Minor Fluctuations in Renal Function May Alter Therapeutic Drug Concentrations Substantially during High-Dose,Continuous-Infusion Beta-Lactam Therapy for Multi-Drug-Resistant Gram-Negative Bacilli

To the Editor:

A 76-year-old male with chronic kidney disease secondary to hypertension, atrial fibrillation, cerebrovascular disease with transient ischemic attacks, and morbid obesity underwent an elective laparoscopic left partial nephrectomy for a renal mass. Immediately after surgery, the patient developed acute respiratory dysfunction and superimposed acute kidney injury, and he was transferred to the surgical intensive care unit for endotracheal intubation and mechanical ventilation.

On postoperative day (POD) 8, while still intubated, the patient developed recurrent hypoxemia; evaluation by bronchoalveolar lavage (BAL) and quantitative microbiology diagnosed Pseudomonas aeruginosa ventilator-associated pneumonia (VAP). The organism had a minimum inhibitory concentration (MIC) of 4 mcg/mL for meropenem, and treatment was initiated under our antibiotic rotation program [1,2] with meropenem 1 g q8h via 3-h infusion for a planned 14-day course. On POD 18 (treatment day 10), the patient's respiratory status worsened further; repeat BAL revealed a multi-drug-resistant (MDR) P. aeruginosa VAP (Table 1). Tracheostomy was performed.

At the time, his serum creatinine concentration was 2.5 mg/dL, increased from a baseline of 1.6 mg/dL (glomerular filtration rate 25 mL/min, estimated by creatinine clearance measured by a 12-h timed urine collection) [3]. To achieve a steady-state serum meropenem concentration >16 mcg/mL, the dose of the drug was increased to 6 g/day by continuous infusion, having achieved that concentration previously, with safety, during treatment of a bla(kpc-2)-positive Klebsiella pneumoniae blood stream infection in another patient with acute kidney injury [4]. Given the clinical deterioration and the resistance phenotype, a two-drug regimen was chosen to maximize the likelihood of giving appropriate therapy. Considering the potential nephrotoxicity of both drugs, polymyxin B 90 mg IV q 12 h was chosen rather than an aminoglycoside. Synergy studies confirmed no antagonism between the meropenem and polymyxin B [5]. Dosing of polymyxin B in renal insufficiency is not well described [6], but reductions may not be necessary, as believed previously. Moreover, retrospective data suggest that polymyxin B monotherapy of Pseudomonas infections may be inferior to comparators [7].

During therapy, the patient's renal function improved minimally. Random serum samples were drawn on PODs 23–26 to determine meropenem serum concentrations by a validated high-performance liquid chromatography method [8] at the Center for Anti-Infective Research and Development (Hartford, CT), along with serial daily re-assessments of kidney function. Assay for polymyxin B was unavailable. Free (unbound) serum meropenem concentration, assuming protein binding of 2% [9], decreased from 32.83 mcg/mL to 15.93 mcg/mL concurrent with the minimal improvement in renal function (Table 2). No dosage adjustments were made, as the patient was responding to therapy. Although subsequent BAL samples demonstrated colonization of the patient's airway with MDR P. aeruginosa, the patient completed 24 days of meropenem therapy and 14 days of polymyxin B and was liberated from the ventilator. His tracheostomy was decannulated, and he was discharged from the hospital.

Pseudomonas aeruginosa, a prevalent and virulent opportunistic pathogen, is known to develop an MDR phenotype during treatment [9,10]. Surveillance data from the United States note that MDR Pseudomonas has increased over the last decade, with resistance especially to fluoroquinolones, third-generation cephalosporins, and carbapenems [11]. Effective treatment must involve thoughtful application of agent-specific pharmacokinetics. Aminoglycosides and polymyxins exhibit concentration-dependent killing and eradicate bacteria most effectively with a peak serum concentration several-fold higher than the MIC for the organism [12,13]. By contrast, beta-lactam antibiotics exhibit time-dependent killing and eradicate bacteria when the free concentration of the drug is higher than the bacterial MIC for a sufficient portion of the dosing interval [14]. Accordingly, continuous or prolonged infusion has shown promise over conventional intermittent dosing for achieving pharmacodynamic targets and was successful for treatment of Pseudomonas pneumonia in a randomized trial [15].

Antibiotic dosing is especially challenging with fluctuating renal function. Clinical trials of continuous-infusion beta-lactams typically have excluded patients with renal insufficiency [16]; fluctuating function adds further complexity. Two studies have examined meropenem in patients receiving renal replacement therapy, estimating that dosing of 1 g q 12 h is adequate for these patients [17,18], but it is unknown specifically whether MDR pathogens were being treated. We have reported success with high-dose, continuous-infusion meropenem for carbapenemase-producing Klebsiella pneumoniae bacteremia in a patient with stable moderate acute kidney injury, achieving a steady-state serum meropenem concentration >20 mcg/mL [4]. However, it has not been described previously how changing renal function affects the serum concentration of antibiotics dosed continuously. The present data suggest that even modest recovery of renal function during treatment may decrease the effective circulating concentration of the antibiotic meaningfully, which could lead to treatment failure as renal function improves. To accommodate changes in renal function, clinicians who use continuous- or prolonged-infusion beta-lactam antibiotics must re-evaluate the treatment regimen constantly. Real-time drug concentration monitoring and frequent assessments of renal function may be crucial for effective alterations.