Antimicrobial dosing in critically ill patients with sepsis-induced acute kidney injury

Abstract

Severe sepsis often leads to multiple organ dysfunction syndromes (MODS) with acute kidney injury (AKI). AKI affects approximately, 35% of Intensive Care Unit patients, and most of these are due to sepsis. Mortality rate of sepsis-induced AKI is high. Inappropriate use of antimicrobials may be responsible for higher therapeutic failure, mortality rates, costs and toxicity as well as the emergence of resistance. Antimicrobial treatment is particularly difficult due to altered pharmacokinetic profile, dynamic changes in patient's clinical status and, in many cases, need for renal replacement therapy. This article aims to describe the appropriate antimicrobial dosing and reviews the factors contributing to the difficulties in establishing precise guidelines for antimicrobial dosing in sepsis-induced AKI patients. Search strategy: Text material was collected by systematic search in PubMed, Google (1978–2013) for original articles.

Introduction

Sepsis is a common heterogeneous clinical entity that is defined by the physiological changes collectively known as a systemic inflammatory response syndrome, which occurs in response to a presumed infectious etiology.[1] Severe sepsis and septic shock are frequent reasons for patient's admission to Intensive Care Units (ICU). In septic shock, patients fail to maintain their blood pressure despite adequate hydration. Severe sepsis is defined as sepsis plus sepsis-induced organ dysfunction or tissue hypoperfusion.[2] Severe sepsis often leads to multiple organ dysfunction syndromes (MODS) with acute kidney injury (AKI).[3] AKI affects approximately, 35% of ICU patients,[4] and around 50% of these are due to sepsis.[5] AKI has an overall mortality rate of 45%, mortality rate of sepsis-induced AKI is much higher, at over 70%.[4,6] A study from the USA by Angus and Wax[7] has reported that there are approximately, 7.5 lakh new cases of severe sepsis annually, with an economic impact approaching $17 billion, and it is the 3^rd leading cause of infectious death and the 10^th leading cause of death overall. There is a paucity of data in India regarding the true incidence and prevalence of AKI. In a study from North India, Kohli et al.,[8] reported the incidence of hospital-acquired AKI was 2.1/1000 admissions and the incidence of community-acquired AKI (CAAKI) was 6.6/1000 admissions. Similarly in recent study by Kaul et al. from North India, the prevalence of sepsis-induced CAAKI was 13.9% and overall mortality rate among patients with CAAKI was 26.2% but sepsis-induced CAAKI had the highest mortality. Majority of patients with CAAKI required dialysis mainly hemodialysis.[9] Thus, it has an enormous impact on resource depleted ICUs in developing the country like India. The only solution of such huge problem is an early institution of appropriate resources. Recent published Surviving Sepsis Campaign (SSC) guideline hoped that over time, particularly through education programs and feedback performance improvement initiatives, the guideline will influence bedside health care practitioner behavior that will reduce the burden of sepsis worldwide.[2] Appropriate antimicrobial treatment in terms of spectrum of activity or dose and frequency of administration will result in better outcomes in such patients.[10] However, because of various alterations in the pharmacokinetics (PK) of antimicrobials during sepsis, standard antimicrobial regimens can result in subtherapeutic serum drug concentrations in such patients.[11,12]

Disturbances of Renal Function in Critically Ill Patients

Critically ill-patients are diagnosed with various stages of AKI. Kidney Disease: Improving Global Outcomes have recently published new stages of AKI and their diagnostic criteria [Table 1].[13] Approximately, two-third of patients are diagnosed within the first 24 h after admission to the ICU.[6,14] It is emphasized that disturbances of renal function are not limited to glomerular filtration rate, but also affect the process of tubular secretion and reabsorption.

Table 1.

Stages of acute kidney injury according to KDIGO

Need for Individualized Antimicrobial Dosing in Sepsis-Induced Acute Kidney Injury

In addition to the in vitro susceptibility of the isolated strains and timely antimicrobial administration,[15] antimicrobial efficacy is dependent on the serum and tissue concentrations of the agent used.[16] Sepsis significantly alters the PK of antimicrobial agents, including increasing the volume of distribution (Vd), protein binding and drug clearance. The effect of hypoproteinemia, organ dysfunction and the presence of augmented renal clearance may lead to unexpectedly high or low plasma antimicrobial concentrations.[17] The problem of optimal antimicrobial doses becomes even more complex when there is concomitant renal failure because drug clearance is reduced, and accumulation of antimicrobials in the blood and tissues may potentially contribute to increased adverse side effects.[16]

Pharmacokinetics of antimicrobials in a heterogeneous group is altered to varying extent compared with a healthy population and their clinical state and drug PK can fluctuate significantly on the day to day basis. Therefore, indicators routinely employed in designing the antimicrobial regimen in individuals without organ dysfunction are entirely inadequate. In such patient, these disparities can result in inappropriate antimicrobial treatment.

Antimicrobial agents are a group of drugs with “silent” pharmacodynamics (PD) (i.e. the pharmacologic effect is not perceivable immediately after administration), it is almost impossible to assess whether therapeutic concentrations are being achieved during the early phase of therapy. Therefore, situations likely to alter antimicrobial PK [Figure 1] and necessitate dosage adjustment are necessary to enable the individualization of antimicrobial therapy.[18]

Figure 1.

Clinical scenarios likely to alter antimicrobial pharmacokinetics in Multiple Organ Dysfunction Syndrome[18]

Factors Affecting Antimicrobial Dosing in Patients with Acute Kidney Injury

Factors contribute to the difficulties in establishing precise guidelines for antimicrobial dosing in critically ill-patients with AKI [Table 2] are mainly: (1) Patient related (2) hemofiltration-related and (3) drug-related variables.

Table 2.

Factors contributing to the difficulties in establishing precise guidelines for antimicrobial dosing in patients with acute kidney injury

Patient-related variables

Most of the antimicrobials are acidic and protein binding is often significantly altered in critical illness due to the fall in serum albumin, decreased systemic pH and the presence of uremic toxins, bilirubin and free fatty acids, all of which may be present in renal failure and sepsis.[19,20,21] Most antimicrobial agents are eliminated via the kidney, and therefore a significant reduction in creatinine clearance may result in an extensive prolongation of the half-life of some antimicrobials. Hepatic metabolism and biliary or gut excretion may substantially increase in the presence of renal failure. Sepsis causes the damage of vascular endothelium with an increase of capillary permeability and redistribution of fluid into the extracellular compartment. As a result, Vd of water soluble antimicrobials increases with a subsequent drop in their concentration to the subtherapeutic level.

Hemodialysis and hemofiltration-related variables

Continuous venovenous hemofiltration (CVVH) removes plasma water by producing an ultrafiltrate and clears molecules of varying sizes by convection - by dragging molecules with the fluid. This process of the molecular clearance is influenced by:

• The sieving coefficient of the molecules removed:

  • Sieving coefficient is defined as the concentration of drug in the ultrafiltrate divided by mean of concentrations in pre-and post-filter blood and it reflects the capacity of a drug to pass through a hemofilter membrane. It varies from 0 (drugs that do not pass) to 1 (drugs that pass freely).

• The ultrafiltration rate:

  • In addition, drug clearance is directly proportional to the ultrafiltration rate; a higher proportion of the drug is removed at higher filtration rates.

• The proportion of replacement fluid given predilution:

  • Transfer of drug across the filter membrane depends on the concentration of drug. Infusion of a proportion of the total replacement fluid before the filter (predilution) may decrease local concentration and results in a decrease in drug clearance.

• Membrane characteristics:

  • Use of large surface area membranes and frequent changes of the filter membrane will increase the amount of drug being removed. Solute-membrane interaction, leading to the formation of plasma protein layers on the membrane and reduce its permeability.[22]

Drug-related variables

Several drug factors [Figure 2] play an important role in determining the final amount of drug removed by hemofiltration, notably:

Figure 2.

Drug related factors affecting antimicrobial dosing in critical ill patients

• The molecular weight of the drug

• Protein binding

• The degree of renal clearance.

Molecular size influences drug clearance, as the contribution of convective transport relative to diffusion increases with increasing molecular weight of the drug.

Dosage Adaptations

Commonly used drug-dosing technique involves calculating the total creatinine clearance rate by adding any estimated residual renal creatinine clearance to the expected extracorporeal creatinine clearance. The extracorporeal creatinine clearance rate can be assumed to be approximately equivalent to the dialyzate, ultrafiltrate, or effluent rate, and medication dosing guidelines specified for the total creatinine clearance can be used to guide dose selection. This method assumes that drugs only undergo glomerular filtration, not tubular secretion or reabsorption.[23] For drugs that do undergo tubular secretion, this method could lead to increased drug exposures, and in patients with impaired reabsorption, underdosing can potentially occur. Thus, drug dosing technique on the basis of creatinine clearance rate is not effective.

Pharmacokinetics and Pharmacodynamic of Antimicrobials in Sepsis-Induced Acute Kidney Injury Patients

Intrinsic PK and PD properties are the major determinants of in vivo efficacy of antimicrobial agents.[24] The PK and PD properties of each antimicrobial agent and the typical susceptibilities of relevant pathogens are considered in Table 3.[25,26,27,28,29,30]

Table 3.

Pharmacokinetic and pharmacodynamic parameters of antimicrobial agents[25,26,27,28,29,30]

Pharmacokinetics

Pharmacokinetics is the study of the interrelationship between drug dose and variations in concentrations in plasma and tissue over time. The most relevant PK parameters [Figure 3] include the following.[31]

Figure 3.

Basic pharmacokinetic and pharmacodynamic parameters

• Cmax: Peak concentration achieved after a single dose

• Cmin: The lowest (trough) concentration that a drug reaches before the next dose is administered

• Vd: The apparent volume of fluid that contains the total drug dose administered at the same concentration as in plasma

• Clearance (CL): Quantification of the irreversible loss of drug from the body by metabolism and excretion

• Elimination half-life: Time required for the plasma concentration to fall by one-half

• Protein binding: Proportion of drug binding to plasma proteins

• Area under the curve (AUC) 0-24: Total area under the concentration curve over 0-24 h.

Pharmacodynamics and pharmacokinetics/pharmacodynamics models

Pharmacodynamics is the study of the relationship between drug concentrations and effect [Figure 4].[31] Several PK/PD models have been constructed using the three most popular parameters: Cmax/minimum inhibitory concentration (MIC), %T > MIC, and AUC24/MIC.

Figure 4.

Interrelationship among pharmacokinetics, pharmacodynamic, and pharmacokinetics/pharmacodynamics[31]

• Cmax/MIC: How many times the peak serum concentration of a given antimicrobial is higher than MIC

• %T > MIC: Percentage of a dosage interval in which the serum drug concentration remains above the MIC

• AUC24/MIC: Area under the concentration curve over 0-24 h-to-minimum inhibitory concentration ratio.

Classification of Antimicrobials Based on Pharmacokinetics-Pharmacodynamics Models Associated with their Optimal Killing Activity[32,33,34,35]

Time-dependent antimicrobial agents

Optimal activity is achieved when unbound plasma concentrations are maintained above the MIC of the bacteria for the longest period, and %T > MIC is the best predictor of their efficacy. Time - dependent antimicrobials, including cephalosporins, carbapenems, and penicillins. After administering the loading dose, timedependent antimicrobials should be readministered in several lower doses per 24 h.[32]

Concentration-dependent antimicrobial agents

Concentration - dependent antimicrobials, including aminoglycosides, fluorochinolones, daptomycin, amphotericin B, should be administered in high doses once per 24 h in order to obtain optimal activity of Cmax/MIC to maximize killing, followed by very low troughs to minimize toxicity.[33]

Time and concentration dependent antimicrobial agents

AUC24/MIC is the most reliable predictor of antimicrobial (vancomycin, linezolid, tetracyclines, azithromycin) efficacy, and it is also related to the type of the pathogen involved.[34,35]

Antimicrobial Dosing in Sepsis-Induced Acute Kidney Injury on Renal Replacement Therapy

In patients with sepsis, sustained oliguria or severe metabolic acidosis, refractory volume overload and severe electrolyte disarray may be the reasons enough to start renal replacement therapy (RRT).[36] In sepsis-induced AKI, therapeutic antimicrobial drugs are often required, but standard dosing regimens are affected by RRT. Continuous renal replacement therapy, particularly CVVH, is becoming more commonly used in the routine management of critically ill patients with AKI.[37] These modalities [Table 4] may change dosing of antimicrobial agents.[38] The SSC recommends that intravenous antimicrobials are begun within the 1^st h after diagnosis of severe sepsis and septic shock.[39] Dosage of antimicrobial by type of RRT are showed in Table 5.[18,26,30,40,41,42,43,44,45,46,47,48,49,50]

Table 4.

Modalities of renal replacement therapy[38]

Table 5.

Dose adjustments of selected intravenous antimicrobials in patients with renal dysfunction and hepatic failure[18,26,30,40,41,42,43,44,45,46,47,48,49,50]

Dosing of Antimicrobials in Sepsis Related Multiple Organ Dysfunction Syndromes

In sepsis-related MODS, homeostasis cannot be maintained without intervention, usually involving two or more organ systems.[51] Hemodynamic alterations lead to sepsis-induced tissue hypoperfusion, which affect PK leading to inadequate dosing of antimicrobials [Figure 1]. Several pathophysiological conditions may increase Vd and an increase in dosage should be considered with the intent of ensuring optimal care [Figure 5]. [52] Dose recommendations and general principles for loading and maintenance dosing of antimicrobial agents [18,53,54,55] in patients with renal failure and hepatic failure are showed in Tables 5 and 6.

Figure 5.

Schematic representation of the pathophysiological or iatrogenic conditions in patients affecting drug distribution and elimination[52]

Table 6.

Antibiotic dosing adjustment for critically ill patients with multiple organ involvement[18,53,54,55]

Conclusion

Sepsis is a common and remains a major cause of multiorgan dysfunction syndrome, indicating a crucial role in efficient antimicrobial treatment. Inappropriate use of antimicrobials may be responsible for higher therapeutic failure, mortality rates, costs and patient toxicity as well as the emergence of resistance. Antimicrobial treatment is particularly difficult due to altered PK profile, dynamic changes in patient's clinical status and, in many cases, need for RRT. Instructions on antimicrobial dosing in package inserts provided by drug manufacturers are frequently insufficient to guide dosing in the critically ill patients appropriately and current guidelines concerning the dosing of antimicrobials are not particularly reliable because they are based on studies involving small and heterogeneous groups of patients, often treated with different RRT modalities.

Shortage of reliable data regarding antimicrobial dosing in critically ill patients possesses an enormous clinical dilemma. Consequently, there is an urgent need for establishing new sets of recommendations corroborated by large-scale prospective clinical studies conducted in homogenous patient populations treated according to the uniform RRT procedures.