Augmented Renal Clearance (ARC)
Augmented renal clearance (ARC) is a term used when a patient’s creatinine clearance (CrCL) reaches or exceeds 130 mL/min/1.73 m2 [1,2]. Commonly, renal impairment is studied in drug development. In contrast, ARC is when drug elimination is accelerated instead of reduced. Among critically ill patients whose plasma creatinine is normal, ARC is most common in the young and after trauma. Age at or below 50 years, a trauma diagnosis, and a modified SOFA score at or below 4 were the independent risk factors in a cohort of 71 septic and multi-trauma patients, where trauma patients met the threshold far more often than septic ones, 85.7% against 39.5% [3]. Men were over-represented among those with ARC, but sex did not survive multivariate adjustment, so trauma rather than sex is the signal to carry. Across four ICUs and 281 patients admitted with a normal plasma creatinine, 65.1% showed ARC on at least one of the first seven days [1]. Critical illness can be caused by e.g. burns or infections. Furthermore, critically ill patients often have rapidly fluctuating renal function.
Standard CrCL estimation equations like Cockcroft-Gault, MDRD, or CKD-EPI, are inaccurate at these high CrCL values [2,4]. Furthermore, standard CrCL estimation equations were developed for steady-state conditions.
The main concern is that the increased renal function will lead to suboptimal therapy with drugs that are primarily eliminated by renal excretion [2]. Most notably, in the context of renally excreted antimicrobials like aminoglycosides. The same holds for β-lactams: across a cohort of critically ill patients, 42% had initial trough concentrations below the MIC, and among those with a measured creatinine clearance at or above the 130 mL/min/1.73 m2 threshold used here, 82% did [5]. Colistin shows the same pattern. Among critically ill patients on a fixed loading dose followed by 4.5 MU twice daily, plasma concentrations ten hours after a maintenance dose exceeded 2 mg/L in 94% of patients with a creatinine clearance at or below 120 mL/min, but in only 44% of those above it [6]. That study split at 120 mL/min uncorrected rather than the body-surface-indexed threshold used here, and I am a co-author on it. How that exposure is referenced to the MIC in the first place is covered under antibacterials for systemic use. Patients with ARC are younger, more often male, and have less organ dysfunction than patients with lower creatinine clearances, so their baseline risk profile is the more favorable one [7]. That does not translate into better measured outcomes. In a nested cohort of 254 severe sepsis patients from the BLING-II trial, the 45 (17.7%) with a day 1 creatinine clearance at or above 130 mL/min had higher unadjusted clinical cure, 73.3% against 55.0%, but the difference was attenuated on multivariable adjustment, and neither alive ICU-free days at day 28 nor 90-day mortality differed [7].
In a study by Andrew Udy et al., 17 of 20 patients (85%) with traumatic brain injury managed with hypertonic saline and/or norepinephrine to maintain cerebral perfusion pressure had augmented clearance [8]. Note that study set the bar higher than the 130 mL/min/1.73 m2 used above, at >150 mL/min/1.73 m2 in women and >160 in men, so the 85% is against a stricter definition than this page’s. A timed urinary creatinine clearance comes closer to the measured value than any estimating equation, but it is not exact either and it cannot give an immediate answer [4]. Brown et al. measured creatinine clearance serially in 50 critically ill postoperative patients [9]. In the trauma patients it was initially higher than expected, 140 to 190 mL/min/1.73 m2, so 190 is the top of that early range rather than a typical value. Fuster-Lluch et al. found glomerular hyperfiltration in 17.9% of 89 critically ill patients on admission, rising to as high as 30% during the first week [10]. That study is from 2008 and reports hyperfiltration rather than ARC as defined here, so the two rates are not directly comparable.
Cytokine release from acute injury, the body’s immune and inflammatory responses to trauma, and aggressive fluid resuscitation may all contribute to increased organ (renal) blood flow and enhanced excretory function.
Understanding ARC is crucial, as it impacts the dosing of renally excreted drugs, potentially making standard dosing regimens insufficient for patients with ARC. Baptista et al. compared measured creatinine clearance against Cockcroft-Gault and the 4- and 6-variable MDRD estimates in 86 critically ill patients with ARC, using the same 130 mL/min/1.73 m2 threshold as here [11]. Every equation significantly underestimated the measured value, Cockcroft-Gault least of all, and none was precise enough for clinical use. CKD-EPI was not among the equations tested.
The key challenges in estimating GFR in ICU patients [4]:
- Rapidly changing renal hemodynamics
- Risk of both overdosing and underdosing due to inaccurate GFR estimates
- Constantly fluctuating patient volume status
- ARC is often overlooked in GFR calculations, leading to inadequate drug dosing
- Common creatinine-based equations are unreliable in critical care settings, and ideal methods like inulin clearance are impractical in the ICU.
Ways to deal with ARC in PK modeling
Introduce a scalar above 130 mL/min. The idea is that this will estimate, and correct for, the bias of estimating equations in ARC. Of course, this assumes that the bias is linear.