Burden of disease of antibiotic resistance The example of MRSA Eva Melander Clinical Microbiology, Lund University Hospital
Discovery of antibiotics Enormous medical gains Significantly reduced morbidity and mortality in bacterial infectious diseases Condition for modern health care Advanced surgery Cytostatic treatment Solid organ and bone marrow transplantations Care of prematurely born children
Antibiotic resistance An inevitable consequence of the use of antibiotics
Resistance history : Staphylococcus aureus Penicillin Penicillinase Methicillin resistance (MRSA) Multiresistance Oxazolidinones Oxazolidinone resistance Vancomycin Penicillinase- stable penicillins Vancomycinresistance (VRSA) 1940 1960 2000
Problem bacteria Hospital care MRSA, VRE, multiresistant gramnegative bacteria Outpatient care Penicillin- and macrolide resistant pneumococci, betalactamase producing H influenzae, trimethoprim resistant E coli MRSA (although often health care related)
Antibiotic resistance = increased morbidity, mortality and costs? To study the attributable effect of antibiotic resistance on morbidity, mortality and costs: compare individuals infected with antibiotic resistant- and susceptible bacteria of the same species
MRSA vs MSSA bacteremia Mortality Literature review: MEDLINE search Articles published 1980 Aug 2005 English Different combinations of these keywords: outcome, mortality, methicillin resistance, bacteremia, endocarditis 202 articles; 47 comparing outcome in MRSA and MSSA bacteremia
MRSA vs MSSA bacteremia Mortality Pointing at different directions Small size Adjustment for confounders poorly/not always performed 2 meta-analyses, including 31 and 9 studies published in 1980-2000: Cosgrove et al, Clin Infect Dis 2003;36:53-9 Whitby et al, Med J Aust 2001;175:264-7
Meta analysis Cosgrove et al, Clin Infect Dis 2003;36:53-9 31 studies, all retrospective, varying methods; 3 963 patients with S aureus bacteremia, 1 360 MRSA 24 no sign. difference in mortality 7 higher all cause in-hospital mortality for MRSA Meta analysis: higher all cause in-hospital mortality, pooled OR 1.93 (95 % CI 1.54-2.42) Heterogeneity subgroup analysis
Meta analysis Cosgrove et al, Clin Infect Dis 2003;36:53-9 Mortality attributable to bacteremia: six studies, pooled OR 2.2 (95 % CI 1.2-3.8) Possible confounders: severity of illness, age, sex, LOS before onset of infection, comorbidities by matching and/ or multivariable regression models; 11 studies; pooled OR 1.88 (95 % CI 1.33-2.69)
Meta analysis Whitby et al, Med J Aust 2001;175:264-7 9 studies, all included in Cosgrove ; 2209 patients, 778 MRSA Only hospital-aquired SA bacteremia studies Similar result: higher all cause in-hospital mortality for MRSA, pooled RR by two different methods 2.12 (95 % CI 1.76-2.57) and 2.03 (95 % CI 1.55-2.65)
MRSA vs MSSA bacteremia Mortality 2001-2005: additionally 14 studies publ. Majority: prospective, larger sample size, better adjustment for confounders Majority: MRSA bactermia means an increased mortality compared to MSSA bacteremia
Why increased mortality? 1. MRSA not more virulent per se 2. Vancomycin less effective than staphylococcal penicillins in severe S aureus infections? 3. Increased risk of getting inadequate empiric antibiotic treatment?
Vancomycin: S aureus bacteremia Vancomycin (compared to staphylococcal penicillins) for S aureus bacteremia and endocarditis, increased risk of: therapeutic failure relapse mortality irrispective of methicillin resistance
Inadequate empiric ab therapy Most of the S aureus bacteremia studies taking inadequate empirical ab treatment into account: Patients with MRSA receive inadequate empirical ab therapy to a higher extent Inadequate ab therapy is an independent predictor (the strongest) of death Patients with MRSA with inadequate empirical ab therapy were at higher risk to die
Inadequate empiric ab therapy In several of the more recent studies comparing outcomes in MRSA vs MSSA bacteremia where no differences in mortality could be shown, the majority of patients had received vancomycin
Number of lab reports of S aureus bacteremia and death certificates mentioning MRSA, England & Wales 1993-2003 16000 14000 12000 No of reports 10000 8000 6000 4000 2000 0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Susceptible No Information Resistant (MRSA)
Mortality SA bacteremia Sweden today vs a 50 % MRSA scenario Conditions: 2250 SA bacteremias in Sweden/year Constant incidence of MSSA bacteremia MRSA bacteremia added to the MSSA bacteremias 5 % attributable mortality in MSSA bacteremia twice as high mortality rate in MRSA bacteremia Today <1 % MRSA: 112 deaths 50 % MRSA scenario: 337 deaths, i.e. 225 extra deaths attributable to MRSA bacteremia
Cost due to more expensive antibiotics Sweden: 20 % of total use of iv staph. penicillins iv vancomycin: 9 million SEK extra 50 % of total use of iv staph. penicillins iv vancomycin: 22,6 million SEK extra 20 % of total use of oral staph. penicillins oral linezolid: 616 million SEK extra 50 % of total use of oral staph. penicillins oral linezolid: 1.54 billion SEK extra (Figures from 2003)
Costs due to MRSA control The EMRSA-16 outbreak in Gothenburg 1997-2001: 147 colonised/infected with MRSA 36 ward units 13 minor outbreaks 4 temporarily closed wards Åhrén, personal communication
Costs due to MRSA control Total cost > 30 million SEK due to: Infectious Disease dept: Extra cost /patient/day (2700 SEK) : 13 million SEK Blockage of ward rooms: 4 million SEK /year MRSA cultures: 9 million SEK Temporary closure of wards (7-10 days): 2-3 million SEK New personnel: physician 1.3 million SEK, 3 nurses 2 million SEK Åhrén, personal communication Björholt et al, Eur J Clin Microbiol Infect Dis 2004;23:688-95
Although antibiotic resistance is considered to be a major public health threat: Lack of data on the extent of the conseqeunces of antibiotic resistance at a society/national level Lack of data in out patient care,nursing homes, rehabilitation clinics Lack of data on loss of working force etc due to antibiotic resistance
In conclusion Data on MRSA may constitute evidence enough for stating that antiobiotic resistance means significant mortality and costs to health care and that it is urgently needed to take action globally.