PHARMACOKINETICS OF POLYMYXIN B IN AN INFANT WITH MULTIDRUG-RESISTANT KLEBSIELLA PNEUMONIAE BACTEREMIA

Abstract

We report our experience with a 9-month-old infant treated with intravenous polymyxin B for multidrug-resistant Klebsiella pneumoniae bacteremia. Serial blood samples were obtained at steady state and serum drug concentrations were determined using a validated liquid chromatography-mass spectrometry method. The elimination half-lives of polymyxin B1 and isoleucine-polymyxin B1 were found to be 3.1 and 4.7 hours, respectively.

An emerging mechanism of multidrug resistance in Enterobacteriacae spp. is the serine-based carbapenemases (eg, KPC). It is of great concern as viable therapeutic options are severely limited.¹ Polymyxin B is increasingly used in this clinical situation, despite a limited understanding of its pharmacologic properties.² There is a paucity of polymyxin B pharmacokinetic data in both adults and children, and optimal dosing of polymyxin B in pediatric patients is unknown. We report our pharmacokinetic observations in a 9-month-old infant treated with intravenous polymyxin B.

CASE REPORT

A 9-month-old girl, born at 24 weeks of gestational age was admitted to the pediatric intensive care unit for treatment of acute hypoxemic respiratory failure. She was a resident of a long-term care facility due to multiple complications of prematurity, including grade I intraventricular hemorrhage, chronic lung disease, gastroesophageal reflux, tracheolaryngomalacia requiring tracheostomy, and volvulus with small bowel obstruction requiring bowel resection and ileostomy.

On admission, she had fever (39.2°C), hemodynamic instability, and significant oxygen requirements with 100% FiO₂ to maintain oxygen saturation above 88%, necessitating therapy with high frequency oscillatory ventilation. A chest radiograph showed bilateral hazy opacities that could have represented bronchial aspiration. Laboratory results included white blood cell count (12.8 × 10³/mm³ with 59% neutrophils and 24% bands), elevated C-reactive protein (16.8 mg/dL), normal creatinine, and hypoalbuminemia (2.4 g/dL). Blood and urine cultures were negative. On day 2 of hospitalization, a bronchoalveolar lavage was performed and cultures were negative for bacteria and fungi. However, a culture from an endotracheal aspirate yielded Acinetobacter baumannii susceptible to piperacillin/tazobactam (minimal inhibitory concentration [MIC] = <4 μg/mL). She completed a 10-day course of piperacillin/tazobactam therapy with significant improvement in respiratory status.

On day 14 of hospitalization, she had recurrence of fever (39°C) with elevated white blood cell count and C-reactive protein value. No increase in the respiratory requirements or hemodynamic instability was noted. An endotracheal culture yielded Stenotrophomonas maltophilia and Pseudomonas aeruginosa. Urine culture was negative. The blood culture grew multidrug resistant Klebsiella pneumoniae. Initial empiric treatment consisted of intravenous meropenem. The Klebsiella was resistant to piperacillin/tazobactam (MIC = >128 μg/mL, resistant-R), ampicillin/sulbactam (MIC = >32 μg/mL, R), aztreonam (MIC = >64 μg/mL, R), ceftazidime and ceftriaxone (MIC = 16 μg/mL, R), meropenem (MIC = 32 μg/mL by E-test, R), and susceptible to polymyxin B (MIC = 1.5 μg/mL by E-test, sensitive-S) and amikacin (MIC = <2 μg/mL, S). On the basis of these studies, meropenem was discontinued and intravenous polymyxin B at 1.5 mg/kg/dose every 12 hours (weight of 4 kg) in combination with amikacin once daily was started. She remained persistently febrile and the blood cultures were persistently positive for 3 days (total days of bacteremia were 6). As a result, the dose of polymyxin B was increased to 2 mg/kg/dose every 12 hours. An echocardiogram was negative for vegetations. Cerebrospinal fluid values were normal and the culture was sterile. She defervesced within 24 hours and completed a 2-week course of polymyxin B. Subsequent blood cultures were sterile. Her inflammatory indices normalized within 10 days and her renal function remained normal.

METHODS AND RESULTS

Polymyxin B sulfate (lot number 204475; APP Pharmaceuticals, Schaumburg, IL) was given as an intermittent intravenous infusion initially at 1.5 mg/kg every 12 hours infused during 90 minutes, and subsequently 2 mg/kg every 12 hours in 60 minutes. Serial blood samples were obtained at steady states (after 3 doses). The serum samples were assayed in duplicate for polymyxin B concentrations using a validated liquid chromatography-tandem mass spectroscopy method³; which was modified to detect various major polymyxin B components concurrently.⁴ The proportions of polymyxin B1 and isoleucine polymyxin B1 in polymyxin B (polymyxin B sulfate) were assumed to be 73.5% and 8.6%, respectively, as previously reported.⁵ A one-compartment linear model was fit to the concentration-time profiles to derive the best-fit pharmacokinetic indices. All model fittings were performed with the ADAPT II program.⁶ Using the best-fit parameter estimates, elimination half-life was derived by Ln 2 divided by eliminating rate constant, clearance was derived by elimination rate constant times volume of distribution, and area under the concentration-time curve over 24 hours was derived by daily dose divided by clearance. Serum protein binding of polymyxin B was not determined.

Both polymyxin B1 and isoleucine polymyxin B1 (>80% of total polymyxin B content) were satisfactorily detected in the serum samples. The overall model fits to the data were reasonable (r² ≥0.92, Fig., Supplemental Digital Content 1, http://links.lww.com/INF/A710), and the best-fit pharmacokinetic indices are as shown in Table 1.

TABLE 1.

Pharmacokinetic Indices of Polymyxin B1 and Isoleucine Polymyxin B1

	Polymyxin B1	Isoleucine Polymyxin B1
Ke (h−1)	0.2258	0.1462
T1/2 (h)	3.07	4.74
V (L)	1.345	1.645
V (L/kg)	0.336	0.411
Cl (L/h)	0.304	0.241
AUC24 (mg.h/L) (1.5 mg/kg q12 h)	29.042	4.291
AUC24 (mg.h/L) (2 mg/kg q12 h)	38.722	5.721
Model fit (r2)	0.98	0.92

The proportions of polymyxin B1 and isoleucine polymyxin B1 in polymyxin B (USP) were assumed to be 73.5% and 8.6%, respectively.

Ke indicates eliminating rate constant; T_1/2, elimination half-life; V, volume of distribution; Cl, clearance; AUC₂₄, area under the concentration-time curve over 24 hours.

DISCUSSION

Polymyxin B is increasingly used for the management of difficult-to-treat infections. However, dosing of polymyxin B is mostly based on convention or anecdotal experience, rather than the results of well-designed pharmacokinetic studies. The conventional weight-based dosing strategy in adults may not be directly applicable in infants, as infants often have a higher percentage of total body water. For a hydrophilic drug, lower drug concentrations would thus be expected if the same mg/kg dose is used. With increasing use of polymyxin B, an optimal treatment and dosing strategy is needed to provide a balance between efficacy and safety in the setting of increasing multidrug-resistant pathogens.^7,8

For optimal therapy, the pharmacokinetic data should be considered in light of pharmacodynamics. Most pharmacodynamic data available for polymyxin B apply to P. aeruginosa; the agent was shown to demonstrate concentration-dependent killing and area under the concentration-time curve/MIC appeared to be most closely linked to bactericidal activity.⁹ It is unknown whether these data can be extrapolated to Klebsiella spp. On the basis of this knowledge the dose in our patient was increased.

To the best of our knowledge, this is the first case reporting the pharmacokinetics of polymyxin B in an infant. Our limited data did not allow us to adequately explore complex pharmacokinetic profiles. Based on our simple analysis, the elimination half-life of polymyxin B1 was found to be 3.1 hours, which was shorter than those previously observed in adults.³ In addition, the pharmacokinetics of isoleucine polymyxin B1 appeared to be slightly different, but such a conclusion would be premature with only limited observations in 1 patient. More studies are warranted to define the disposition of this agent in pediatric patients.