HIV and Community-Acquired MRSA

Methicillin-resistant Staphylococcus aureus (MRSA) has been recognized as an important nosocomial pathogen since at least the 1960s. Although most MRSA infections continue to be hospital-acquired, the past decade has brought a sudden, unrelenting rise of community-acquired infections, both in the general population and among HIV-positive people in particular.1^,2

Background

Methicillin resistance develops in S. aureus when the mecA gene, which is carried on a mobile element called staphylococcal chromosomal cassette mec (SCCmec), is acquired and integrated into the staphylococcal chromosome. This gene encodes a penicillin-binding protein (PBP2A) that has reduced affinity for all β-lactams, resulting in methicillin resistance.

Community-acquired MRSA strains are genetically distinct from hospital-acquired ones in several important ways: First, they tend to harbor SCCmec type IV, which is smaller and therefore likely more transmissible than the types carried by hospital-acquired MRSA. Second, they are more likely to produce virulence factors, most notably Panton-Valentine leukocidin (PVL). PVL has been implicated in the pathophysiology of severe necrotizing pneumonia, but the true relevance of this and other toxins is still unclear.3 Finally, community-acquired MRSA strains have greater pathogenic potential than hospital-acquired strains, but they are usually susceptible to more non–β-lactam antibiotics.1^,2

Community-acquired MRSA infection was first reported in the U.S. about 10 years ago. Since then, outbreaks have been described in numerous groups with direct physical contact, including football teams, children in daycare centers, inmates in correctional facilities, and men who have sex with men (MSM). Identified risk factors include crowding, crystal methamphetamine use, antibiotic use, and skin disruption.

Most of the observed increase in community-acquired MRSA infection is attributed to the spread of certain SCCmec type IV–containing clones, the most common of which is the USA300 strain. Now, this community-acquired strain has spread into hospitals: In a 2004 study conducted at Grady Memorial Hospital in Atlanta, 34% of MRSA bloodstream infections were caused by the USA300 clone.4

Epidemiology of MRSA and HIV Infections

In recent years, many HIV specialists have observed an alarming increase in the number of patients with community-acquired MRSA infections. Formal studies of this phenomenon, however, have been somewhat limited: Most, because of their retrospective nature, have used clinical history alone, rather than pulsed-field gel electrophoresis (PFGE) or SCC testing, to distinguish community-acquired from hospital-acquired MRSA. This approach might overestimate the proportion of infections acquired in hospitals, given the frequent, extensive contact that some HIV-positive people have with the healthcare system and the high rate of antibiotic use in this population.

The few studies that have used PFGE or SCC testing clearly demonstrate that most MRSA infections in HIV-positive people are now community-acquired. First, Miller and colleagues evaluated 500 former and active intravenous drug users in 1999 and 2000 in New York City and found that HIV infection was a major risk factor for MRSA infection. Fifty-six (29%) of the HIV-infected individuals had nasal cultures positive for S. aureus; eight of the isolates (14%) were MRSA, and at least seven harbored SCCmec type IV, indicating community origin.5 Second, Carleton and colleagues evaluated S. aureus isolates obtained from HIV-infected adults in San Francisco and found that the USA300 clone represented >90% of isolates in 2004 and 2005, compared with only 33% in 2002.6 Finally, Lee and colleagues studied MRSA skin infections among 35 HIV-positive MSM in Los Angeles and found that the predominant strain was indeed community-acquired.7

The high risk for MRSA infection among HIV-positive people, compared with the general population, could be a reflection of immunodeficiency, certain lifestyle practices, or an increased prevalence of MRSA in the general population. Although evidence exists to support each of these scenarios, the dramatic increase in case rates suggests that we are seeing an unfortunate synergistic coevolution of several epidemiologic factors.

Immunodeficiency

Impaired cellular immunity is a hallmark of HIV infection, but patients with advanced HIV infection may also have inadequate humoral immunity and abnormal chemotaxis, resulting in increased susceptibility to bacterial pathogens including S. aureus. Several studies among HIV-infected patients have indicated that lower CD4-cell counts are an independent risk factor for MRSA colonization5^,8^,9 and infection.10^,11

Lifestyle Factors

High-risk sexual behavior and injection-drug use (IDU) are well documented among HIV-infected patients, particularly MSM, and these behaviors undoubtedly contribute to the elevated rate of MRSA infection in this population. In a retrospective cohort study of HIV-positive adults in San Diego, Mathews and colleagues identified male-to-male sexual contact and IDU as independent risk factors for community-acquired MRSA infections,10 and a similar study pointed to a history of syphilis as being predictive.11 Surveys in hospitals and outpatient clinics in San Francisco and Boston also showed that male-to-male sexual contact was a strong risk factor for multidrug-resistant, community-acquired MRSA infection.12 Overall, these data suggest that high-risk sexual activities, abrasive sex, and IDU will predispose to the spread of MRSA and to increasingly resistant strains in the future.

Nasal Colonization

Nasal colonization with S. aureus is relatively common among HIV-positive people and seems to be associated with increased risk for S. aureus infection. In two cross-sectional studies of HIV-infected patients in Asia, rates of nasal colonization with S. aureus were between 20% and 30%, and 5% of isolates were MRSA.8^,13 In a prospective Veteran Affairs study of more than 200 HIV-infected patients who were followed for at least 2 years, 49% had at least one nasal culture positive for S. aureus.14 Those who had at least two positive cultures, and CD4 counts <100 cells/mm³, had an estimated 10% risk for S. aureus infection per 6-month period. Notably, these and most other previous studies have been limited to cultures of the anterior nares and thus could have underestimated the true prevalence of S. aureus colonization, given that community-acquired MRSA is known to preferentially colonize other regions of the body. In addition, the transient nature of S. aureus colonization is well recognized, and spot checks, such as those performed in the Asian studies, are not reflective of the true burden of the bacteria or the likelihood of it causing disease. Finally, few of these studies distinguished between community- and hospital-acquired MRSA.

A more recent prospective study, conducted in Dallas in 2005, addressed some of these limitations.9 Investigators obtained nasal samples from both the anterior nares and axilla in 146 HIV-infected patients and then tested any identified MRSA strains using PFGE. The prevalence of nasal colonization with MRSA was 10.3%; axillary colonization was rare. Most isolates harbored SCCmec type IV, consistent with community acquisition.

Recent Antibiotic Use

Recent antibiotic use is a known risk factor for MRSA infection in the general population and may also be a risk factor among HIV-positive people. In particular, fluoroquinolones and β-lactams have been shown to increase risk for MRSA colonization and infection, whereas trimethoprim-sulfamethoxazole (TMP-SMX) has been shown to be protective, even against community-acquired strains.6^,7^,9^,10^,11

Clinical Presentation of MRSA Infection

Similar to the general population, HIV-positive people with MRSA typically present with skin and soft-tissue infections (SSTIs). In a retrospective study of clinically significant MRSA infections among HIV-positive people in San Diego from 2000 through 2003, 83% of events were SSTIs, and fewer than 10% were bacteremias.10 Given the high mortality rate associated with bacteremia,14^,15 these data seem reassuring, but a separate retrospective analysis, from the Johns Hopkins HIV clinic, shows that the incidence of MRSA bacteremia has actually been rising — from 5.3 per 1000 person-years in 2000 through 2001 to 11.9 per 1000 person-years in 2003 through 2004.16 Significant risk factors in that study included IDU, end-stage renal disease, and CD4 counts <200 cells/mm³.

Unfortunately, community-acquired MRSA infection seems to be associated with a disturbingly high rate of recurrence (defined as clinical infection at a new site), particularly in people with HIV. In one recent study of MRSA SSTIs among such patients, the rate of recurrence for a 6-month period was 27%.17 Reasons for the elevated recurrence rate in immunocompromised individuals are unclear but might be related to autoinfection and colonization.

Management of MRSA Infection

A substantial proportion of community-acquired MRSA SSTIs will resolve with incision and drainage alone, but antibiotics are usually required for patients with advanced immunodeficiency or clinically significant perilesional cellulitis. Although USA300 strains are typically resistant to β-lactams, they usually retain susceptibility to many other drug classes. In 2006, however, Diep and colleagues described an alarming case of a USA300 isolate that is resistant not only to β-lactams but also to fluoroquinolones, tetracycline, macrolides, clindamycin, and mupirocin.18 More recently, this research group described the spread of multidrug-resistant USA300 among MSM in San Francisco and Boston; risk was independent of HIV infection.12 Use of clindamycin and mupirocin might have contributed to the development of this multidrug-resistant strain, but high-risk behaviors, such as sex with multiple partners, are postulated to have been important factors in disseminating it. Dissemination of such strains would have major implications for the empirical treatment of SSTIs in HIV-infected patients engaging in high-risk sexual activities, but, fortunately, TMP-SMX and rifampin resistance remain rare, at least for now.

Clindamycin remains a commonly used treatment option for community-acquired MRSA, especially in the pediatric population. Notably, many strains that are erythromycin-resistant but test clindamycin-sensitive have an inducible clindamycin resistance, rendering the clindamycin less effective against such strains. Before considering clindamycin use, clinicians should run a so-called D-test to rule out inducible resistance in all erythromycin-resistant, clindamycin-sensitive isolates.

Conclusion

Community-acquired MRSA infections in HIV-infected people can usually be treated in the outpatient setting. However, given the dramatic increase in cases and the high recurrence rates, such infections will continue to pose a major problem for HIV providers and their patients.