Sage Journals: Discover world-class research

Abstract

Acute kidney injury is a frequent complication in critically ill patients and often necessitates kidney replacement therapy. Optimising antimicrobial dosing in this population is essential yet challenging owing to altered pharmacokinetics in critical illness and the wide variability of prescriptions utilised in kidney replacement therapy. This case report investigates the pharmacokinetics of cefazolin in a critically ill patient receiving prolonged intermittent kidney replacement therapy, offering insights into dosing considerations in this clinical setting.

Keywords

Cefazolin pharmacokinetics renal replacement therapy prolonged intermittent kidney replacement therapy critically ill

Introduction

Septic shock causing acute kidney injury requiring kidney replacement therapy is associated with a 90-day mortality of 56%.¹ In the critically ill septic patient, multiple pharmacokinetic variables are altered, including an increased volume of distribution, alteration of protein binding capacity and organ dysfunction affecting clearance.^2–5 Severe kidney dysfunction necessitating renal replacement therapy may further complicate antimicrobial dosing, especially when using prolonged intermittent kidney replacement therapy (PIKRT) with its wide variability of prescriptions, such as frequency, duration, blood and dialysate flow rates, all of which significantly impact clearance of highly dialysed antimicrobials such as cefazolin.^4,6,7 Given the elevated mortality in this patient population and complexities of altered pharmacokinetics, there is a critical need to optimise antimicrobial dosing strategies to improve therapeutic efficacy and clinical outcomes.

Cefazolin is a first-generation cephalosporin which exhibits bactericidal action through bacterial cell wall synthesis inhibition.^2,8 It is a front-line treatment option for methicillin-susceptible Staphylococcus aureus (MSSA) bacteraemia and endocarditis including patients with non-severe penicillin hypersensitivity.^9–11 Emerging evidence supports cefazolin as an appropriate agent for the treatment of patients undergoing intermittent haemodialysis with MSSA bacteraemia.^6,12,13 As cefazolin exhibits time-dependent pharmacodynamics, the bactericidal effect is improved with greater proportion of time (T) that the unbound drug concentration (f ) remains over the causative organism’s minimum inhibitory concentration (MIC).^2,14 A suggested pharmacodynamic target for cefazolin would be 60–70% f T > MIC, with higher targets such as 100% f T > MIC also being proposed.^2,8,14 Routinely, cefazolin susceptibility is inferred from cefoxitin or oxacillin susceptibility testing or by molecular detection of mecA or mecC gene.^11,15 According to the European Committee on Antimicrobial Susceptibility Testing, the majority of MSSA infections have an MIC of ⩽2 mg/L to cefazolin.¹⁵

Cefazolin has a molecular weight of 475 Da and is primarily renally excreted with 70–80% excreted as unchanged drug.^8,13 In patients with end-stage kidney disease, the clearance is approximately 0.2 L/h.⁶ Cefazolin is highly protein bound (80–90%) in healthy individuals, with a volume of distribution of 0.13 L/kg in patients with normal renal function, increasing to 0.16 L/kg in patients with end-stage kidney disease.^5,6 This case report investigates cefazolin pharmacokinetics in a critically ill patient undergoing PIKRT.

Case report

A 77-year-old female (156 cm height, 49 kg weight) presented via the emergency department with disseminated MSSA bacteraemia associated with obstructive urosepsis of the left kidney, polyarticular septic arthritis, tricuspid valve endocarditis and septic pulmonary emboli. MSSA was isolated from multiple sets of blood cultures, bilateral knee and shoulder aspirates as well as urine and endotracheal specimens. The patient had a known penicillin hypersensitivity manifesting as nausea, vomiting and rash.

Operative management included insertion of left ureteric stent on day 1 of hospitalisation followed by bilateral arthroscopic knee and open right shoulder washouts on day 5. The patient was admitted to the intensive care unit (ICU) within 24 h of hospitalisation with septic shock and an established acute kidney injury (serum creatinine 215 µmol/L, estimated glomerular filtration rate (eGFR) 19 mL/min per 1.73 m² from baseline serum creatinine of 75 µmol/L, eGFR 66 mL/min per 1.73 m²) which did not resolve with resolution of hydronephrosis, and necessitated kidney replacement therapy. During the sampling period, the patient remained oliguric with a urine output of 5–15 mL/h, with serum creatinine of 184 µmol/L and eGFR 22 mL/min per 1.73 m². The first negative blood cultures occurred on day 7 of hospitalisation.

The patient was commenced on cefazolin from day 1 of the admission. During the sampling period, the patient was administered either intravenous (IV) cefazolin 1 g (~20 mg/kg) over 30 min every 12 h (while not undergoing PIKRT), or IV cefazolin 2 g (~40 mg/kg) over 30 min (while undergoing PIKRT).

PIKRT was prescribed in accordance with the ICU’s standard practices using the Fresenius 5008 with a 1.4 m² filter (Fresenius, Sydney, Australia). On the sampling day, the dialysis session ran over 8 h with a blood flow rate at 200 mL/min; dialysate flow at 240 mL/min; predilutional substitution rate at 50 mL/min and a net total ultrafiltration of 2360 mL. This delivered a urea reduction ratio of 62.94% and a Kt/V (K = clearance, t = time, V = volume of distribution) of 1.31.

MSSA-directed therapy was continued for a total of 12 weeks owing to magnetic resonance imaging showing evidence of L2–L4 osteomyelitis/discitis. After 21 days of cefazolin treatment, the patient had several weeks of broader antimicrobial therapy including cefepime, meropenem and moxifloxacin owing to concerns of hospital acquired bacterial infections. Kidney recovery occurred approximately four weeks into admission and following approximately five months in hospital, including 17 days in ICU and over two months of rehabilitation, the patient was successfully discharged to a nursing home facility with stable kidney function (serum creatinine 78 µmol/L).

Informed consent was obtained to report this case.

Analytical methods

A total of eight blood samples were taken over a 24 h period which included one PIKRT session. These samples were centrifuged within 6 h at 1800 to 3000 rev/min for 10 to 15 min. Plasma specimens were stored at −80°C prior to assay. Total and unbound cefazolin concentrations were measured in plasma specimens by validated high performance liquid chromatography mass spectrometry at the University of Queensland Centre for Clinical Research, Brisbane, Australia. Analytes were validated from 1 to 500 mg/L for total cefazolin and 0.5 to 500 mg/L for unbound cefazolin, with an intra- and inter-run precision of <5% for total cefazolin and <10% for unbound cefazolin.

Compartmental modelling of unbound cefazolin concentrations was undertaken using NONMEM (v.7.5.1),¹⁶ testing both one- and two-compartment models, with elimination from the central compartment and several residual error models (additive, proportional, combined). The model included assessment of two components of clearance to represent the endogenous (off-PIKRT) clearance and the PIKRT-related clearance. Given that no pre-dose concentration was available, and prior dosing included a period on continuous kidney replacement therapy, the pre-dose concentration was estimated as a part of the modelling procedure. Simulations were performed within NONMEM using the final model (detailed in Supplementary material supplement 1 online) to obtain time versus concentrations profile and percentage of drug removed during PIKRT of proposed dose regimens.

Results

The unbound cefazolin pharmacokinetic data is contained in Table 1, the fitted pharmacokinetic curve is plotted along with the observed concentrations in Figure 1. Modelling proceeded with two compartments and a proportional error model. There was a significantly better fit as compared with a one-compartment model (change in objective function value –34.795, P < 0.001, df = 2), with reduction in the residual error and reduced bias in the fit to the observed data (Figure 1). The component of additive error was estimated to be negligible, consistent with concentrations modelled being many times higher than the lowest limit of quantification of the assay (0.5 mg/l) and a similar co-efficient of variability of 5–10% performance in the range of measured concentrations in this case report. The baseline concentration was estimated to be 111 mg/L within the model. The inclusion of a separate parameter for clearance during PIKRT was significant (P < 0.001). Clearance during PIKRT was 37.8 times higher than endogenous clearance (11.15 vs 0.295 L/h, respectively) with a terminal elimination half-life of 3.36 h on-PIKRT compared with 80.2 h off-PIKRT. Simulations determined that for second daily dosing of 1 g cefazolin post-PIKRT, with an 8-h PIKRT session, 82% of the unbound cefazolin present at the commencement of a PIKRT session was removed with a trough concentration of 2.6 mg/L.

Table 1.

The influence of prolonged intermittent kidney replacement therapy (PIKRT) on unbound cefazolin pharmacokinetics (PK).

PK parameter	Value (SE%)	(95% CI)
CL_native (L/h)	0.295 (36)	(0.0801 to 0.535)
CL_PIKRT (L/h)	11.2 (1)	(9.90 to 12.1)
V_C (L)	13.5 (7)	(8.94 to 17.7)
Q (L/h)	8.33 (6)	(5.27 to 10.7)
V_P (L)	20.2 (1)	(17.7 to 22.7)
RV (%)	1.8 (13)	(1.2 to 3.3)
	Secondary PK parameters for 1000 mg 48 hourly, post 8 h PIKRT
t_{½, β} native (h)	80.2
t_{½, β} PIKRT (h)	3.36
Trough unbound concentration (mg/L)	2.60
% removed per 8 h PIKRT	82

CL_native = native clearance (without PIKRT), CL_PIKRT = clearance during PIKRT, V_C = central volume of distribution, Q = inter-compartmental clearance, V_P = peripheral volume of distribution, F_PIKRT = bioavailability of dose given partially during PIKRT, RV = residual variability of PK model, t_{½, β} = terminal elimination half-life, SE = standard error determined from $COV step in NONMEN, CI = confidence interval determined from log-likelihood profiling.

Figure 1.

Time concentration profile of unbound cefazolin in a patient undergoing prolonged intermittent kidney replacement therapy (PIKRT). Time scale (in hours) is relative to the dose prior to the first sample. Solid line represents the time concentration profile based on the PK model parameters for a) 1-compartment model and b) 2-compartment model, crosses represent observed concentrations and shaded area represent times PIKRT was applied. Dose amounts (in milligrams) are included along the time axis at times doses were given.

Discussion

Cefazolin is removed by both intermittent haemodialysis and continuous kidney replacement therapy compared with poor native clearance in severe kidney impairment.^3,5–7,17 Although cefazolin clearance is variable based on individualised dialysis parameters, generally a 15–20 mg/kg dose (or 1–2 g) dose is recommended after dialysis for intermittent haemodialysis, though these studies were not performed in critically ill patients with acute kidney impairment.^6,7 A higher dose of cefazolin ranging from 1 g to 2 g twice daily is suggested for patients undergoing continuous kidney replacement therapy.^3,5 In this patient, PIKRT enhanced cefazolin clearance (11.15 L/h) compared with when the patient was off PIKRT (0.295 L/h), which is consistent with other forms of kidney replacement therapy in patients with end-stage kidney disease.⁶ Pharmacokinetic modelling using our patient data demonstrates the fT > MIC of unbound cefazolin would exceed 99.9%, if only 1 g of cefazolin was administered every 48 h immediately post-PIKRT. Given individual patient and PIKRT variabilities and the dynamic nature of a critically ill patient, however, further data is required to provide dosing recommendations that can be extrapolated.

We observed a high unbound fraction in the plasma samples for this patient (mean of 56% unbound fraction while not on PIKRT), which could be due to reduced protein binding due to hypoalbuminemia or the competitive binding of excess urea.^2,6 Only the unbound fraction of cefazolin is expected to be cleared by dialysis, therefore, a higher unbound fraction might result in greater clearance.² Volume of distribution of unbound cefazolin determined by modelling was 33.7 L (0.69 L/kg), which is higher than in healthy patients with normal or impaired kidney function. Volume of distribution of hydrophilic antimicrobials, such as cefazolin, are likely increased in sepsis owing to administration of large volumes of fluids required for resuscitation and treatment as well as greater distribution into the interstitial space due to reduced protein binding, which can increase tissue penetration.^4,6 Though higher concentrations might be seen in tissues, the peak plasma concentration will likely be reduced and the half-life will be prolonged, thus also leading to lower clearance by dialysis.² These dynamic and often competing pharmacokinetic alterations highlight the complexity of antimicrobial dosing in critically ill patients.

Supratherapeutic dosing resulting in elevated cephalosporin plasma concentrations has been associated with increased risk of central nervous system adverse drug reactions including encephalopathy, seizures, confusion and coma.^18,19 Other risk factors for cephalosporin induced neurotoxicity include renal impairment, older age and pre-existing central nervous system disease.^18,19 The balance between ensuring adequate plasma and tissue concentrations while avoiding supratherapeutic exposure should also be considered, particularly in patients with risk factors for neurotoxicity.

Limitations of this case report include the absence of tissue concentration measurements, which would reflect tissue penetration of cefazolin. Additionally, the pharmacokinetic model represents a point in time of this patient’s ICU course; given the rapidly evolving physiology in critical illness, ongoing reassessment of antimicrobial dosing is essential as patient specific variables change.

Supplemental Material

sj-txt-1-aic-10.1177_0310057X251396229 – Supplemental material for Cefazolin pharmacokinetics during prolonged intermittent kidney replacement therapy

Supplemental material, sj-txt-1-aic-10.1177_0310057X251396229 for Cefazolin pharmacokinetics during prolonged intermittent kidney replacement therapy by Jessica H Xu, Sam Salman, David JR Morgan, Hayoung Won, Steven C Wallis, John Dyer, Jason A Roberts and Matthew DM Rawlins in Anaesthesia and Intensive Care

Footnotes

Author contribution(s)

Jessica H Xu: Conceptualization; Data curation; Investigation; Project administration; Writing – original draft; Writing – review & editing.

Sam Salman: Conceptualization; Formal analysis; Investigation; Methodology; Software; Writing – original draft; Writing – review & editing

David JR Morgan: Writing – review & editing.

Hayoung Won: Data curation; Formal analysis; Resources; Writing – review & editing.

Steven C Wallis: Data curation; Formal analysis; Resources; Validation; Writing – review & editing.

John Dyer: Writing – review & editing.

Jason A Roberts: Writing – review & editing.

Matthew DM Rawlins: Conceptualization; Project administration; Supervision; Writing – review & editing.

Declaration of conflicting interests

The authors have no conflicts of interest to declare.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Jessica H Xu

Sam Salman

Supplementary material

Supplementary material for this article is available online.

References

Barbar

Clere-Jehl

Bourredjem

, et al. Timing of renal-replacement therapy in patients with acute kidney injury and sepsis. N Engl J Med 2018; 379: 1431–1442.

Corona

Cattaneo

Latronico

Antibiotic therapy in the critically ill with acute renal failure and renal replacement therapy: A narrative review. Antibiotics 2022; 11: 1769.

Trotman

Williamson

Shoemaker

, et al. Antibiotic dosing in critically ill adult patients receiving continuous renal replacement therapy. Clin Infect Dis 2005; 41: 1159–1166.

Hoff

Maker

Dager

, et al. Antibiotic dosing for critically ill adult patients receiving intermittent hemodialysis, prolonged intermittent renal replacement therapy, and continuous renal replacement therapy: An update. Ann Pharmacother 2020; 54: 43–55.

Heintz

Matzke

Dager

WE.

Antimicrobial dosing concepts and recommendations for critically ill adult patients receiving continuous renal replacement therapy or intermittent hemodialysis. Pharmacotherapy 2009; 29: 562–577.

Duke

Parker

Zam

, et al. Population pharmacokinetics of unbound cefazolin in infected hospitalized patients requiring intermittent high-flux haemodialysis: Can a three-times-weekly post-dialysis dosing regimen provide optimal treatment? J Antimicrob Chemother 2024; 79: 2980–2989.

Sowinski

Mueller

Grabe

, et al. Cefazolin dialytic clearance by high-efficiency and high-flux hemodialyzers. Am J Kidney Dis 2001; 37: 766–776.

Tsai

Soto

Crawford

, et al. Important antimicrobial dosing considerations for transitions of care: Focus on thrice-weekly dosing in hemodialysis. Am J Health Syst Pharm 2025; 82: 203–210.

Echevarria

Traugott

KA.

ß-lactam therapy for methicillin-susceptible Staphylococcus aureus bacteremia: A comparative review of cefazolin versus antistaphylococcal penicillins. Pharmacotherapy 2017; 37: 346–360.

10.

Gatti

Viaggi

Rossolini

, et al. Targeted therapy of severe infections caused by Staphylococcus aureus in critically ill adult patients: A multidisciplinary proposal of therapeutic algorithms based on real-world evidence. Microorganisms 2023; 11: 394.

11.

Sritharan

Zhang

, et al. Clinical significance of cefazolin inoculum effect in serious MSSA infections: A systematic review. JAC Antimicrob Resist 2024; 6: dlae069.

12.

Renaud

Lin

Subramanian

, et al. High-dose cefazolin on consecutive hemodialysis in anuric patients with Staphylococcal bacteremia. Hemodial Int 2011; 15: 63–68.

13.

Ahern

Possidente

Hood

, et al. Cefazolin dosing protocol for patients receiving long-term haemodialysis. Am J Health Syst Pharm 2003; 60: 178–181.

14.

Legg

Davis

Roberts

JA.

Optimal drug therapy for Staphylococcus aureus bacteremia in adults. Curr Opin Crit Care 2023; 29: 446–456.

15.

The European Committee on Antimicrobial Susceptibility Testing (EUCAST). Breakpoint tables for interpretation of MICs and zone diameters 2024, https://www.eucast.org/bacteria/clinical-breakpoints-and-interpretation/clinical-breakpoint-tables/ (2024, accessed 9 October 24).

16.

Beal

Sheiner

LB.

NONMEM user guides (I to VIII). San Francisco, CA: University of California at San Francisco, 1998.

17.

Maynor

Carl

Matzke

, et al. An in vivo-in vitro study of cefepime and cefazolin dialytic clearance during high-flux hemodialysis. Pharmacotherapy 2008; 28: 977–983.

18.

Schliamser

Cars

Norrby

SR.

Neurotoxicity of ß-lactam antibiotics: Predisposing factors and pathogenesis. J Antimicrob Chemother 1991; 27: 405–425.

19.

Kingra

Ariano

Sharma

Seizures associated with high-dose cefazolin in a patient with renal dysfunction: A case report. J Pharm Pract 2025; 38: 529–533.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB