were placed in the same PFGEtype. Results

The 74 MRSA strains investigated were collected from infected patients at a large Polish University Hospital during two eight-month periods in 1991 and 1994, respectively. In order to study the epidemiology, all strains were characterized by phage-typing, antibiotic resistance patterns and DNA-restriction maps after cleavage with SmaI enzyme and PFGE. Phage- typing yes&s A total of 65 strains (88%) were typable, and could be subdivided into 13 different types, as shown in Table I and designated from A-M. Type H was further subdivided in a number of different subtypes (Hi-H,). Among typable strains, the ones belonging to group III were predominant (45 strains plus six strains of type 81) and 11 strains belonged to mixed group, I and III types. The general picture was as seen in the Table I than except

Typing

of MRSA

155

Table II. The occuYvence of MRSA according to DNA pattern in diffeerent wards. Comparison of DNA, phage and resistance patterns (othev than penicillin and methicillin resistance)

DNA pattern 1

Phagetype E E : NT NT

Resistance pattern

S, T S, T S, T S, T S, T S, T

G, E, R, G, E, R, G, E, R, G, E, R, G, E, R, G, E, R, S. T. E

F, F, F, F, F, F,

C C C C C C

S; T; E S, T E, F S. T. E. C S; T; E; C S, T, G, E T

S, T

T E, T

Hz or H4 K H6 A S, Streptomycin; C, Ciprofloxacin; Transplantation;

S, T G, E S, T G, E S, T G, E S, T G, E S, T G, E S, T, G, E, F, C S, T E

No

Ward

4 1 1 1 1 1

ICU Gyn Orth ICU Trans Derm Orth Derm Derm Trans Surg Surg Surg Neo Neo Neo Neo ICU Surg Orth Orth End Orth Surg

; ; 1 :. 1 7 1 2 1 : 2 1 ;

T, Tetracycline; G, Gentamicin; E, Erythromycin; R, Rifampicin; Orth, Orthopaedics; Derm, Dermatology; Surg, Surgery; Neo, ICU, Intensive care; Gyn, Gynaecology; End, Endocrynology.

Year

;: 94 94 94 94 91 91 91 z:: i;’ 91 91 5: 91 8: 94 94 94 91 F, Fusidic acid; Neonatal; Tram,,

for strains of type 84u, two different MRSA populations occurred in the two observation periods. In 1991 strains of type 81, 84u, 53/84u and mixed type dominated (74%) whereas strains of type 75/77 fu, 84u, and NT occurred more frequently (5 1%) in 1994. DNA-patterns Among the 74 strains, 37 different PFGE-patterns were observed. Six of the obtained PFGE patterns occurred among two or more isolates and as shown in Table II, 43 of the isolates belonged to these six patterns which are also shown in Figure 1. These six patterns were designated with numbers from l-6. Among strains belonging to PFGE-patterns 3, 5, and 6 all isolates of the same PFGE pattern belonged to the same phage-type (Hz and H3 are not significantly different). Among the nine strains of PFGE pattern 1, six belonged to phage-type E (75/77 +u), two were non-typable and one was also of group III but had more extended type (42E/47/75/ 93 +u). Among the strains of PFGE pattern 2, the majority belonged to

156

G. Mlynarczyk

et al.

Figure 1. Pulsed-field gel electrophoresis (PFGE) patterns of chromosomal DNA restriction fragments (SmaI) resolved in 1% Life Technologies agarose for six typical MRSA isolates from infected patients of Warsaw Hospital (pulse times, 5-60s; running time, 23 h). Lanes 2-7, DNA patterns from l-6, respectively; Lanes 1 and 8, /z ladder standard (Bio-Labs).

phage-type J (29/77/81/83A/84/93u 1000 x), but two strains demonstrated type H4 (f83A/93u) and the not significantly different H5 (93u), respectively. Four of the six strains of PFGE pattern 4 belonged to the H phage-types, whereas two strains were members of the other types of group III. The phage-type and the antibiotic resistance patterns of the remaining strains with non-identical PFGE-patterns are shown in Table III. Antibiotic resistance patterns The antibiotic resistance patterns of the strains with similar PFGE-pattern are shown in Table II, whereas the patterns of the remaining strains are shown in Table III. There were 13 strains, resistant to all antibiotics investigated except vancomycin, nine of PFGE-pattern 1 and four phagenon-typable strains of different PFGE-patterns, all from 1994 (Tables II and III). Seven strains resistant to PMSTGEC and, sometimes resistant to fusidic acid, occurred among two PFGE-pattern 5 strains and five strains

Typing Table

III.

The resistance

Resistance pattern

S, T G, E, R, C, F S, T G, E, C, F S, T, G, E, C S, T, G, E

of MRSA

patterns (more than penicillin and methicillin) MRSA strains with unique PFGE-types Phagetype

No. of isolates

NT NT XT E

3 1

H*

JJT

HI G M A

1 1 1

T E, R S, T, E, R S, T E, C S, T, E

1

T

1

H3 Hz T F S, Streptomycin; C, Ciprofloxacin; Transplantation.

157

K

H2

1

1 1 1

Ward

and phage-types

01

Year

Orth Trans Orth Orth Orth Orth Orth Surg Surg Surg Trans Derm Orth Surg Surg Surg Surg Derm Derm Derm Surg Orth Surg Derm Neo Trans Orth Surg

T, Tetracycline; G, Gentamicin; E, Erythromycin; R, Rifampicin; F, Fusidic acid; Orth, Orthopaedics; Derm, Dermatology; Surg, Surgery; Neo, Neonatal; Trans,

with unique PFGE-patterns. They represented phage-types H and one strain was phage-type E and two were non-typable. Five strains were from 1994 and two from 1991. Fourteen strains were resistant to PMSTGE and occurred among six PFGE-pattern 4 strains, one of PFGE-pattern 2, and seven with unique types. Ten of them were from 1994. Among remaining 40 less resistant strains the majority (29) occurred in 1991. Ten belonged to PFGE-pattern 2, thirteen to PFGE-pattern 3, two to PFGE-pattern 6 and the remaining to unique types. Strains of PFGE-pattern 2 are all resistant to PMSTE and three have acquired resistance to ciprofloxacin, two resistance to fusidic acid and one resistance to gentamicin. Epidemiology Among the 35 strains from 1991, 11 originated from the newborn ward and all of them belonged to PFGE-pattern 3 and were resistant to few but

158

G. Mlynarczyk

et al.

variable antibiotics. The first three strains from the neonatal ward were isolated in May, followed by six isolates in August/September; after the ward was closed and cleaned, however, two new cases occurred in December together with two cases in surgery. In 1991 strains of PFGE-pattern 2 spread with four cases in dermatology and three in orthopaedic nearly simultaneously in August and November. No relationship could be established between the PFGE-type 2 strains isolated in 1991 and the four strains isolated in 1994; three of them had acquired ciprofloxacin resistance and one gentamicin resistance. Strains of PFGE-type 5 and 6 both occurred with only two cases each and with only one ward involved for each type. The strain of PFGE-type 1 was first isolated in April 1994 from a patient at the gynaecology ward. The patient was then transferred to intensive, where the strain spread to five other patients, as well as to patients in dermatology, orthopaedics and general surgery. Discussion

The present investigations clearly confirm the occurrence among Polish one with resistance to PM and a few other MRSA of two populations,i4 antibiotics such as tetracycline, streptomycin or erythromycin and another population with in vitro resistance to gentamicin and often ciprofloxacin and rifampicin. The MRSA from 1994 were generally more antibiotic resistant in vitro than the isolates from 1991. Resistance to ciprofloxacin had increased from 5.7% in 1991 to 58.9% in 1994 probably due to increased ciprofloxacin usage, which in turn has played an important role in the selection of the more resistant strains, as shown by others.” In the outbreaks with strains of PFGE-patterns 1, 4, 5 and 6, the antibiotic resistance pattern was constant. It is, however, worth noticing that among strains with the same PFGE-pattern, variation in antibiotic resistance did occur. In the neonatal outbreak with strains of PFGE-pattern 3, the majority of the strains were resistant to PMT, but two were PM, one PME and one PMST resistant. Amongst the strains of the PFGEpattern 2 the majority was resistant to PMSTE, but resistance to ciprofloxacin, gentamicin and fusidic acid also occurred. The usage of antibiotic resistance pattern as an epidemiological typing method should therefore be performed with great care and should be confirmed with other typing methods. Phage-typing of MRSA with the international set of routine phages has in some but not all areas given a low typability. In this study, non-typable strains occurred, especially among the strains isolated in 1994 resistant to PMSTGERCF. Among the less resistant strains, the typability was high and there was a good correlation between PFGE-pattern and phage-type, except that with strains of PFGE-pattern 2 and 4, some changes did occur over time, probably due to loss or gain of prophages. The PFGE typing was, however, more discriminatory than the phage-typing among the present

Typing

of MRSA

159

set of MRSA. Addition of experimental phages, especially designed for MRSA, might add to the value of phage-typing. It is, however, questionable as shown in a recent study by Noble and Howel12’ whether phage-typing or DNA typing with PFGE should be regarded as the golden standard when discrepant results are obtained. The present investigations confirm that occurrence of MRSA in the hospital both in 1991 and 1994, was not a spread of one or two clones; but the occurrence of a few well characterized strains (types) which spread together with quite a number of sporadic occurring strains of different types, of which some might have been developed from the more commonly occurring ones. It is quite evident from the typing results, that all MRSA in the two periods did not have just one or two origins, and that typing is an essential tool in order to be able to obtain epidemiological data useful for a relevant eradication programme. The typing results also indicate that although the MRSA populations were mainly different in 1994, compared with 1991, some of the strains might have been present in the hospital during both observation periods such as strains of the PFGE-pattern 2 and type 4. References 1. Voss A, Milatovic resistant

D, Wallrauch-Schwarz Staphylococcus aweus in Europe.

C, Rosdahl VT, Braveny I. MethicillinEur J Clin Microbial Infect Dis 1994; 13:

50-55. 2. Kreiswirth B, Kornblum J, Arbeit R, Eisner W, Maslow JN, McGeer A, Low DE, Novick RP. Evidence of clonal origin of methicillin-resistance in Staphylococcus auveus. Science 1993; 259: 227-230. 3. Hryniewicz W, Trzcinski K, Tyski S, Zareba T, Jeljaszewicz J. Antimicrobial susceptibility patterns of some important pathogens in Poland 1991-1994. ZbZ B&t 1994;

281: 263. 4. Trzcinski K, Hryniewicz W, Claus H, Witte W. Characterization of two different clusters of clonally related methicillin-resistant Staphylococcus auyeus strains by conventional and molecular typing. J Hosp Infect 1994; 28: 116-123. r Blair JE, Williams REO. Phage typing of staphylococci. Bull WHO 1961; 24: 771-784. ;: Branger C, Goullet P. Genetic heterogenity in methicillin-resistant strains of Staphylococcus auveus revealed by esterase electrophoretic polymorphism. J Hasp Infect 1993; 14: 125-134. 7. Costas M, Cookson BD, Talsinia HG, Owen RJ. Numerical analysis of electrophoresis protein patterns of methicillin resistant strains of Staphylococcus aweus.J Clin Microbial 1989; 27: 2574-2581. 8. Gaston MA, Duff PS, Naidoo J, Ellis K, Roberts JIS, Richardson JF, Marples RR, Cooke EM. Evaluation of electrophoretic methods for typing methicillin-resistant Staphylococcus aweus. J Med Micvobiol 1988; 26: 1 X9-1 97. 9. Struelens MJ, Bax R, Deplano A, Quint WG, Van Belkum A. Concordant clonal delineation of methicillin-resistant Staphylococcus auyeus by macrorestriction analysis and polymerase chain reaction genome fingerprinting. J Clin Microbial 1993; 31: 1064-1970. 10. Trilla A, Nettleman MD, Hollis RJ, Fredrickson M, Wenzel RP, Pfaller MA. Restriction endonuclease analysis of plasmid DNA f rom methicillin-resistant Staphylococcus aweus: clinical application over a three-year period. Infect Control Hasp Epidemiol 1993; 14: 29-35. 11. Van Belkum A, Bax R, Peerbooms P, Goessens WH, Van Leeuwen N, Quint WG. Comparison of phage-typing and DNA fingerprinting by polymerase chain reaction for

160

G. Mlynarczyk discrimination

of methicillin-resistant

et al.

Staphylococcus

aureus strains.

J Clin

Microbial

1993; 31: 798-803.

JF, Reith S. Characterization of a strain of methicillin-resistant Staphylococcus uureus (EMRSA-15) by conventional and molecular methods. J Hosp Infect 1993; 25: 45-48. gel electrophoresis 13. Bannerman TL, Hancock GA, Tenover FC, Miller MJ. Pulsed-field as a replacement for bacteriophage typing of Staphylococcus auveus. J Clin Microbial

12. Richardson

1995; 33: 551-555. 14. Saulnier P, Bourneix 15.

16. 17. 18.

C, Prevost G, Andremont A. Random amplified polymorphic DNA assay is less discriminant than pulsed-field gel electrophoresis for tvping strains of methicillin-resistant Staphylococcui aureus. ‘j C%n Microbial 1993; 31: 982-985. Schlichtine C. Branger C. Fournier TM. Witte W. Boutonnier A. Wolz C. Goullet aweus by pulsed-field gel electrophoresis, P, Doring-G.’ Typing of ‘Staphylocokus’ zymotyping, capsular typing, and phage-typing: resolution of clonal relationships. r Clin Microbial 1993; 31: 227-232. Casals JB, Petersen OG. Tablet sensitivity testing: a comparison of different methods. APMIS Sect B 1972; 80: 806-816. Parker MT. The significance of phage-typing patterns in Staphylococcus aureus. In: Eastmon CSF, Adam C, Eds. Staphylococci and Staphylococcal Infections. Vol. 1, London: Academic Press. 1983; 33-62. Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BE, Persing DH, Swaminathan B. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. ._ I- 7 Clin Microbial

i995; 33: 22f3-2239. 19. Witte W, Grimm H. Occurrence

nosocomial infection. Eoidemiol 20. Noble WC, Howell SA. Labile Infect 1995; 31: 135-141.

of quinolone resistance in Staphylococcus aweus from Infect 1992: 109: 413-421. antibiotic resistance in Staphylococcus auveus. J Hasp