Case Finding

Case-players were defined as team A members with MRSA culture-confirmed SSTIs or SSTIs presumably caused by the USA300 strain in the outbreak period August 5 to September 5, 2003. Because we suspected that disease exposure occurred during camp, we chose the study period from the start of training camp to ≈2 weeks after the end. Our experience with other SSTI outbreaks found that in most persons lesions develop within 2 weeks postexposure to CA-MRSA. To find case-players, we reviewed the trainer’s treatment log to identify players with skin lesions who required medical or surgical interventions. We asked the staff to conduct skin inspections of all players. Players were encouraged to report any skin lesion. In addition, we queried the student health center to determine if these infections were prevalent on campus.

Nasal Carriage Study

As soon as the current outbreak was recognized, a returning player (player X) was suspected to be the source of infection. Player X had 1 of the 2 cases of CA-MRSA SSTIs discovered in 2002. His locker was directly across from the index case-player, and he was a roommate, during camp, of another case-player. Trainers obtained a nasal culture from player X on August 25. On September 3, trainers obtained cultures from the anterior nares of 99 available team members for a nasal carriage study.

Laboratory Study

MRSA isolates from case-players and nasal carriers were characterized by using pulsed-field gel electrophoresis (PFGE) with the SmaI and EagI restriction enzymes (12,13). PFGE patterns of the isolates were compared with the USA300 strain responsible for other SSTI outbreaks in Los Angeles County (14). This strain was previously determined to contain SCCmecIV by the Centers for Disease Control and Prevention (L. Yasuda, pers. comm.). We also characterized a sample of methicillin-susceptible Staphylococcus aureus (MSSA) isolates from players’ nasal cultures.

Case-Control and Carrier-Control Studies

On the basis of anecdotal reports of players sleeping in the locker room on used towels and delaying treatment of cuts and abrasions, we hypothesized that poor hygiene habits and compromised skin integrity might predispose players to infection. We designed a standard questionnaire to collect data on player demographics, living situation, football activities, exposure to persons with skin infections, hygiene practices, histories of skin lesions, and clinical symptoms. Trained health department employees administered the questionnaires in person.

We conducted unmatched case-control and carrier-control studies. Controls were selected by jersey numbers, by using a random-number generator, from asymptomatic teammates without nasal carriage of MRSA. Teammates with positive nasal cultures for MRSA were considered carriers. Carrier-players were defined as carriers with matching PFGE pattern to the USA300 strain. We excluded non-USA300 strains carriers to remove players who might represent the background prevalence of MRSA in the community. Players who were not available for interviews were not included. Bivariate analysis was completed by using Fisher exact test in Epi Info version 3.3 (CDC, Atlanta, GA, USA). Statistical significance was defined as p values <0.05. Because of the small sample size and zero-valued cells, similar risk factors from the bivariate analysis were grouped into categories. Multivariate analysis was completed by using the conditional exact test in SAS version 8 (SAS Institute Inc., Cary, NC, USA).

Outbreak Control Interventions

Upon recognition of the outbreak on August 25, team A instituted daily hexachlorophene showers for all players, increased the frequency of cleaning the facilities and athletic gear, disinfected the whirlpool tubs, provided more towels, and posted hand-hygiene signs in the locker room. Once nasal culture results were available, team physicians attempted to decolonize carriers with intranasal mupirocin (15). We recommended improving the timeliness of wound care, barring case-players from playing unless wounds were covered, discouraging the sharing of personal items and tubs, prohibiting sleeping in the locker room, and checking laundry procedures. We also disseminated CA-MRSA educational materials to staff and team members (16).

Characteristics of Case-players

Figure 1

Figure 1. Epidemic curve of clinical and methicillin-resistant Staphylococcus aureus skin and soft tissue infections among players on a college football team by date of diagnosis, Los Angeles County, August–September 2003.

Figure 2

Figure 2. Football field positions; see Table 2 for position-specific attack rates. S, safety; LB, linebacker; CB, cornerback; L, lineman; WR, wide receiver; TE, tight end; QB, quarterback; TB, tailback,...

We identified 11 case-players out of 107 team members for an attack rate of 10%. Cases were diagnosed during or within 2 weeks of the end of training camp (Figure 1). The first case was diagnosed on August 15, the last on September 1. With 1 exception, infections occurred before the first scheduled game on August 30. The most common sign was a boil (Table 1). The elbow was the most common body site infected. No infection was at a current site of skin trauma or occurred at >1 body location simultaneously. Before hospitalization, the index and second case-players were given cephalexin and levofloxacin, respectively, for their infections without any clinical improvement. In total, 4 case-players were hospitalized and treated with parenteral vancomycin. Subsequent nonhospitalized case-players were treated with doxycycline and rifampin. Lesions of 9 players required surgical incision and drainage. All case-players ultimately responded to treatment with resolution of their infections. The median age of case-players was 20 years, with a median tenure of 2 years on team A. Linemen had the highest attack rate (18%) among all field positions (Figure 2, Table 2). No quarterbacks, wide receivers, or special team players (kickers, punters) were affected. All were healthy men without underlying illnesses. Eight (80%) case-players interviewed reported having never worn elbow pads, and 6 (60%) usually did not have cuts or abrasions covered until >1 hour postinjury.

Characteristics of Carriers

Figure 3

Figure 3. Distribution of locker locations for case-players, methicillin-resistant Staphylococcus aureus (MRSA) nasal carriers, and methicillin-susceptible S. aureus (MSSA) nasal carriers.

Nasal cultures were obtained from 99 (93%) of 107 team members. Twenty-six (26%) cultures were positive for Staphylococcus aureus, among which 8 (8%) were positive for MRSA, including player X. Player Y’s nasal culture was negative. The median age of carriers was 20 years (range 18–21 years), and median tenure on the team was 2.5 years (range 1–5 years). MRSA carriage was highest in linemen (38%). We identified 1 case-player with nasal carriage of MRSA. However, trainers obtained nasal cultures after all case-players had begun antimicrobial treatment. Locker room assignments showed clustering of case-players and carrier-players, notably the proximity of the potential source player (player X) to the index case-player (Figure 3). Among MSSA carriers (n = 18), no clustering of locker locations was seen. MSSA carriage was highest among linemen (28%) and cornerbacks/safeties (28%).

Laboratory Results

Four (57%) of seven MRSA isolates from culture-confirmed case-players were available for PFGE analysis. All were indistinguishable from each other, the USA300 strain found in Los Angeles County, and the isolates from 2 cases (players X and Y) in 2002. We denoted this genotype as strain A. Of 6 (75%) available MRSA isolates from 8 carriers, 4 (67%) were indistinguishable from strain A. Two carriers had unique MRSA genotypes (strains B and C) with >7 bands difference between them and between strain A. Strains A, B, and C, player X and Y’s isolates, demonstrate community-associated antimicrobial susceptibility phenotypes (Table 3). Among 5 MSSA isolates characterized, all had >7 bands difference among themselves as well as from the USA300 strain.

Case-control and Carrier-control Study Results

Ten of 11 case-players were enrolled in the study; 1 was unavailable for interview. During camp, case-players were 15 times more likely than controls to have shared bars of soap with teammates and more likely to have had preexisting cuts or abrasions (Table 4).

Five of 6 carrier-players were available for interviews. Carrier-players were 60 times more likely than controls to have had a locker adjacent to or across from a teammate with an SSTI and 47 times more likely to have shared towels with teammates (Table 4). Carrier-players were more likely than controls to lived on campus in a dormitory or fraternity house. Among carrier-players and controls, players who lived on campus had a higher mean number of roommates than those who lived in off-campus apartments (2.3 vs. 1.5, p = 0.046).

Potential risk factors were grouped into 3 categories: “sharing” (sharing soap/towels with teammates), “skin injury” (cuts, abrasions), and “close contact” (locker adjacent to case-players, living on-campus). Multivariate analysis including these categories indicated that sharing was a significant risk factor for CA-MRSA infection (OR 12.1, 95% CI 1.83–108, p = 0.006) and carriage (OR 17.4, 95% CI 1.03–undefined, p = 0.047).

Postintervention Surveillance

Daily hexachlorophene showers were in use from August 25 to September 19. No new infections were reported during the 4 weeks after the discontinuation of the hexachlorophene showers. From October 20 to November 9, MRSA SSTI developed in 4 players: a lineman with a chin abscess, a linebacker (player Y from 2002) with an elbow boil, a quarterback (player Z) with folliculitis on a leg, and a tight end with a gluteal boil. Three MRSA isolates (except from the tight end) were available for PFGE; all matched strain A. The lineman in this cluster shared bars of soap with his roommate, a case-player.

Because of ongoing disease transmission and to identify potential reservoirs of MRSA, all 28 staff and student trainers and managers were nasally cultured on November 3; 11 (39%) were positive for MSSA. None was positive for MRSA. On November 22, we observed an official game. Previously unidentified lapses in hygiene practices occurred on the sidelines. We observed that student trainers reused hand towels between players, and players shared towels among themselves. Subsequently, the team switched to single-use towels on the sidelines. No new infections were reported for the remainder of the 2003 season. In the following season (August–December 2004), no MRSA SSTI outbreak occurred on team A. However, player Z had a recurrence of MRSA pustules on the forearm and leg in October 2004. He responded to outpatient treatment with doxycycline, rifampin, and incision and drainage of the lesions. His MRSA isolate was not available for PFGE. Throughout the last 3 football seasons, we received no reports of SSTI outbreaks among opposing athletes after playing this team.