Trends in antibiotic use in Swedish Intensive Care Units1999-2009 - PowerPoint PPT Presentation

Trends in antibiotic use in Swedish Intensive Care Units1999-2009 M. Edström, H. Gill, G. Fransson, S. Walther, H. Hanberger ICU-Strama and Swedish ICU registry www.icuregswe.org www.strama.se

Background • Swedish ICU-Strama started 1999 • ICU-Strama collects aggregated ICU data on – antibiotic use – antibiotic resistance – infection control procedures. • From 2005 joint project between ICU-Strama and the Swedish Intensive Care Registry (SIR) • SIR collects data from individual patients. www.icuregswe.org www.strama.se

Members of Swedish Intensive Care Registry (SIR) 2008 • 70 members • 18 non members • 52 ICUs reported number of bed days 2008 www.icuregswe.org

Materials and methods • Hospital pharmacies provided data on delivered amounts of antibiotics to the ICUs. • ATC-classification of antibiotics were used. • Annually updated DDDs calculated by WHO Collaborating Centre for Drug Statics and Methodology were used. • Number of bed days were based on all admissions to the ICUs. • Data were analysed using linear regression and Spearman’s rank correlation. www.whocc.no

Mean Total Antibiotic Consumption in Swedish ICUs 1999-2009 • 1999: 1216 defined daily doses per 1,000 occupied bed days (DDD 1000 ) • 2009: 1425 DDD 1000 • P=<0,001 • Trend=31 DDD 1000 /year 1800 1600 1400 1200 DDD1000 1000 800 600 400 200 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 www.icuregswe.org www.strama.se

Total antibiotic consumption 2009 - Swedish ICU:s 3000,00 Range: 680-2698 DDD 1000 2500,00 2000,00 D D D 1 0 0 0 Median: 1354 DDD 1000 1500,00 Median 1000,00 500,00 0,00 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 ICUs 1-48 ICUs 1-48 www.icuregswe.org www.strama.se

Carbapenems Trend=7.5 DDD 1000 /year 250 200 P=<0.001 D D D 1 0 0 0 150 100 50 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 www.icuregswe.org www.strama.se

Piperacillin-Tazobactam Trend=10 DDD 1000 /year 140 120 100 DDD1000 80 P=<0.001 60 40 20 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 www.icuregswe.org www.strama.se

Triazole Trend=7 DDD 1000 /year 160 140 120 P=0.001 100 D D D 1 0 0 0 80 60 40 20 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 www.icuregswe.org www.strama.se

Cephalosporins P=0.2 (NS) 400 350 300 250 0 0 0 1 200 D D D 150 100 50 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 STRAMA launches campaign to reduce cephalosporin consumption www.icuregswe.org www.strama.se

Increase in antibiotic consumption DDD 1000 - Average weighted by number of bed days Trend (DDD 1000 Antibiotic 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 p /Year) Aminoglycosides 29 26 25 25 32 38 36 36 41 43 43 1,91 <0.001 Beta-lactamase sensitive penicillins 49 51 62 57 63 60 56 59 63 73 79 2,17 0,002 Carbapenems 125 125 130 120 143 150 149 156 178 191 192 7,47 <0.001 Cephalosporins (all) 336 311 317 325 332 362 356 362 333 262 203 -6,52 0,16 Combinations of penicillins, incl. beta- lactamase inhibitors 22 25 20 27 34 43 46 61 79 102 120 9,58 <0.001 Fluoroquinolones 92 100 104 121 114 118 107 105 101 99 83 -0,88 0,43 Glycopeptides 27 25 28 25 22 34 31 33 36 40 48 1,94 <0.001 Imidazoles (J01XD) 61 60 64 70 67 70 61 65 62 52 41 -1,38 0,08 Isoxazolyl-pc 204 224 230 272 248 260 223 197 230 245 211 -0,5 0,84 Lincosamides 41 39 41 48 48 46 41 44 47 48 44 0,47 0,15 Macrolides 58 52 43 41 32 36 40 42 48 45 42 -0,77 0,28 Triazole derivatives 73 81 65 84 97 96 99 113 151 130 129 7,22 <0.001 other 99 95 120 115 127 146 117 151 175 144 192 Total 1216 1214 1249 1330 1359 1459 1362 1424 1544 1474 1425 30,69 <0.001

Trends in antibiotic use in Swedish Intensive Care Units1999-2009 Major Findings • Reduced cephalosporin consumption (last 3 years) • Increased carbapenem consumption • Increased piperacillin-tazobactam consumption What about resistance to these drugs?

Trends towards increased cephalosporin and pip-tazo resistance among E. coli isolated from Swedish ICU patients 2005 - 2008 n=3900 2005 2006 2007 2008 2009? 10 % 10 % 15%

Correlation between AB consumption and standardised mortality rate (SMR) based on SAPS3 - 31 ICU:s 2009 Correlation of SMR to DDD1000 P=0.95 (NS) 2500 2000 1500 DDD1000 1000 500 0 0 0,2 0,4 0,6 0,8 1 SMR www.strama.se www.icuregswe.org

Correlation between AB consumption and standardised mortality rate (SMR) based on SAPS3 - 31 ICU:s 2009 Correlation of SMR to DDD1000 P=0.95 (NS) 2500 2000 Conclusion Swedish ICUs with high (upper 1500 DDD1000 range) antibiotic consumption may reduce it without increased 1000 risk for critically ill patients 500 0 0 0,2 0,4 0,6 0,8 1 SMR www.strama.se www.icuregswe.org

Antibiotic Use based on drugs delivered by each hospital pharmacy Potential errors • drugs may be delivered but not administered to patients • dosing in ICU – not always standard WHO doses - increased dosing due to life-threatening disease (sepsis/meningitis etc) - decreased dosing in renal failure - lower doses in children • differences in reporting bed days

Why continue to collect data based on drugs delivered by each hospital pharmacy? • Easy access • Low cost • Not time consuming • Sustainable • Useful for bench marking • May be used for measuring effect of interventions = Time series analysis

Measurable goals for antibiotic use in ICUs 1. Appropriate antibiotics without delay 2. Culture before start of antibiotics 3. Narrow spectrum based on culture results 4. Compliance to local and national treatment guidelines 5. Shorter courses 6. Reduce broad spectrum