A reply:

Professor Lipworth suggests that the tissue distribution of lipophilic corticosteroids is a more important factor than bioavailability in determining systemic effects. The steady-state volume of distribution, a measure of the partitioning of drugs between plasma and tissues, is high for all intranasal corticosteroids. The values range from 103 l [1] for triamcinolone acetonide to 424 l for beclomethasone-17-monopropionate (the active metabolite of (beclomethasone dipropionate) [2]. All these values are large compared with the volume of plasma (3 l) so the majority of the absorbed dose will be distributed to the tissue (97–99%) for all intranasal corticosteroids. For example, triamcinolone acetonide has a smaller volume of distribution than fluticasone propionate (318 l), however, the plasma and tissue exposures are estimated to be greater than for fluticasone propionate because of the higher bioavailability of triamcinolone acetonide. For triamcinolone acetonide (bioavailability 46%, dose 220 µg) 101 µg will be absorbed following each dose and 97% of this (98 µg) will distribute to the tissues. In contrast, for fluticasone propionate (bioavailability 0.5%, dose 200 µg) 1 µg will be absorbed following each dose and although 99% of this will distribute to the tissues this amounts to < 1 µg. This higher exposure to triamcinolone acetonide, in both tissues and plasma, persists for approximately 20 h during a dose interval at steady-state despite its shorter elimination half-life (Figure 1). The relationship between tissue and plasma distribution for other intranasal corticosteroids is also shown in Figure 1. The conclusions from this analysis are firstly that for all intranasal corticosteroids most of the drug is found in the tissues during multiple dosing. Secondly, the bioavailability rather than the volume of distribution is clearly the most important factor in determining tissue exposure. Finally, even when allowance is made of the lower potency of triamcinolone acetonide compared with fluticasone propionate lower systemic effects are still predicted for fluticasone propionate [3].

Figure 1.

Calculated amounts of corticosteroid in plasma (….) and tissues (—) at steady-state for once daily dosing of intranasal corticosteroid aqueous nasal spray formulations at their clinical doses (triamcinolone acetonide 220 µg ○, beclomethasone dipropionate 336 µg ▿, budesonide 128 µg □, mometasone furotae 200 µg ◊ and fluticasone propionate 200 µg ▵). Values are based on published values of volume of distribution at steady-state, systemic bioavailability and clearance. (source [3]).

The low absolute bioavailability of intranasal fluticasone propionate is due to low absorption from both the nose and the gut and its measurement is not influenced by the volume of distribution. Fluticasone propionate is administered as an aqueous suspension and its absorption directly in the nose appears to be limited by dissolution [4]. The aqueous solubility of fluticasone propionate is very low, therefore most of the drug is cleared by the nasal cilia and swallowed before dissolution occurs. The swallowed dose is subject to almost complete first pass metabolism resulting in negligible oral bioavailability [5], whereas for intranasal corticosteroids with higher systemic absorption oral bioavailablity provides a major contribution to systemic exposure and higher aqueous solubility results in greater nasal absorption [3].

Lipworth cites a study by Wilson et al. [6] as evidence that 200 µg day⁻¹ intranasal fluticasone propionate has significant systemic effects. However, the reference does not support this as overnight urinary cortisol excretion corrected for creatinine, morning serum cortisol and ACTH stimulation tests were not different from placebo. There were changes in overnight urinary cortisol excretion (9.5 h collection period) uncorrected for creatinine but there is no justification for using this parameter and ignoring the others. There are a number of reasons why the reported 43% reduction in overnight urinary cortsiol uncorrected for creatinine is not a robust result and a detailed account of the perceived flaws in that study has been published in a letter to another journal [7]. As a follow-up to this correspondence the author replicated the Wilson et al. study [6] using LC-MS/MS assay methodology and failed to show any significant change in overnight cortisol excretion following repeat daily dosing with 200 µg intranasal fluticasone propionate [8]. Furthermore, in a double blind, cross-over placebo controlled study [9], using a robust measurement of HPA axis effects (24 h serum cortisol AUC), repeat daily dosing with 200 µg intranasal fluticasone propionate produced no change relative to placebo (ratio 1.01, 90% CI 0.90, 1.14). Other studies have also failed to show significant changes in cortisol using much higher doses than those used by Wilson et al. [10, 11].

An examination of the literature in this area illustrates a lack of robustness in the methodology where urinary cortisol has been used to assess systemic exposure to corticosteroids. There are a number of possible explanations for this including: lack of assay sensitivity, cross-reactivity with corticosteroids and their metabolites, circadian influences on the choice of sampling period and nonlinear responses to exogenous corticosteroids exposure [12].

The other study Lipworth uses to support his arguments [13] was not placebo controlled so changes in cortisol excretion and the other parameters measured over time are difficult to interpret, but nevertheless the authors concluded that ‘according to serum and urinary cortisol levels the hypothalamic-pituitary-adrenal function remained intact’. The biochemical and cellular changes reported in the study cannot be interpreted as an indicative of systemic exposure to corticosteroids as they could arise following topical exposure and subsequent cell migration. In the context of long-term safety a more significant observation is that long-term exposure to intranasal corticosteroids with high systemic exposure such as BDP can have effects on growth velocity in children [14]. However, this has not been reported for corticosteroids with low systemic bioavailability such fluticasone propionate and mometasone furoate [10, 15].

Knowledge of the systemic bioavailability and potency of intranasal corticosteroids is a more logical basis for predicting relative systemic effects. Studies relying on HPA axis effects alone are not useful in this regard, not only due to concerns surrounding the methodology [16] but also because no link has been established between these measurements and the long-term safety of inhaled corticosteroids. Against this background it is difficult to see the value of such studies in guiding the prescribing physician. However, based on systemic bioavailability and potency data, relevant HPA axis effects are not likely when intranasal corticosteroids are administered at their recommended doses [3].