Pharmacokinetics and Kidney Clearance: A Nephrologist's Role in Drug Safety and Toxicology

The kidneys play a vital role in eliminating drugs and their metabolites, making it essential for nephrologists to understand pharmacokinetic changes in kidney injury and diseases. Drugs may be excreted by the kidney by glomerular filtration or by tubular secretion via active drug transporters. They may also be reabsorbed from the filtrate across the renal tubular epithelial lining, usually by passive diffusion. Total clearance is the sum of glomerular filtration and tubular secretion minus tubular reabsorption (1). The proximal tubule is the site of elimination for small molecules such as antibiotics, antivirals, diuretics, nonsteroidal anti-inflammatory drugs, and antidiabetic agents (2).

Nephrologists are uniquely positioned to assess kidney function and provide guidance on appropriate drug dosing through daily review of medications and documentation in the problem list (a summary of the patient's relevant clinical information). By evaluating a patient's glomerular filtration rate (GFR) and considering other factors such as age, weight and muscle mass, and comorbidities, nephrologists can recommend the most appropriate biomarkers to order for GFR estimation, provide their estimate of a patient's GFR, and collaborate with pharmacists to optimize the dose to maximize therapeutic efficacy while minimizing toxicity (3). For example, consider a patient with chronic kidney disease (CKD) taking the antiretroviral combination agent dolutegravir/tenofovir alafenamide/emtricitabine, in which dolutegravir interferes with tubular secretion of serum creatinine (SCr) leading to elevations in baseline SCr. In this situation, serum cystatin C may be ordered to better estimate GFR.

In CKD, as GFR declines, changes in pharmacokinetics occur (1). Protein binding may be altered due to conformational changes in albumin-binding sites or occupation of those sites by uremic toxins, impairing binding and increasing the free drug concentration of drugs like phenytoin. Alterations in the drug metabolism have been seen in CKD due to a decreased intrarenal and hepatic metabolism and are generally proportional to the reductions in GFR. In transplant recipients, calcineurin inhibitors are metabolized by cytochrome P450 3A4, and drug–drug interactions can lead to changes in drug exposure. Common interacting drug classes include calcium channel blockers, azole antifungals, rifampin, macrolide antibiotics, and anticonvulsants. Lastly, drugs eliminated by the kidney will require dose adjustment as GFR declines. Drugs such as gabapentin, baclofen, or cefepime are eliminated by glomerular filtration, necessitating dose adjustment to avoid accumulation and adverse effects such as drowsiness, falls, edema (for gabapentin), and neurotoxicity (for baclofen and cefepime) (4–6). The dose of baclofen should be reduced by 50% for CKD stage 3 and avoided in stages 4–5. Risk of cefepime-associated neurotoxicity increases in CKD stage 4 and is minimized with therapeutic drug monitoring keeping trough concentrations <7.5 mg/L (6).

Acute kidney injury (AKI) can significantly alter drug pharmacokinetics (7). In critical illness, absorption is impaired due to changes in gastrointestinal pH, perfusion, transit time, and intestinal atrophy, resulting in lower drug exposure. Fluid shifts, volume overload, and changes in protein binding can alter drug distribution. In patients who are critically ill, the measured volume of distribution (Vd) can differ from the estimated Vd, as massive hemorrhage, ascites, and other fluctuations in the fluid status commonly occur. Many patient-related factors can alter protein binding, including hypoalbuminemia, systemic pH, hyperbilirubinemia, uremic products, heparin therapy, and other drugs that may act as competitive displacers. In AKI, cytochrome P450 expression is variable, and the degree of reduction in the drug metabolism does not appear to be as great in AKI as in kidney failure (8). Many patients may recover within 48 hours (9), making dose adjustments sometimes unnecessary. Consider temporarily discontinuing certain medications, reducing doses, or switching to alternative therapies until kidney function improves.

Estimating GFR in AKI is challenged by inaccurate estimating equations; nonsteady state equations, measured iohexol clearance, or timed creatinine clearances (CrCls) can be used (10). In critical illness, SCr may be inaccurate, especially in patients with cachexia, muscle wasting, or a prolonged intensive care unit stay; SCr may be lower than expected, leading to an overestimation of kidney function (11). In these cases, nephrologists may need to order cystatin C to identify patients at risk for erroneous dosing, necessitating additional therapeutic drug monitoring. Cystatin C has its own limitations and may be impacted by glucocorticoid use, cancer, thyroid status, inflammation, and obesity (3). Lastly, augmented renal clearance is measured as CrCl > 130 mL/min/1.73 m² and is seen in 20%–65% of younger patients who are critically ill with trauma, major surgery, or severe burns. Prompt recognition of augmented renal clearance is paramount to ensuring adequate antimicrobial dosing and avoiding subtherapeutic concentrations (12).

The extent of drug removal by renal replacement therapy (RRT) depends on drug factors such as molecular weight, protein binding, and Vd (7). Most drugs have a molecular weight <500 Da and can be removed by conventional hemofilters with a cutoff of 1500 Da (13). Only the unbound fraction of a drug is cleared by dialysis, and the degree of protein binding is the most important factor to determine whether a drug needs dose adjustment. A drug that is >90% bound to plasma proteins (e.g., ceftriaxone and warfarin) is unlikely to be removed by dialysis; however, it is very likely to be removed by pheresis. The drug sieving coefficient is the capacity of the drug to pass the membrane by convection and can be estimated by the fraction unbound (13). The sieving coefficient is then used to determine the drug clearance (Table). A large Vd reflects distribution outside of the plasma space, resulting in low removal by dialysis and pheresis. Vd is the most important determinant for removal by pheresis therapy. A drug with Vd <1 L/kg is more likely to be removed by RRT; <0.3 L/kg, likely to be removed by pheresis; and >2 L/kg, less likely to be removed by RRT or pheresis (13). Finally, water-soluble drugs are more likely to be removed by dialysis. Antimicrobials such as ceftazidime, ceftriaxone, vancomycin, or aminoglycosides should be administered after pheresis procedures if feasible (14). The important therapy-specific factors include RRT modality, blood and effluent flow rates, and hemofilter characteristics. Continuous RRT may result in more consistent and prolonged drug removal compared with intermittent hemodialysis (IHD), whereas IHD may be favored in intoxications to remove drugs more rapidly. In peritoneal dialysis, the peritoneal membrane's permeability and the drug's physicochemical properties influence removal with conditions such as peritonitis, leading to alterations in drug transport. Therapy-dependent factors such as blood flow rate, effluent flow rates, and administration of replacement fluids before or after filter will affect drug clearance (Table). Lastly, high-flux, high-efficiency dialyzers will have the greatest capacity for removal and can remove drugs with larger molecular weight (e.g., vancomycin, daptomycin). The estimated CrCl for the type of dialysis is summarized in the Table.

Table

Estimated drug clearance by renal replacement therapy

View Table

Loading doses should be adjusted based on changes in Vd (1). For example, in volume overload, the loading dose of hydrophilic drugs may need to be increased to achieve therapeutic concentrations (13). In the case of digoxin, which has a large Vd, the Vd is altered in CKD, requiring different loading doses. Maintenance doses are adjusted depending on drug pharmacokinetic and dynamic factors (13). For drugs in which a therapeutic effect should be maintained throughout the interval, the dose is reduced, and the interval is maintained (e.g., cephalosporins and time above minimum inhibitory concentration). Sampling blood concentrations should account for removal by the dialysis procedure and re-equilibrium between plasma and tissue compartments (e.g., aminoglycosides, vancomycin). Consider sampling after re-equilibrium is established.

Nephrologists play a vital role in collaborating with other health care professionals and educating patients about their estimated GFR and the optimal selection of drugs and doses to maximize efficacy and to reduce adverse effects.