DOES THEORETICAL TARGET ATTAINMENT REFLECT REAL-WORLD TARGET ACHIEVEMENT? AN EVALUATION IN INTENSIVE CARE UNIT PATIENTS RECEIVING PIPERACILLIN AS CONTINUOUS INFUSION - PAGE Meeting (Population Approach Group Europe)

Johannes Starp ¹, Johanna Sedlmeyr ¹, Sebastian Greppmair ¹, Balázs Poros ², Martin Dunker ², Sarah Koudaimi ², Andreas Bauer ², Michael Paal ^3,4, Stefan Hagel ^5,6, Alexander Brinkmann ⁷, Otto Frey ⁸, Uwe Liebchen ¹

1 Department of Anaesthesiology, LMU University Hospital, LMU Munich (, Germany), 2 Department of Anesthesiology and Intensive Care Medicine, RoMed Klinikum Rosenheim (, Germany), 3 Institute of Laboratory Medicine, LMU University Hospital, LMU Munich (, Germany), 4 SYNLAB Medizinisches Versorgungszentrum (, Germany), 5 Institute for Infectious Diseases and Infection Control, Jena University Hospital-Friedrich Schiller University Jena (, Germany), 6 Center for Sepsis Control and Care (CSCC), Jena University Hospital-Friedrich Schiller University Jena (, Germany), 7 Department of Anaesthesiology, General Hospital Heidenheim (, Germany), 8 Department of Clinical Pharmacy, General Hospital Heidenheim (, Germany)

Introduction: Model-informed precision dosing (MIPD) aims to individualise antimicrobial therapy using pharmacokinetic models. These models need to be evaluated prior to implementation, but robust and clinically interpretable metrics for comparing predictive performance remain limited. Conventional error-based metrics (e.g. mean predictive error) quantify statistical bias and precision – yet, they do not directly reflect therapeutic consequences. This highlights the need for decision-oriented evaluation approaches. Theoretical target attainment (TTA) was therefore introduced as a new model evaluation metric and has been used in external model comparisons previously [1,2]. However, its ability to reflect actual target attainment (ATA) achieved through prospective use of MIPD in real-world settings has not yet been examined. Therefore, this study evaluates the concordance between TTA and ATA.
Methods: TTA was calculated retrospectively in a published dataset of intensive care unit (ICU) patients receiving piperacillin as continuous infusion [3,4]. Concentration predictions (PRED) were obtained in NONMEM® using the population pharmacokinetic model of Kim et al. [5], informed by population characteristics only (a priori), one therapeutic drug monitoring (TDM) sample (Bayesian 1), and two TDM samples (Bayesian 2). For each prediction–observation (OBS) pair, the expected concentration (CE) was calculated as CE=T×OBS/PRED, with T being the target of 55 mg/L (geometric mean of the target range 32-96 mg/L). CE values were categorised as underdosed, within target, and overdosed.
ATA was determined in an independent prospective cohort of 20 ICU patients (46 samples) treated with continuous infusion supported by TDMx (www.tdmx.eu) as an educational MIPD software tool. One sample per patient per day was classified using the same target range. Category distribution for each scenario were compared between TTA and ATA using Fisher’s exact test.
Results: The TTA group included 223 ICU patients (669 samples): 66% male, median age of 67 years [range: 19-90], weight of 80 kg [45-203], Cockcroft-Gault creatinine clearance of 62 mg/L [11-218], and C-reactive protein of 20 mg/L [2-49]. The ATA group included 20 ICU patients (46 samples): 60% male, median age of 69 years [27-80], weight of 81 kg [43-130], Cockcroft-Gault creatinine clearance of 76 [21-173], and C-reactive protein of 9 mg/L [0-242].
TTA within-target proportions increased from 71% (159/223) in the a priori scenario to 87% (195/223) and 92% (205/223) in the Bayesian 1 and 2 scenarios. Corresponding ATA proportions were 80% (16/20), 81% (13/16), and 90% (9/10), respectively. Category distributions (under/target/over) were: a priori: TTA 22/159/42 vs ATA 1/16/3; Bayesian 1: TTA 14/195/14 vs ATA 1/13/2; Bayesian 2: TTA 14/205/4 vs ATA 1/9/0. No statistically significant differences were detected in the distribution of exposure categories between TTA and ATA in a priori (p=0.80), Bayesian 1 (p=0.49), or Bayesian 2 (p=0.58).
Conclusions: TTA approximated the ATA achieved under MIPD based on an external dataset, supporting its external validity as a decision-oriented evaluation metric. The concordant increase in target attainment across days suggests that TTA captures the effect of Bayesian forecasting on dose optimisation. Limitations include the small ATA cohort size and the deterministic nature of TTA, which assumes linear dose-concentration relationships and ideal model application. Further evaluations across other drugs and dosing regimens are warranted.

References:
[1] Greppmair S. Towards model-informed precision dosing of piperacillin: multicenter systematic external evaluation of pharmacokinetic models in critically ill adults with a focus on Bayesian forecasting. Intensive Care Med 2023;49:966–76. https://doi.org/doi:%252010.1007/s00134-023-07154-0.
[2] Starp J, Leonhardt A, Zoller M, Scharf C, Zander J, Paal M, et al. Towards model-informed precision dosing of intravenous linezolid: a multicentre external evaluation of pharmacokinetic models in critically ill adults. Clinical Microbiology and Infection 2026;32:80–6. https://doi.org/10.1016/j.cmi.2025.08.032.
[3] Hagel S, Bach F, Brenner T, Bracht H, Brinkmann A, Annecke T, et al. Effect of therapeutic drug monitoring-based dose optimization of piperacillin/tazobactam on sepsis-related organ dysfunction in patients with sepsis: a randomized controlled trial. Intensive Care Med 2022;48:311–21. https://doi.org/10.1007/s00134-021-06609-6.
[4] Richter DC, Frey O, Röhr A, Roberts JA, Köberer A, Fuchs T, et al. Therapeutic drug monitoring-guided continuous infusion of piperacillin/tazobactam significantly improves pharmacokinetic target attainment in critically ill patients: a retrospective analysis of four years of clinical experience. Infection 2019;47:1001–11. https://doi.org/10.1007/s15010-019-01352-z.
[5] Kim YK, Kim HS, Park S, Kim H, Lee SH, Lee D-H. Population pharmacokinetics of piperacillin/tazobactam in critically ill Korean patients and the effects of extracorporeal membrane oxygenation. Journal of Antimicrobial Chemotherapy 2022;77:1353–64. https://doi.org/10.1093/jac/dkac059.

Reference: PAGE 34 (2026) Abstr 12286 [www.page-meeting.org/?abstract=12286]

Poster: Clinical Applications