Bloom S, Fendrick AM, Chernew ME, Patel P
Clinical and economic effects of mupirocin calcium on
preventing Staph. aureus infection in hemodialysis patients
Am J Kidney Dis
(May) 27:687-694 1996

Medicare spends $6 billion annually on dialysis and related care.
Half of all dialysis patients are nasal carriers of S. aureus
infection in a given year, and half of all carriers develop S.
aureus infections during a given year. Sequelae of infections are
expensive.
The study is based on the premise that eliminating nasal carriage
decreases the risk of S. aureus infections and therefore
decreases cost. Mupirocin calcium is a new topical medication that
eliminates carriage of S. aureus. The study used decision
analysis to determine the cost effectiveness of three strategies of
use of mupirocin calcium.
The first strategy was to screen all patients every three months for
S. aureus nasal carriage and treat with intranasal mupirocin
if positive. The second strategy was to treat ALL HD patients weekly
with intranasal mupirocin without testing for nasal carriage. The
third strategy was not to screen or treat for nasal carriage, treating
secondary (primarily access site) S. aureus infections as
appropriate.
The outcome was the number of infections prevented or treated and the
cost attributed to the outcome. There were 1000 theoretic patients
pre strategy. A decision tree was developed that incorporated the
sensitivity and specificity of screening, estimated compliance and
probability of S. aureus infection. These numbers were derived
from the medical literature. Outcome cost was derived from the Health
Care financing commission. A sensitivity analysis was then performed
to determine the effect of input variables on the cost, number of
infections and cost effectiveness. It was assumed that 75% of S.
aureus infections were related to intranasal carriage. Worst case
scenarios were tested for each alternative (i.e. the lowest rate of
nasal carriage and infection, lowest rate of access loss, and highest
cost of mupirocin).
The decision analysis revealed that a screening and eradication
strategy relative to the non-use of mupirocin theoretically would have
saved Medicare $90 million in direct medical costs in 1994 and result
in a 45% reduction in S. aureus infection). The treatment of
all patients regularly with intranasal mupirocin would have decreased
infections by 55% relative to the non-use of mupirocin, and would have
saved Medicare $130 million dollars.
Comment: The finding that weekly use of intranasal mupirocin
ointment in all patients without screening of carriage of S.
aureus is most cost effective creates a concern that such
treatment will result in the development of resistant strains of S.
aureus and may, in the end cost more. There are few data on
resistance of mupirocin but this is a valid concern that must be
considered when deciding which strategy is most appropriate.
As noted in the discussion, there are several pitfalls in a decision
analysis, for the most part related to the initial assumptions being
made. First, the role of S. aureus nasal carriage in
subsequent infection is not completely clear. Second, there are
discrepancies in the reported rates of S. aureus infection,
carriage and sequelae. Third, the optimal regimen for treating
carriage is unknown. Fourth, there is an assumption that the quality
of life and daily function will improve with treatment of infection.
Fifth, the study did not differentiate between non-MRSA and MRSA and
the related cost differences of infection. Last, the cost of death,
time off from work, etc. were not incorporated into the model.
The study was nevertheless well designed and conservative in its
assumptions and should be helpful in deciding strategies for use of
mupirocin calcium when it becomes available.
(Catherine Stehman-Breen, UW Nephrology Clinical Research Training
Group, Seattle, WA)