Measurement of tobramycin and gentamicin in saliva is not suitable for therapeutic drug monitoring of patients with cystic fibrosis

Abstract

We read with interest the paper by Madsen et al. [[1]Madsen V. Lind A. Rasmussen M. Coulthard K. Determination of tobramycin in saliva is not suitable for therapeutic drug monitoring of patients with cystic fibrosis.J Cyst Fibros. 2004; 3: 249-251Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar] who reported that tobramycin could not be detected in saliva within the first 6 h after once daily dosing, concluding that plasma cannot be substituted with saliva for the therapeutic drug monitoring(TDM) of patients with cystic fibrosis(CF). Madsen et al. mention in their paper two short reports describing the use of saliva as an alternative to serum for monitoring once daily gentamicin in children [2Berkovitch M. Bistritzer T. Aladjem M. Burtin P. Dagan T. Chen-Levi Z. et al.Clinical relevance of therapeutic drug monitoring of digoxin and gentamicin in the saliva of children.Ther Drug Monit. 1998; 20: 253-256Crossref PubMed Scopus (12) Google Scholar, 3Berkovitch M. Goldman M. Silverman R. Chen-Levi Z. Greenberg R. Marcus O. et al.Therapeutic drug monitoring of once daily gentamicin in serum and saliva of children.Eur J Pediatr. 2000; 159: 697-698Crossref PubMed Scopus (12) Google Scholar]. These studies confirmed a correlation between salivary and serum levels; however, none of these subjects had toxic levels, and therefore, this system has not been tested for its ability to detect such levels. Berkovitch et al. proposed that it takes longer for gentamicin to enter and equilibrate with the saliva, hence the correlation between serum and saliva levels only when gentamicin was given once daily as opposed to a twice daily or thrice daily regime in their study [2Berkovitch M. Bistritzer T. Aladjem M. Burtin P. Dagan T. Chen-Levi Z. et al.Clinical relevance of therapeutic drug monitoring of digoxin and gentamicin in the saliva of children.Ther Drug Monit. 1998; 20: 253-256Crossref PubMed Scopus (12) Google Scholar, 3Berkovitch M. Goldman M. Silverman R. Chen-Levi Z. Greenberg R. Marcus O. et al.Therapeutic drug monitoring of once daily gentamicin in serum and saliva of children.Eur J Pediatr. 2000; 159: 697-698Crossref PubMed Scopus (12) Google Scholar]. We wish to describe our own experience of using saliva as an alternative to plasma for TDM of children with CF. We studied 24 children who were prescribed once daily gentamicin (n=19) or tobramycin (n=5) at a dosage of 10–12 mg/kg/day for a chest exacerbation. Median (range) age was 10 years (6.9 to 16.75). Trough serum and saliva samples were obtained 48 h after drug initiation, prior to administration of the 3rd dose. Citric acid crystals were used to stimulate saliva flow if necessary. In addition, in an attempt to assess the ability of the test to detect ‘toxic’ levels, 4 patients (tobramycin (n=3) and gentamicin (n=1)) also had paired plasma and saliva samples taken at various post-dose time points (1, 3, 5 and 7 h) when serum levels were predicted to be high. Levels were estimated using the Abbott TDx system (Abbott laboratories Ltd., Maidenhead, UK), which uses fluorescence polarization immunoassay (FPIA) technology. The lower limits of detection (distinguished from zero with 95% confidence) in our laboratory were 0.27 mg/L for gentamicin and 0.18 mg/L for tobramycin. Our results showed that pre-dose correlation was poor (R=0.213 and p=0.213). Post-dose (up to 7 h) all serum levels were high, range (2.1–22.0); despite this, none of the saliva samples were >1 (0.07–0.94), i.e., toxicity was not detected in a single case. Although aminoglycosides could be detected in saliva, there was no correlation between absolute levels of saliva and serum of either tobramycin or gentamicin levels in patients with CF at any time point measured. Most importantly, in every case, a high serum level would have been missed on salivary testing. Plasma cannot be substituted by saliva for TDM of tobramycin or gentamicin in patients with CF.