Community-Associated Methicillin-Resistant Staphylococcus aureus Isolated in Finland in 2004 to 2006

A nationwide population-based study on community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) in Finland during 2004 to 2006 showed that both incidence (1.9/100,000 population) and strain variation increased in comparison to years 1997 to 1999. There were 7 community-associated epidemic and 25 sporadic MRSA strain types. Half of these had Panton-Valentine leukocidin genes.Few population-based estimates of the burden of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) have previously been published (1-3, 12, 14). In Finland, the incidence of MRSA has been low. During 1997 to 1999, one-fifth of the 526 Finnish MRSA isolations were from persons without any connection to a hospital, and three strain types were associated with community acquisition (14). However, the total number of MRSA findings reported to the National Infectious Disease Register in Finland in 2004 to 2006 increased eightfold since 1997 to 1999. This increase has mostly been due to outbreaks and active screening in hospitals and hospital-associated strains (6, 15).In this study, we estimated the proportion of CA-MRSA strains among the MRSA isolates obtained either by screening or from a clinical specimen and determined the incidence and type of clinical CA-MRSA infections during 2004 to 2006. The MRSA strain types obtained from persons with and without hospitalization were compared.In Finland (population, 5.3 million), all clinical microbiology laboratories notify MRSA findings, including only the first isolate from a person, to the National Infectious Disease Register and send the corresponding isolates to the reference laboratory. During 2004 to 2006, a total of 4,054 (97%) isolates from 4,166 newly detected MRSA-positive persons were received.For each MRSA isolate, pulsed-field gel electrophoresis (PFGE) and antimicrobial drug susceptibility testing were performed (6, 10, 15). Multilocus sequence type (MLST), spa type, and SCCmec determinations were performed for isolates with a PFGE type shared by five or more persons (15). Panton-Valentine leukocidin (PVL) genes (lukS-PV, lukF-PV) were tested for by PCR in all CA-MRSA isolates. Identical or closely related PFGE types, MLSTs, and spa types (4) defined a strain type designated by a FIN number. A sporadic strain was shared by five or fewer persons.For all 4,030 persons with an MRSA isolate from 2004 to 2006, excluding 24 persons with an erroneous national identity code, data on previous hospitalizations were retrieved from the National Hospital Discharge Register. Background information was obtained for each person with MRSA isolated outside a hospital setting or within 2 days of hospital admission and who had not been hospitalized within 2 years of a positive MRSA culture by sending a questionnaire to infection control nurses at the relevant health care facilities. These data included risk factors for MRSA acquisition, the reason for obtaining the culture (screening or infection), and the type and treatment of a possible infection as recorded in medical charts by the patient''s primary health care provider. The Ministry of Social Affairs and Health, the Finnish data protection authority, and the National Research and Development Center for Welfare and Health approved the use of data from the National Hospital Discharge Register.Community association was calculated for each MRSA strain type carried by at least 10 persons. The chi-square test with Yates correction or Fisher''s exact test, as appropriate, was used for categorical variables. The means and medians of the continuous variables were compared by Student''s t test or the Mann-Whitney U test, depending on the sample distribution.A total of 570 persons without previous hospitalization in the last 2 years were identified. Based on the survey, 94 health care workers, 158 long-term care facility residents, and 20 newborns less than 28 days old were excluded. Thus, 298 (7.4%) of the 4,030 MRSA-positive persons and isolates were community associated (range by year, 5.8 to 8.9%); 185 (62%) and 113 (38%) were obtained as clinical specimens and by screening, respectively. Ultimately, 191 (64%) had a clinical infection. The mean annualized incidences of all CA-MRSA findings and CA-MRSA infections, respectively, were 1.9 and 1.2/100,000 population (ranges of annual incidences, 1.6 to 2.1 and 1.0 to 1.4). Persons with a CA-MRSA isolate were younger (median, 37 versus 75 years, P < 0.001) and more often males (56 versus 49%, P < 0.05) than those with previous hospitalization.Among the 4,030 MRSA isolates, our typing scheme showed a total of 148 strain types, of which 109 (74%) were sporadic and 39 (26%) were shared by at least five persons. Twenty-five strain types, each of which was isolated from ≥10 persons, represented 3,971 (99%) of the 4,030 isolates. Seven of these strain types, including two of the three old CA-MRSA strain types and 25 sporadic strain types, were associated with community acquisition (Table ​(Table11).

TABLE 1.

MRSA strain types significantly more often isolated in persons without previous hospitalization, i.e., CA-MRSA strain types, in Finland in 2004 to 2006

Strain type(s)	SCCmeca	MLST	spa type	Antimicrobial resistanceb	PVLc	No. (%) of isolates from persons:
						Without hospitalization (n = 298)	With hospitalization (n = 3,732)
FIN-4	IV	375	t172	OXA	−	56 (19)	379 (10)
FIN-11	IV	80	t044	OXA (ERY, TET, CLIN, CIP)	+	42 (14)	21 (1)
FIN-12	IV	22	t022	OXA, ERY, CIP	−	18 (6)	65 (2)
FIN-25	IV	8	t008	OXA, ERY, CIP	+	19 (6)	14 (0.3)
FIN-5, -5b	IV	30	t018/019	OXA, GEN, TOB, ERY	±	10 (3)	22 (1)
FIN-19	IV	1	t127	OXA	±	9 (3)	13 (0.3)
FIN-30	V	8	t754	OXA, TRI-SUL (ERY, IND CLIN)	−	7 (2)	8 (0.2)
Sporadic	IV, V, NT				±	25 (8)	84 (2)

Open in a separate window^aSCCmec, staphylococcal cassette chromosome mec.^bERY, erythromycin resistance; TET, tetracycline resistance; IND CLIN, inducible clindamycin resistance; CIP, ciprofloxacin resistance; GEN, gentamicin resistance; TOB, tobramycin resistance; TRI-SUL, trimethoprim-sulfamethoxazole resistance. The antimicrobial resistance in parentheses occurs in some isolates of the corresponding strain type.^c+, positive; −, negative; ±, some positive and some negative isolates.Ninety (30%) of the 298 CA-MRSA isolates were PVL positive (range by year, 23 to 34%), including 4 of the epidemic CA-MRSA strain types, as well as 11 sporadic types. During 1997 to 1999, only one PVL-positive strain type (FIN-11, ST80:IV, t044) was community associated (5). In the present study, a USA300 variant (FIN-25, ST8;IV, t008) which first appeared in 2004 was the third most common CA-MRSA strain type. Since sporadic CA-MRSA also harbored PVL genes, these strains may have been generated de novo in the community, as suggested earlier (9, 13).Patients with PVL-positive isolates were more likely to have clinical infections than patients with PVL-negative isolates (82/90 [90%] versus 109/208 [52%]; P < 0.01) (Table ​(Table2).2). The clinical picture of CA-MRSA infections was similar to those reported from other countries (1, 3, 12). However, pneumonias were rare and no bacteremias occurred. As anticipated by the findings of the few existing population-based comparative studies (3, 7, 11), PVL-positive clones were associated with abscesses, a need for surgery, and systemic antimicrobial treatment.

TABLE 2.

Clinical characteristics of persons with PVL-positive and PVL-negative CA-MRSA infections

Category	No. of persons with indicate infection type/total no. of persons (%)		P value
	PVL+ (n = 82)	PVL− (n = 109)
Skin and soft tissue infections	72/82 (88)	84/109 (77)	0.087
Impetigo	7/72 (10)	6/84 (7)	0.771
Abscess	54/72 (75)	11/84 (14)	<0.001
Erysipelas	3/72 (4)	6/84 (7)	0.428
Nail bed	3/72 (4)	7/84 (8)	0.291
Wound	3/72 (4)	38/84 (45)	<0.001
Other	2/72 (3)	16/84 (19)	0.002
Bursitis or arthritis	5/82 (6)	6/109 (6)	0.862
Eye	2/82 (2)	6/109 (6)	0.294
Ear	2/82 (2)	5/109 (5)	0.435
Throat	0/82 (0)	5/109 (5)	0.500
Urinary tract	1/82 (1)	4/109 (4)	0.295
Pneumonia	0/82 (0)	1/109 (1)	0.392
Chronic/relapsing skin and soft tissue infections	19/71 (27)	36/80 (45)	0.311
Surgery for skin and soft tissue infections	41/72 (57)	27/84 (32)	0.003
Hospitalization	11/82 (13)	19/109 (17)	0.579
Systemic antimicrobial treatment	74/82 (90)	76/109 (70)	0.001

Open in a separate windowFamily members had a similar type of skin or soft tissue infection more often if the person had a PVL-positive rather than a PVL-negative strain (24/72 [33%] versus 11/84 [13%]; P < 0.01). The person or his/her family member was an immigrant more often if the strain was PVL positive (19/90 [21%] versus 21/208 [10%]; P < 0.05).Our nationwide population-based study covering virtually all (97%) of the MRSA isolates in Finland showed that between the two 3-year periods, 1997 to 1999 and 2004 to 2006, the number of CA-MRSA isolations tripled, from around 100 to nearly 300. However, the proportion of CA-MRSA decreased from 21 to 7%. Moreover, the diversity of CA-MRSA strain types increased compared with our previous findings.During 2004 to 2006, the average annualized incidence rate of CA-MRSA infections (1.2/100,000 person years) in Finland was far less than that reported in, for instance, the United States (25.7 and 18.0/100,000 population in 2001 and 2002 in Atlanta and Baltimore, respectively) (3) or Australia (391/100,000 population) (3, 12). However, CA-MRSA data are largely dependent on how CA-MRSA is defined. We used a 2-year rather than a 1-year time limit since any previous hospitalization (8).Our study have some other limitations. We included both clinical MRSA isolates and those obtained by screening in the analysis, which may affect the proportions of hospital-associated and CA-MRSA strains. However, this approach allowed comparison to our previous study (2). In addition, we did not interview the patients for health care-associated risk factors, symptoms, recurrences, surgery, later hospitalizations, or treatment. The patient charts were reviewed by the infection control nurses only around the time and place when MRSA was first isolated.