AMS Basic Course ESCMID_TDM of antibiotics

• Geneeskunde
• WO
• Studiejaar 6

For which antibiotics do you perform TDM?

Dutch guideline on AMS (2016)

• Strong recommendation to perform TDM in patients treated with aminoglycosides, glycopeptides, posaconazole or voriconazole
• Yet, quality of evidence is low (TDM reduces length of hospital stay) to very low (TDM reduces mortality, therapy failure, (nephro)toxicity, costs)

https://swab.nl/exec/file/download/81

Why TDM of vancomycin?

1. Optimal Cmin = optimal efficacy
2. Optimal AUC = optimal efficacy
3. Optimal Cmin = optimal efficacy + low risk toxicity
4. Optimal AUC = optimal efficacy + low risk toxicity

Why TDM of vancomycin: efficacy

Targets:

• AUC_0-24 / MIC > 400 (=AUC_0-24 > 400 at MIC of 1 mg/l or smaller)
• Cmin as surrogate for AUC_0-24: 10 - 20 mg/l

What is your current practice: which vancomycin target is applied in your institution?

1. Cmin between 15 and 20 mg/l
2. Cmin between 10 and 15 mg/l
3. AUC > 400
4. Something else

Why TDM of vancomycin: efficacy

Targets:

• AUC_0-24 / MIC > 400 (= AUC_0-24 > 400 at MIC of 1 mg/l or smaller)
  
• Before 2020: Cmin as surrogate for AUC_0-24: 10 - 20 mg/l
• Since 2020: estimate AUC_0-24 with 1 (Cmin) or 2 (Cmin+Cmax) samples and Bayesian estimation or 1st order PK equations

Rybak AmJHealthSystPharm2020;77:835

www.tdm-monografie.org/vancomycine

Why TDM of vancomycin: efficacy

Targets (Rybak, AmJHealthSystPharm2020;77:835):

• AUC_0-24 / MIC > 400 
• Estimate AUC_0-24 with 1 (Cmin) or 2 (Cmin+Cmax) samples and Bayesian estimation or 1st order PK equations
• Samples to be collectes within 48 h after start of vancomycin; already after 1st dose when eGFR<50 ml/min
• NB Target is based on evidence on MRSA bacteremia

Poll: when treating an infection with Staph. epidermidis with an MIC of 2 mg/l, the vancomycin dose should be increased to target an AUC of 800 (e.g. 2 gram twice daily) as the risk for nephrotoxicity is still acceptable at that AUC

1. True
2. False

Why TDM of vancomycin: toxicity

• Lowest toxic concentration: 

        Cmin: >15 mg/L, >20mg/L, 
        AUC0-24: >600, >800, >1300 
        Css: >30 mg/L

• Bayesian estimated AUC based dosing is associated with lower risk of nephrotoxicity
• Guideline ASHP/IDSA: AUC0-24/MIC target = 400-600

Rybak AmJHealthSystPharm2020, AnnPharmacother2019, 2022, 2023

Apart from the narrow therapeutic window, what's another reason why vancomycin levels must be measured?

1. Pharmacogenetic variability with large effect on PK
2. Large INTERpatient variability in PK
3. Large INTRApatient variability in PK
4. All anserws are correct

TDM is of added value when:

• Interpatient variability in PK is high
  
• Therapeutic window is narrow (small difference between lowest effective conc and lowest toxic conc)

TDM is of added value when:

• Interpatient variability in PK is high AND intrapatient variability in PK is low
  
• Therapeutic window is narrow (small difference between lowest effective conc and lowest toxic conc)
• Measurement of drug concentration is not (too) complex
• Efficacy/toxicity of the drug can not be easily measured/predicted in another way

Poll: what's your current practice with regard to sample collection and interpretation when performing TDM of vancomycin?

1. Bayesian estimation of AUC with 1 (Cmin) or 2 samples (also Cmax)
2. Interpretation of Cmin WITH PK model
3. Interpretation of Cmin WITHOUT PK model
4. Interpretation of level under continuous infusion (without PK model)

A patient (f, 63 j, 79 kg, eGFR 37) is being treated with iv vancomycin, starting yesterday at 9.00am. Starting dose was 1000 mg/24h continuous iv. The plasma level, drawn at 8.30 am today is 22.9 mg/l. Poll: this is an adequate level for this patient

1. True
2. False

Vancomycin: continuous dosing

• Target during continuous iv: C_ss= 17 - 25 mg/l
  (C_ss=AUC / dosing interval -> 400 / 24 = 17 mg/l)

Vancomycin: continuous dosing

• Target during continuous iv: C_ss= 17 - 25 mg/l
  (C_ss=AUC / dosing interval -> 400 / 24 = 17 mg/l)
• Be aware that steady-state has been reached; otherwise estimate AUC with Baysian estimation

Patient (m, 67 j, 73 kg, eGFR 72) starts with vancomycin 1000 mg twice daily at 12.00am and 12.00pm for treatment of a line infection. Sample is drawn on day 2 at 8.00am: 11.1 mg/l. What's next?

1. Stay with 1000 mg twice daily
2. Increase dose to 1250 mg twice daily
3. Switch to 2000 mg continuous infusion
4. I'll ask my TDM consultant

Vancomycin: continuous vs intermittant dosing

• Target during continuous iv: C_ss= 17 - 25 mg/l
  (C_ss=AUC / dosing interval -> 400 / 24 = 17 mg/l)
• Be aware that steady-state has been reached; otherwise estimate AUC with Baysian estimation


• TDM is easier with continuous iv dosing (81% of C_min with intermittant dosing is collected >1 h before new dose, 61% >2 h) [Neely AAC2014])

Patient (m, 79 kg, 1.78 m, eGFR 82 ml/min) starts with gentamicin 7 mg/kg once daily for treatment of sepsis. Which levels would you want to measure and when?

1. Peak and trough before 3rd dose
2. Trough before 3rd dose.
3. Peak and trough after 1st dose.
4. Trough after 1st dose

Why TDM of aminoglycosides: efficacy

• Killing rate of aminoglycosides correlates in vitro with Cmax
  
•  Concentration-dependent killing
• PKPD target = Cmax/MIC 8-10
• MIC of Gram - bacteria for gentamicin and tobramycin is 0.1-2 mg/L, so Cmax target is generally 15-20 mg/L
• Time above MIC not of relevance

Moore et al JInfectDis1987;155:93, Moore et al AmJMed1984;4:657

What is your current practice: in which situations do you measure aminoglycoside peak levels?

1. All patients
2. Only ICU patients
3. Only in selected cases
4. Never

Patient (m, 79 kg, 1.78 m, eGFR 31 ml/min) starts with gentamicin 7 mg/kg once daily for treatment of sepsis. Which levels would you want to measure and when?

1. Peak and trough before 3rd dose
2. Trough before 3rd dose
3. Peak and trough after 1st dose
4. Trough after 1st dose

Why TDM of aminoglycosides: toxicity

• The higher Cmin, the higher the risk for nephrotoxicity
  
• Target Cmin = <0.5 mg/L (also reported <1 mg/L or <2 mg/L)
• Long dosing intervals (24 h or longer)
• Gentamicin may be more nephrotoxic than tobramycin
• Measure Cmin before 3rd dose, unless eGFR<50 mL/min, than already after 1st dose

Nicolau AAC1995;3:650, Hodiamont ClinPharmacokinet2022;61:1075 

What is the availability regarding a TDM service for aminoglycosides and vancomycin at your institution?

1. <24h between sample collection and TDM result
2. >24h, but <72h between sample collection and result
3. No service available locally, samples go to nearby center
4. No TDM service available at all

A patient is treated with ceftazidime 2 grams three times daily for a pneumonia caused by Pseudomonas aeruginosa (ECOFF 8 mg/L). The patient has a eGFR of 55 ml/min. Would you recommend performing TDM in this case?

1. Yes, collect a Cmin after 4th dose
2. Yes, collect a sample 4h after 4th dose
3. No
4. Something else

TDM is of added value when:

• Interpatient variability in PK is high AND intrapatient variability in PK is low
  
• Therapeutic window is narrow (small difference between lowest effective conc and lowest toxic conc)
• Measurement of drug concentration is not (too) complex
• Efficacy/toxicity of the drug can not be easily measured/predicted in another way

TDM is of added value when:

• Interpatient variability in PK is high AND intrapatient variability in PK is low
  
• Therapeutic window is narrow (small difference between lowest effective conc and lowest toxic conc)
• Measurement of drug concentration is not (too) complex
• Efficacy/toxicity of the drug can not be easily measured/predicted in another way

A patient is treated with ceftazidime 2 grams twice daily for a pneumonia caused by Pseudomonas aeruginosa (ECOFF 8 mg/L). The patient has a eGFR of 160 ml/min. Would you recommend performing TDM in this case?

1. Yes, collect a Cmin after 4th dose
2. Yes, collect a sample 4h after 4th dose
3. No
4. Something else

TDM of beta-lactams: efficacy?

• ±1/3 of ICU patients doesn't attain 100% of the Time above MIC (T>MIC)
• TDM should be able to get more patients on 100%T>MIC 

• 3 RCT's: TDM vs no TDM of beta-lactams:

        2 RCT's indeed showed more patients with 100%T>MIC (DeWaele                          IntensiveCareMed2015; Hagel IntensiveCareMed2022), 1 RCT did not                 (Ewoldt IntensiveCareMed2022)

TDM of beta-lactams: efficacy?

• Which T>MIC target to use: 50% or 100%; 1xMIC or 4xMIC? 

       (depending on target population)

• Total or unbound concentration? fT>MIC
• Which sample to collect then?  

        Cmid if 50%(f)T>(4x)MIC

        Cmin of 100%(f)T>(4x)MIC
        Css when continuous infusion
        Model based TDM?

TDM of beta-lactams: efficacy?

• ±1/3 of ICU patients doesn't attain 100% of the Time above MIC (T>MIC)
• TDM should be able to get more patients on 100%T>MIC 

• 3 RCT's: TDM vs no TDM of beta-lactams:
    2 RCT's indeed showed more patients with 100%T>MIC (DeWaele                      IntensiveCareMed2015; Hagel IntensiveCareMed2022), 1 RCT did not            (Ewoldt IntensiveCareMed2022)

• May be TDM is only useful in selected (ICU) cases?

Patient (m, 120 kg, 1.78 m, eGFR 82 ml/min) starts with gentamicin 7 mg/kg ABW once daily for treatment of sepsis. Which levels would you want to measure and when?

1. Peak and trough before 3rd dose
2. Trough before 3rd dose
3. Peak and trough after 1st dose
4. Trough after 1st dose

Why TDM of aminoglycosides: toxicity

• The higher Cmin, the higher the risk for nephrotoxicity
  
• Target Cmin = <0.5 mg/L (also reported <1 mg/L or <2 mg/L)
• Long dosing intervals (24 h or longer)
• Gentamicin may be more nephrotoxic than tobramycin
• Measure Cmin before 3rd dose, unless eGFR<50 mL/min or when the patient is obese, than already after 1st dose 

Nicolau AAC1995;3:650, Hodiamont ClinPharmacokinet2022;61:1075

Have you ever requested or measured a beta-lactam serum/plasma level for TDM?

Yes

No