濫用抗生素之衝擊 衛生署疾病管制局中區傳染病防治醫療網王任賢指揮官

Transcription

1 濫用抗生素之衝擊 衛生署疾病管制局中區傳染病防治醫療網王任賢指揮官

2 濫用抗生素之定義 目前沒有人對此下過定義 但由抗生素使用的目的可見出端倪 抗生素使用之目的 : 將致病細菌殺死, 並不對人體產生重大副作用及誘導出抗藥性菌株 濫用抗生素之定義 : 抗生素之使用若無法有效將致病細菌殺死, 但卻對人體產生重大副作用 或誘導出抗藥性菌株者稱為濫用抗生素

3 濫用抗生素之種類 一. 二. 三. 無法有效將致病細菌殺死 1. 處方無效的抗生素 2. 處方有效的抗生素時沒有遵守藥物動力學之原理 對人體產生重大副作用 1. 處方抗生素時未依肝 腎或其他人體因素調整 2. 多種抗生素合併使用 誘導出抗藥性菌株 1. 抗生素使用期間過長 2. 固定處方少數抗生素

4 濫用抗生素之種類 一. 二. 三. 無法有效將致病細菌殺死 1. 處方無效的抗生素 2. 處方有效的抗生素時沒有遵守藥物動力學之原理 對人體產生重大副作用 1. 處方抗生素時未依肝 腎或其他人體因素調整 2. 多種抗生素合併使用 誘導出抗藥性菌株 1. 抗生素使用期間過長 2. 固定處方少數抗生素

5 抗生素處方之模式 Empirical therapy De-escalation therapy: 開始治療即處方有效抗生素, 病人變好後即停藥或換藥 Escalation therapy: 開始治療先處方無效抗生素, 待培養出來或病人變壞再換藥 Target therapy

6 Dudas Study Predictors of Mortality: Multivariate Analysis Variable Change in initial antibiotics ICU admission >8 hr to administration of first antibiotic Age (Decades) SCr (1.0 mg/dl) RR (10 Breaths/Min) (1.5 to 2.4) WBC 10K/mm (1.1 to 1.9) 2nd/3rd generation CEPH or -lactam/ - lactamase inhibitor + macrolide (non-icu) 2nd/3rd generation CEPH or -lactam/ - lactamase inhibitor + macrolide (ICU) p Value Odds Ratio (95% CI) 3.3 (2.1 to 5.1) 2.5 (1.4 to 4.7) 2.6 (1.3 to 4.9) 1.5 (1.3 to 1.8) 1.2 (1.0 to 1.4) 0.4 (0.2 to 0.8) 0.5 (0.2 to 1.6) Annals Pharmacotherapy 2000;34:

7 Lower Mortality for Patients Who Received Initial Adequate Antimicrobial Therapy In a prospective cohort study of ICU patients with infection (n=655), lower mortality was observed in patients who received initial adequate therapy Hospital Mortality (%) p< p< All Causes Inadequate Therapy 0 Infection-Related Adequate Therapy Adapted with permission from Kollef MH et al. Chest 1999;115:

8 Mortality* Associated with Initial Inadequate Therapy in Critically Ill ICU Patients with HAP or Sepsis Alvarez-Lerma, 1996** Luna, 1997 Rello, 1997 Kollef, % 24.7% 15.6% 38% 37% 33.3% 60.8% 91% Initial adequate therapy Initial inadequate therapy Ibrahim, 2000*** 28.4% 61.9% Harbarth, 2003*** Valles, 2003*** 24% 39% 31% 63% Mortality 0% 20% 40% 60% 80% 100% *Mortality refers to crude or infection-related mortality. **Includes patients with HAP. ***Patients had blood stream infections rather than pneumonia as in the other studies. Alvarez-Lerma F et al. Intensive Care Med 1996;22: Luna CM et al. Chest 1997;111: Rello J et al. Am J Respir Crit Care Med 1997;156: Kollef MH et al. Chest 1998;113: Ibrahim EH at al. Chest 2000;118: Harbarth S et al. Am J Med 2003;115: Valles J et al. Chest 2003;123:

9 Treatment Can Affect Mortality in Patients With Sepsis: Three Interventions % Mortality % 25% % 53% 0 0 Activated C Hydrocortisone 2 Adequate antibiotic protein 1 No Yes therapy 3 * Yes indicates that patients received the specified treatment, No indicates that they did not. 1. Bernard GR et al. N Engl J Med 2001;344: Annane D et al. JAMA 2002;288: Valles J et al. Chest 2003;123: * % 31%

10 Delayed Therapy May Be Inadequate Therapy: Results from a Single-Center Study in VAP Early appropriate therapy, before bacteriologic data are known, leads to an improved outcome. % Mortality % p< % 71% p=ns 70% Adequate ATB therapy Inadequate ATB therapy 0 Pre-BAL (n=68) Post-BAL (n=65) ATB = antibiotic; BAL = bronchoalveolar lavage Adapted from Luna CM et al. Chest 1997;111:

11 Correct Timing of Antibiotic Administration May Improve Survival In a prospective surveillance study of 107 patients with VAP: % (33 of 107) had initially delayed appropriate antibiotic therapy (IDAAT; therapy delayed for >24 hours after meeting diagnostic criteria for VAP). Hospital mortality rate of 69.7% in patients with IDAAT versus 28.4% in patients without IDAAT. In a retrospective cohort study of pneumonia in 14,069 Medicare patients: 2 Administering antibiotics within 8 hours of hospital arrival and collecting blood cultures within 24 hours was associated with improved survival 1. Iregui M et al. Chest 2002;122: Meehan TP et al. JAMA 1997;278:

12 Antibiotic Usage Linked to Bacterial Resistance: A Prospective Study A greater percentage of VAP episodes was caused by potentially drug-resistant bacteria* in patients with prior antibiotic therapy. % VAP episodes* episodes of VAP 0 With Prior Antibiotic Therapy (n=96) Without Prior Antibiotic Therapy (n=39) *Methicillin-resistant Staphylococcus aureus, P. aeruginosa, A. baumannii, S. maltophilia Trouillet J-L. Am J Respir Crit Care Med 1998;157:

13 處方有效的抗生素時沒有 遵守藥物動力學之原理

14 PK/PD and Antimicrobial Efficacy 3 patterns of bacterial killing Concentration dependent with prolonged persistent effect Aminoglycosides, quinolones Correlated with AUC/MIC, Peak/MIC Time dependent with no persistent effect Betalactams Correlated with Time above MIC (T>MIC) Time dependent with moderate to prolonged persistent effect Macrolides, azalides, clindamycin, tetracyclines, glycopeptides, oxazolidinones Correlated with AUC/MIC Craig, 4 th ISAAR, Seoul 2003

15 Time-dependent killing The relationship of time above MIC and the reduction in bacterial count in a neutropenic mouse model of Klebsiella pneumoniae for cefotaxime. (Craig WA. Diagn Microbiol Infect Dis. 1995;22:89 96.)

16 Time-dependent Killing Penicillins Cephalosporins Mortality after 4 days of therapy (%) Mortality of animals infected with pneumococci was 100% when T>MIC = or less than 20% Survival was 90% - 100% when T>MIC exceeded 40%- 50% Time above MIC (%) Craig. Diagn Microbiol Infect Dis 1996; 25:

17 Time-dependent killing Clinical cure rates in otitis media and sinusitis was higher than 80% when the T>MIC for betalactam antibiotics exceeded 40% of the dosing interval. (Dagan etal. J Antimicrob Chemother 2001; 47: )

18 Concentration-dependent killing In a rat model of pneumococcal pneumonia, reliable killing by fluoroquinolones was achieved when the AUC/MIC > 25 (Berry et al J Antimicrob Chemother 2000; 45 [Suppl 1] : 87-93)

19 Levofloxacin PK/PD correlations 134 hospitalized patients with respiratory tract, skin or complicated urinary tract infections treated with 500 mg qd for 5-14 days No. of patients Bacteriologic outcome Success Failure 3 4 AUC:MIC <25 Peak:MIC < AUC:MIC Peak:MIC AUC:MIC >100 Peak:MIC >12 Bacteriologic failure rate 43% 11.5% 1% Preston et al., JAMA 1998, 279:

20 Concentration-dependent killing Probability graph for temperature normalization for Cmax/MIC ratio for aminoglycosides in 78 patients with cultureproven nosocomial gram-negative pneumonia. From Kashuba et al. Interscience Conference on Antimicrobial Agents and Chemotherapy, September 1996 Abstract A100.

21 AUIC and Resistance Thomas JK et al. Antimicrob Agents Chemother. 1998;42: Probability of remaining susceptible AUIC< Days from initiation of Therapy AUIC>101

22 濫用抗生素之種類 一. 二. 三. 無法有效將致病細菌殺死 1. 處方無效的抗生素 2. 處方有效的抗生素時沒有遵守藥物動力學之原理 對人體產生重大副作用 1. 處方抗生素時未依肝 腎或其他人體因素調整 2. 多種抗生素合併使用 誘導出抗藥性菌株 1. 抗生素使用期間過長 2. 固定處方少數抗生素

23 Myelosuppression of Linezolid (n=828) Total <14 days days >28 days Adverse Event (n/%) (n/%) (n/%) (n/%) Thrombocytopenia 62 (7.2) 12 (1.4) 26 (3.1) 24 (2.9) Anemia 35 (4.2) 3 (0.4) 8 (1.0) 24 (2.9) Leukopenia 18 (2.2) 1 (0.12) 5 (0.6) 12 (1.5) Pancytopenia 2 (0.2) 0 2 (0.2) 0 All heme events 117 (14.1) 16 (1.9) 41 (4.9) 60 (7.2) Patients treated > 28 days = 272; = 301; < 14 days = 255

24 Important Safety Considerations (1) Myelosuppression (including anemia, leukopenia, pancytopenia, and thrombocytopenia) has been reported in patients receiving linezolid When linezolid was discontinued, the affected hematologic parameters rose toward pretreatment levels If significant myelosuppression occurs during linezolid therapy, treatment should be stopped unless it is considered absolutely necessary to continue therapy ZYVOX [summary of product characteristics], New York, NY: Pfizer Inc; August 2005.

25 Important Safety Considerations (2) Close monitoring of blood counts is recommended in patients who: 1. Have pre-existing anemia, granulocytopenia, or thrombocytopenia 2. Are receiving concomitant medications that may decrease hemoglobin levels, depress blood counts, or adversely affect platelet count or function 3. Have severe renal insufficiency 4. Receive more than 10 to 14 days of therapy Discontinuation of therapy with linezolid should be considered in patients who develop or have worsening myelosuppression ZYVOX [summary of product characteristics], New York, NY: Pfizer Inc; August 2005.

26 Impact of Increasing Vancomycin Dosage Recommended vancomycin trough level mg/l or mg/l Achievable by 15 mg/kg ever 12 hour Need of a loading dose: 25 mg/kg Baddour LM, et al. Circulation 2005;111:e Gemmell CG, et al. J Antimicrob Agent 2006;57: Wang JT, et al. J Antimicrob Agent 2001;47:246 Higher trough level? mg/l: no outcome difference More renal toxicity Wysocki M, et al. Antimicrob Agents Chemother 2001;45:2460 7

27 Voriconazole PK Variability Total Daily Dose Trough, median (range) Undetect. < 0.5 g/ml 200 mg 2.16 (0-3.07) 25% 25% 400 mg 1.09 ( ) 15% 28% 500 mg 1.67 (0-5.9) 10% 20% 600 mg 1.57 (0-6.75) 17% 22% 800 mg 1.65 (0-12.5) 25% 38% Relationship between dose per kg and trough weak (r = 0.26) Trifilio et al. Cancer 2007; 109: Trifilio et al. Bone Marrow Trans 2005; 35: % undetectable; 25% 62% may be subtherapeutic

28 Voriconazole - Neurotoxicity Serious neurological adverse events in five patients Confusion, agitation, toxic encephalopathy, myoclonies, hallucinations Voriconazole trough concentrations correlated with neurotoxicity Odds ratio 284 (95% CI ,407) for severe toxicity with 2-fold increase in voriconazole levels Encephalopathy VRC trough < 5.5 (N=36) VRC trough > 5.5 (N=16) Incidence 0 5 (31%) Interval after VRC start (Range) NA 9 days (5-30) *Association between neurologic adverse events and elevated voriconazole levels supported by other studies. Pascual et al. Clin Infect Dis 2008; 46: 201. Imhof et al. Swiss Med Wkly 2006; 136: 739. Tan et al. J Clin Pharmacol 2006; 46: 235.

29 濫用抗生素之種類 一. 二. 三. 無法有效將致病細菌殺死 1. 處方無效的抗生素 2. 處方有效的抗生素時沒有遵守藥物動力學之原理 對人體產生重大副作用 1. 處方抗生素時未依肝 腎或其他人體因素調整 2. 多種抗生素合併使用 誘導出抗藥性菌株 1. 抗生素使用期間過長 2. 固定處方少數抗生素

30 Lack of Microbiologic Benefit from Prolonged Antibiotic Therapy Results from a prospective study of 27 VAP patients in a single ICU. Initial empiric therapy was appropriate in all cases (even though Enterobacter and Pseudomonas were the cause of 67% of infections). Major findings: Clinical response to VAP therapy occurred within the first 6 days. Acquired colonization, predominately with resistant pathogens such as P. aeruginosa or Enterobacteriaceae, usually occurred in Week 2 and frequently preceded a recurrent episode. Persistence of endotracheal colonization with these Gram-negative bacteria was likely despite susceptibility to antibiotics. Conclusion: 7-day duration of treatment may be sufficient for patients with VAP. Dennesen PJW et al. Am J Respir Crit Care Med 2001;163:

31 Resolution of VAP Leukocyte count Results from a prospective study of 27 VAP patients in a single ICU showed that parameters of infection improved over time Days 8 (6) Temperature o C Dennesen PJW et al. Am J Respir Crit Care Med 2001;163: Days 5 (3) Day 0 Day 6 Mean days to resolution (median) PaO 2 /FIO Days 6 (2)

32 Treatment of VAP with a Clinical Guideline with a Shorter Course of Therapy Results of a protocol, prospective study of patients (N=102) with clinical diagnosis of VAP with tracheal aspirate or bronchial cultures. Before period: therapy as per treating physician (n=50). After period: patients with VAP received antibiotic treatment according to treatment guidelines; empiric treatment for P. aeruginosa; MRSA with vancomycin, imipenem/ciprofloxacin (selected based on local susceptibility data) (n=52). Modify therapy per culture after hours; try to STOP therapy after 7 days unless clinically indicated otherwise. After initiation of guideline: Similar results in mortality, ICU or hospital length of stay. Lower occurrence of secondary VAP (p=0.03). Ibrahim EH et al. Crit Care Med 2001;29:

33 % Mortality Optimizing De-escalation: Reducing Occurrence of Second Episode of VAP Results of a single-center, prospective clinical study in patients with VAP showed application of a clinical guideline requiring 7 days of treatment increased the administration of initial adequate antibiotic therapy and decreased overall duration of treatment. Before (n=50) Mortality After (n=52) % of patients Before (n=50) 2nd VAP episode No. occurrences *p<0.001, **p=0.03, Mean APACHE II = 25.6, Mean CPIS = 6.7 After (n=52) Before = before institution of guideline; after = after institution of guideline. Guideline included coverage for potentially resistance P. aeruginosa and MRSA. Adapted from Ibrahim EH et al. Crit Care Med 2001;29: * * Days Adequate treatment Before (n=50) Days ** After (n=52)

34 Comparison of Shorter versus Longer Duration of Therapy for VAP* Results from a prospective, multicenter, randomized, double-blind study of 401 patients with VAP evaluating 8-day (n=197) vs. 15-day (n=204) course of antibiotics found: Among patients who had received appropriate initial empiric therapy, comparable clinical efficacy (mortality, recurrence, antibiotic freedays) was observed with 8- and 15-day treatment regimens. Multiresistant pathogens emerged more frequently in recurrences from patients treated for 15 days. The 8-day group had less antibiotic use (p<0.001). In patients with infection due to nonfermenting Gram-negative bacillus, there was a trend towards higher relapse rates with shortduration therapy. *In non-immunocompromised patients. Chastre J et al. JAMA 2003;290:

35 Impact of Prolonged Surgical Prophylaxis DESIGN: Prospective POPULATION: CABG patients (N=2641) Group 1: pts who received < 48 hours of AP Group 2: pts who received > 48 hrs of AP

36 Impact of Prolonged Surgical AP OUTCOMES Incidence of SSI Isolation of a resistant pathogen RESULTS: 43% of patients received AP > 48 hr SSI Incidence <48 hrs group: 8.7% (131/1502) vs >48 hrs group: 8.8% (100/1139), p=1.0 Antimicrobial resistant pathogen OR 1.6 (95% CI )

37 Duration of antibiotic therapy and Microbiologist/Infectious diseases specialist input 254 ITUs, 34 countries The greater the specialist input, the shorter the duration of therapy (P < ) Corona et al. JAC 52 (5): 849. (2003)

38 抗生素固定處方習慣 與抗藥性

39 抗生素固定處方習慣之案例 一碰到發燒就處方 cefa + GM 一碰到 cellulitis 就處方 unasuyn/augmentin 一碰到 MRSA 就處方 vancomycin 一碰到病人休克就處方 Tienam

40 MRSA 的治療用藥 Vancomycin/Teicoplanin Daptomycin Linezolid Tygacil Quinolone + rifampin Baktar + rifampin Fosfomycin

41 ESBL 之治療選擇 Carbapenems (Tienam, meropenem, ertapenem) 2 nd generation cephalosporins Tygacil

42 MDR-AB 之治療選擇 Colistin Tygacil Carbapenem + aminoglycoside Rifampin + carbapenem, tobramycin, colistin

43 Cycling vs. Mixing and Selective Pressure on a Bacterial Clone Cycling offers greater heterogeneity at the level of the ward Mixing offers greater heterogeneity at the level of the individual patient Bergstrom CT, et al. Proc Natl Acad Sci. 2004;101:

44 固定某些抗生素之使用可誘導細菌 對本身及其他抗生素產生抗藥性 Cephalosporins VRE ESBLs (thus MDR) Penicillins and combination inhibitor resistance Clostridium difficle Metronidazole VRE All gram negative Quinolones MRSA VRE ESBLs

45 Antibiotic Usage Impacts Bacterial Resistance Three studies found: In a single-center retrospective study, an increase in VRE (54 cases/10,000 admissions) was associated with third-generation cephalosporins (p<0.001), metronidazole (p=0.008), and longer duration of quinolone use (p=0.03). 1 In a multicenter, prospective study exposure to a -lactam antibiotic containing an oxyimino group (cefuroxime, cefotaxime, ceftriaxone, ceftazidime, aztreonam) was associated with ESBL production (RR 3.8, CI, 1.1 to 13.8). 2 In a single-center retrospective study, emergence of broad-spectrum cephalosporin-resistant Enterobacter spp. in 10% (49/477) of patients with previously susceptible isolates, was explained by antibiotic use leading to resistance due to Type I -lactamase expression. 3 VRE = vancomycin-resistant Enterococcus 1. Carmeli Y et al. Emerg Infect Dis 2002;8: Paterson D et al. Ann Intern Med 2004;140: Kaye KS et al. Antimicrob Agents Chemother 2001;45:

46 Relationship between the use of vancomycin and the development of VRE in U.S.A. % resistance % Vancomycin used resistance Vancomycin used (kg) UCL Seminaire de Pathologie Infectieuse 26 Oct 2000

47 Vancomycin and VRE: Stratified No stratification 3rd G cephalopsporin and metronidazole: No Vancomycin Rx Cases 36% 21% Vancomycin Rx Controls 24% 21% 3rd G cephalopsporin and/or metronidazole: Yes 44% 42%

48 Association of Antecedent IV Antibiotic Treatment and VRE Case Control Unadjusted Adjusted IV drug OR p OR p Vancomycin 29% 19% Metronidazole 20% 9% rd G Cephalosporins 30% 5% Penicillins 29% 21% Beta-inhibitors 21% 15% Clindamycin 8.6% 7.9% Imipenem 8.2% 4.2% Quinolones 21% 11%

49 Effect of duration of IV treatment and VRE Regimen Vancomycin Metronidazole 3rd G Cephalosporins Quinolones RR p

50 Meta-analysis on the effect of Anti-GNB agents on VRE Beltrami Bhavnani Bonten D'Agata Falk Handwerger Henning Hwang Loeb Loeb Lucas Morris (study 2) Ostrowsky 1999 Ostrowsky 2001 Peset Singh-Naz Slaughter Stosor Tokars Tornieporth Combined Bhavnani D'Agata Falk Gordts Loeb Lucas Morris (study 1) Morris (study 2) Ostrowsky Shay (study 1) Stosor Tokars Combined OR Cephalosporins rr Quinolones Harbarth S. AAC 2002

51 Quinolone use and resistance Dramatic increase in quinolone use Accompanied by increase in proportion of Quinolone-resistant gram negative MRSA C. difficle ESBL

52 Secular trends in quinolone resistant Enterobacteriaceae

53 Fluoroquinolones Resistance in USA PA (r=0.976; p<0.001) GNB (r=0.891; p<0.001) Neuhauser MM, JAMA 2003;289:885

54 Differences Between Various Quinolones in Selecting QR-PA Group QR-PA QS-PA Control N Ciprofloxacin exposure 9.5% 8% 7.5% Levofloxacin 36%* 14% 22% exposure * OR 2.0, p=0.03 Kaye K. ICAAC 2004

55 Quinolones and S. aureus MRSA vs. MSSA MRSA MSSA Weber S. EID 2003

56 Mechanisms of Resistance: Pseudomonas and Efflux Pumps Efflux System Exit Portal (OprM) Outer Membrane Meropenem is pumped out while imipenem is not Imipenem and meropenem enter here Porin Periplasm Linker Lipoprotein (Mex A) Cytoplasmic Membrane Efflux System Pump (Mex B) Adapted with permission from Livermore DM. Clin Infect Dis 2002;34:

57 抗生素管制之目的 讓病患獲得最適當的抗生素治療 減少細菌產生抗藥性

58 In 2002, out of 89 new drugs approved by U.S. FDA, no new antibiotics were approved.

59 Antibacterial Agents Approved by U.S. FDA, '02-04 Source: Speilberg et al. CID May 1, 2004

60 懇請賜教