VISA—Not Welcome Everywhere


Synopsis: At the very least, the message of vancomycin-resistant Staphylococcus aureus reinforces the need to rationally and sparingly use the antibiotics we already have, as they are not a renewable resource.

Sources: Tabaqchali S. Vancomycin-resistant Staphylococcus aureus: Apocalypse now? Lancet 1997;350:1644-1645; Hiramatsu K, et al. Dissemination in Japanese hospitals of strains of Staphylococcus aureus heterogeneously resistant to vancomycin. Lancet 1997;350:1670-1673; Detecting vancomycin-intermediate Staphylococcus aureus: Increased vigilance required by European microbiologists. Eurosurveillance Weekly January 8, 1998. ( eurosurv/e2-1.html)

For antibiotic resistance, 1997 may have been a watershed year since the first clinical isolate of MRSA with intermediate resistance to vancomycin was reported in Japan (Infect Dis Alert 1997;16:179.) This strain, MRSA Mu50, showed intermediate susceptibility to vancomycin (MIC 8 mg/L compared with 1 mg/L for typical, fully susceptible S. aureus)—hence, VISA (vancomycin intermediate Staphylococcus aureus). The strain also exhibited homogeneous insensitivity to vancomycin since all the individual bacterial cells in the culture population expressed resistance. By contrast, a second VISA strain, MRSA Mu3, isolated from the sputum of a patient treated in the same hospital with post-surgical pneumonia after 12 days of vancomycin, exhibited an MIC of 3 mg/L vancomycin. Only a small fraction (1 in a million) of the cell population proved resistant with sub-clones showing MICs from 2 to 9 mg/L. This strain was, therefore, designated as showing heterogeneous resistance and has been dubbed "hetero VISA." Similar vancomycin-resistant MRSA have also been reported in the United States (Infect Dis Alert 1997;17:12) and now in Slovakia, but is it all doom and gloom?

In addition to exploring the nature of resistance in the Japanese VISA strains, Hiramatsu et al go on to demonstrate that the problem in Japan is one of clonal spread. A survey of Japanese hospitals around Tokyo and Hiroshima showed the VISA strains (both homogeneous and heterogeneous resistance) all belonged to a single clone, type II-A—providing clear evidence that it has spread. University hospitals also seemed most affected; 12 (9.3%) hetero-VISA were found among the 129 isolates of MRSA affecting four of the seven Japanese universities sampled, whereas only 13 (1.3%) such strains were found among the 970 MRSA isolated in non-university institutions affecting 12 of the 129 hospitals sampled. Importantly, the transmissable vanA and vanB genes responsible for high-level resistance in the enterococci are not involved. Rather, the VISA strains examined so far show cell-wall thickening and increases in PBPs 2 and 2' and in murein monomer precursors, indicating resistance to be due to enhanced cell-wall synthesis rather than alterations in cell-wall composition. This would also correspond with the gradual selections of spontaneously occurring mutants reported by Hiramatsu et al who concluded, not unreasonably, that the hetero-VISA strains probably represent precursors of the homogeneous VRSA.

By contrast, the sporadic reports of VISA in the United States and Europe seem analogous to what occurs in the Petri dish. Independent cultures (infections) involving large numbers of bacteria are being exposed to sub-lethal concentrations of vancomycin that give selective advantage to spontaneously occurring mutants. These gradually increase in number, resulting in the appearance of colonies on Petri dishes (persistent positive cultures in patients). Ironically, this may provide the most crucial clue as to how to avoid the appearance of more VISA: Don’t persist in treating any MRSA infection in an occluded site with vancomycin when there is little or no response. Rather, change therapy. The question is, to what? Hiramatsu et al succeeded with ampicillin/sulbactam and arbekacin. This, or an equivalent beta-lactam/beta-lactamase/ aminoglycoside regimen, seems a reasonable starting point.

The next question is: How do we assess the risk of VISA. Although formal studies to identify risk factors are not yet possible, as there have been so few reports of VISA outside Japan, we can make an educated guess as to what they might be. University hospitals are more at risk than are their county counterparts probably because they offer complex surgery including transplantation as well as intensive care, have sicker patients referred to them, and, importantly, have easier access to all classes of antibiotics including those under investigation and often without commensurate controls. (See figure.) They are also more likely to use vancomycin empirically as well as for specific indications. However, the prevalence of MRSA is probably the most critical factor. VISA are obviously much less likely to evolve where prevalence is low, such as in Scandinavia where it is less than 1% (Voss, et al. Eur J Clin Microbiol Infect Dis 1994;13:50-55), but have become increasingly more likely as the prevalence rises (e.g., Italy and the United States). Hospitals that have already encountered infections due to vancomycin-resistant enterococci are probably prime candidates for VISA if they also have endemic MRSA, since these VRE infections indicate both a breakdown in hospital hygiene and excessive use of vancomycin.


At risk for VISA

Therefore, far from every hospital immediately unleashing screening programs for VISA, the evidence suggests that only those with endemic MRSA need to do so. Those units within the affected hospitals should act as sentinels; all isolates of S. aureus from patients nursed in them should be screened for resistance to vancomycin, but laboratories have to use the right test. The PHLS Antibiotic Reference Unit in London tested VISA strains Mu50 (homogeneous resistance) and Mu3 (heterogeneous resistance) reported by Hiramatsu et al and found that E-tests or breakpoint dilution tests with 4 mg/L vancomycin were needed, as the routine disc diffusion was unreliable. Therefore, MRSA should be tested for intermediate vancomycin resistance by culturing on agar containing 4 mg/L drug. VISA strains should then be sent to an appropriate national reference laboratory for confirmation and to determine the mechanism for resistance.

Should a VISA infection occur, strict infection-control measures should be implemented to minimize the risk of these strains ever becoming endemic. The PHLS paper also recommends that affected patients are isolated or at least nursed in cohorts to prevent further spread, and that all others on the ward should be screened by culturing swabs from the nose, perineum, wounds, and any manipulated sites (e.g., around catheters). All isolates of S. aureus should obviously be screened for resistance to 4 mg/L vancomycin. It may also be necessary to undertake a wider survey and to examine health care workers for skin lesions and to screen their nose and perineum by culture for VISA—especially on high-risk units. Last, but by no means least, antibiotic policies should be reviewed to limit overprescribing of antibiotics in general and to restrict the use of glycopeptides in particular.

Given the nature of its resistance, the emergence of VISA clearly shows that antibiotic prescribing is to blame. Exposing a large mass of largely dormant bacteria to any drug that requires active growth to be effective is inviting trouble. All bacterial populations throw up mutants spontaneously, and the larger the microbial mass, the greater the number of mutants and the more likely it is that one of these mutants will be able to gain a selective advantage despite the adverse conditions that are obtained during antibiotic treatment. Bacteria are, after all, the ultimate survivors, and S. aureus is one of their most accomplished species. We may draw some comfort from the fact that the mechanism of vancomycin resistance in VISA involves a subtle change in cell-wall synthesis rather than any transmissible gene, making explosive outbreaks less likely. But we should not be complacent. At the very least, the message of VISA reinforces the need to rationally and sparingly use the antibiotics we already have, as they are not a renewable resource. Perhaps the fact that VISA is not a unique acronym may be of some help not only in raising awareness but also as a means of persuading everyone to adhere to the guidelines directly by exacting a penalty from offenders and investing the money in a vigilance program that also rewards those who actively support it. —jpd

University hospitals are more at risk for VISA than their county counterparts probably because they:

a. have sicker patients referred to them.

b. have easier access to all classes of antibiotics, including those under investigation.

c. offer complex surgery, including transplantation and surgery.

d. all of the above.