High Levels and Safety of Oseltamivir Carboxylate Plasma Concentrations after Nasogastric Administration in Critically Ill Children in a Pediatric Intensive Care Unit

Abstract

During the 2009 H1N1 influenza pandemics, the concentrations of oseltamivir (O) and its active metabolite (oseltamivir carboxylate [OC]) were determined in 11 children (1 month to 16 years of age) admitted to intensive care units for presumed severe H1N1 infection. They received oseltamivir phosphate (OP) nasogastrically at doses between 1.5 and 6.8 mg/kg of body weight. High OC concentrations were found, with a mean level of 678 ± 535 μg/liter. The mean OP concentration was 27 ± 52 μg/liter. No marked side effect was reported.

Influenza virus is a leading cause of acute respiratory infection associated with significant annual morbidity and mortality. The recent outbreak of 2009 pandemic influenza A (H1N1) was distinguished by its severity (17). Children under the age of 2 years, in particular those less than 6 months of age, exhibited the greatest hospitalization rates and mortality (7, 10, 13, 14, 16, 17). Oseltamivir (Tamiflu) is an influenza virus neuraminidase inhibitor, which is actually a prodrug (oseltamivir phosphate [OP]) rapidly and extensively converted by hepatic esterases to a single active metabolite, oseltamivir carboxylate (OC) (8, 11, 12). Oseltamivir has been shown to be well tolerated and effective in reducing the severity and duration of the symptoms of influenza in adults and healthy children who are older than 1 year of age (5, 6, 15, 20, 21). Regarding the pharmacologic management of the severe 2009 H1N1 influenza infection, some guidelines were recommended by the World Health Organization for adults requiring oseltamivir nasogastric administration. However, no guidelines were recommended for critically ill children's care. Therefore, when the 2009 H1N1 pandemics began in France, pediatric physicians in intensive care units (ICU) feared that their patients could be underexposed to the drug when administered nasogastrically and thus be at risk of insufficient drug efficacy and worsening of their clinical status. Between November 2009 and January 2010, 11 critically ill children were admitted in the ICU of the Necker Enfants Malades Hospital (Paris, France) with suspected or confirmed H1N1 influenza infection. Peripheral blood samples were collected at different times after nasogastric administration of oseltamivir. Plasma concentrations of O and OC were monitored as part of clinical care to ensure that doses of oseltamivir were sufficient. They were quantified with a method combining high-performance liquid chromatography and tandem mass spectrometry (ADME Bioanalyses Laboratory, Vergèze, France). The children's ages were distributed as follows: 1 to 3 months (3 children), 11 to 17 months (4 children), 3 to 4 years (1 child), 10 to 16 (3 children). Oseltamivir therapy consisted of nasogastric administration of oseltamivir twice daily with doses between 1.5 and 6.8 (mean dose ± standard deviation, 3.7 ± 1.5) mg/kg of body weight for a mean duration of 3.0 ± 2.5 (range, 0.5 to 7) days. Treatment was decided because of clinical presentation, other cases diagnosed in the family and/or preexisting severe comorbidities, including lung or heart diseases, and immunosuppression. The median length of ICU stay was 6.7 ± 3.8 (range, 2 to 12) days. At admission to the ICU, fever and upper respiratory tract symptoms (cough, retraction signs, and hypoxia) were the most common clinical signs (10 and 9 children out of the 11, respectively). Six patients were mechanically ventilated. Leukopenia, anemia, and elevated C-reactive protein (CRP) levels were the most frequent biological abnormalities. Sixteen plasma concentrations of O and OC were quantified during the ICU stay of the 11 hospitalized children, with a mean delay between the last OP administration and blood sampling of 9.6 ± 11.8 (range, 2 to 51) hours. The mean concentration of O was 27 ± 52 (range, 0.2 to 215) μg/liter. Oseltamivir carboxylate concentrations were much higher, with a mean level of 678 ± 535 (range, 79 to 1,871) μg/liter. The ratio of O to OC concentrations was highly variable, with a mean of 309 ± 852, ranging from 5 to 3,464. The maximum ratio was observed in a patient who developed a multiple-organ dysfunction syndrome, including acute renal failure. Results are summarized in Table 1. Oseltamivir was well tolerated up to the highest dose (6.8 mg/kg). Three children developed mild diarrhea during oseltamivir treatment. One of them was affected by mucoviscidosis, while the two others had diarrhea after stopping morphine treatment. No other clinically relevant changes or abnormalities in vital signs or laboratory tests were observed. In the end, only three children, aged 10 to 16 years, were diagnosed to be really infected by influenza A (H1N1) by using reverse transcription (RT)-PCR. Two children fully recovered, while one child with H1N1-associated encephalitis kept aftereffects (epilepsy). The present study thus provides new pharmacological data regarding O and OC concentrations after nasogastric administration of OP in critically ill children. Indeed, we reported high effective therapeutic OC plasma concentrations after nasogastric administration of OP at doses between 1.5 and 6.8 mg/kg in this population. The average OC concentration we found dramatically exceeded, by about 3,400- to 6,800-fold, the 50% inhibitory concentration (IC₅₀) against influenza A isolates, demonstrated to be around 0.1 to 0.2 ng/ml (1, 3, 17). The 90% inhibitory concentration is not available for isolates of the 2009 pandemic (H1N1) influenza. However, if one assumes a ratio between the maximal IC₅₀ and the maximal IC₉₀ similar to those found for the other influenza A viruses (i.e., about 10), the plasma values of OC in our study were still 34- to 68-fold greater than likely values for the 90% maximal inhibitory concentration (4). In adults, Taylor et al. also reported high OC plasma concentrations after nasogastric oseltamivir administration, yet only for three subjects (19). Regarding the use of oseltamivir in children younger than 1 year, very few data exist. This may explain the use of broad dosing ranges of oseltamivir in these children and thus support the need for pharmacokinetic (PK) data to provide dosing guidelines for this very young population (13). In the present work, infants who were younger than 1 year received doses between 3.0 and 6.8 mg/kg that were associated with high effective OC plasma concentrations and a very good tolerance. Actually, all our results confirmed the good overall tolerability of oseltamivir. Moreover, the lack of oseltamivir side effects despite the high variability of OC plasma concentrations, probably partly due to the variability of oseltamivir doses and times of blood sample collection, also confirmed its wide clinical safety margin previously reported in numerous studies (1, 8, 9, 18). Two limitations of our study are the low number of patients and the fact that the results of this paper cannot be used directly for dose selection in children because of the lack of PK study design. Indeed, we determined the concentrations of oseltamivir and its metabolite as part of clinical care for H1N1 infection in children rushed to the ICU in a critical situation, but with concern about protecting children from worsening their clinical status because of an inappropriate dose of oseltamivir. In this context, a prospective PK study would have been completely impractical. Conversely, Acosta et al. recently conducted a well-designed PK study leading to data regarding drug selection for oseltamivir in neonates (2). However, the context was totally different. Indeed, 32 neonates were exposed to the virus at the same time, in the same place, which enabled the setting up of a prophylactic treatment that could be completely organized and controlled (2). As a conclusion, the present study showed that nasogastric administration of OP at doses between 1.5 and 6.8 mg/kg in 11 children who were 1 month to 16 years old enabled us to reach sustained effective therapeutic plasma concentrations of OC, the active metabolite, in this pediatric population. Our work also confirmed the high clinical safety margin of oseltamivir, for which drug therapeutic monitoring appears to be unnecessary.

TABLE 1.

Average plasma concentrations of oseltamivir (O) and oseltamivir carboxylate (OC) measured in children after nasogastric administration of oseltamivir

Concn or ratio	Mean ± SD (range)	Median (interquartile range)
Plasma concn (μg/liter) (n = 16) of:
O	27.2 ± 52.2 (0.2-215)	11.5 (3.4-26.1)
OC	678 ± 535 (79-1,871)	591 (315-919)
O/OC ratio	309 ± 852 (5-3,464)	43 (20-134)