Sublingual administration of furosemide: new application of an old drug

Abstract

What is already known about this subject

• Furosemide is an effective diuretic, but its absorption may be too slow to allow oral treatment in certain patients.

What this study adds

• In healthy volunteers, sublingual administration is associated with a higher C_max, a higher bioavailability and a more accentuated initial natriuretic response than oral furosemide. Sublingual administration may offer advantages over oral administration of furosemide in certain clinical situations.

Background

In patients with decompensated heart failure, absorption of orally administered furosemide may be delayed, possibly leading to impaired pharmacodynamic effects. Sublingual administration may represent an alternative in such situations.

Methods

In a crossover study including 11 healthy men, 20 mg furosemide was administered intravenously, orally and sublingually on three different days. Pharmacokinetics and pharmacodynamics were assessed from repeated blood and urine samples.

Results

Compared with oral administration, sublingual administration was associated with 43% higher C_max[difference 215 ng ml⁻¹, 95% confidence interval (CI) 37, 392], a higher urinary recovery (8.9 vs. 7.3 mg, difference 1.6 mg, 95% CI 0.3, 2.9), an 28% higher AUC (difference 328 ng h⁻¹ ml⁻¹, 95% CI 24, 632) and a higher bioavailability of furosemide (59 vs. 47%, difference 12.0%, 95% CI −1.2, 25.2). Sodium excretion was higher after sublingual compared with oral administration (peak excretion rate 1.8 vs. 1.4 mmol min⁻¹, P < 0.05), whereas urine volume did not differ significantly between the two application modes. In comparison, intravenous administration showed the expected more rapid and intense response.

Conclusion

Sublingually administered furosemide tablets differ in certain kinetic and dynamic properties from identical tablets given orally. Sublingual administration of furosemide may offer therapeutic advantages in certain groups of patients.

Introduction

Furosemide is a rapidly acting loop diuretic used in oedematous states associated with renal, hepatic and particularly cardiac failure [1]. So far, the routes of administration are oral and intravenous (i.v.). Intravenously, furosemide is used primarily in decompensated heart failure, when rapid diuretic action is required and intestinal absorption of furosemide may be delayed due to gastrointestinal oedema [2–5]. Thus, lag time, time to peak, and peak of serum concentration may differ in compensated compared with decompensated patients after oral furosemide intake, whereas elimination half-life and area under the serum concentration curve are similar [2, 6].

Additionally, bioavailability of furosemide exhibits a large interindividual variability, primarily due to limited absorption [7]. In healthy subjects, the median bioavailability is approximately 50% with a wide range of 20–80% [8]. Most of this variability derives from the absorption process, which is slow and influenced by gastric emptying and ingestion of food [9–12], contributing to the well known intra- and interindividual differences of the diuretic effect of oral furosemide [9]. In clinical practice, it would be very useful to possess a safe and predictable route of administration not requiring i.v. injection. Sublingual application may fulfil this requirement, as intestinal absorption is circumvented. We therefore compared the pharmacokinetics and pharmacodynamics of sublingual furosemide with its oral administration, using the i.v. route as control.

Methods

Eleven healthy male volunteers without alcohol, tobacco or drug abuse (age 26 ± 4 years) were recruited after clinical and biochemical exclusion of any disorder, and written informed consent was obtained. All subjects abstained from medication 1 week prior to and throughout the study. As an analgesic, only paracetamol up to 4 g daily was allowed. They also abstained from alcohol 48 h prior to and throughout the study. The study protocol had been approved by the Ethics Committee of the State of Basel.

The study had an open, crossover design. All laboratory tests were performed by investigators blinded to the group allocation. Subjects were treated with a single dose of 20 mg furosemide on three different days either sublingually, orally or intravenously in random order and 7 days apart. They had to restrain from any intake of food 8 h prior to study drug intake to optimize intestinal absorption of furosemide [11].

A venous catheter was inserted into a cubital vein of the nondominant arm. The study drug was administered at 08.00 h. Orally, a 20-mg tablet furosemide (Lasix®, Sanofi-Aventis, Geneva, Switzerland) was swallowed immediately with 180 ml water. Sublingually, the 20-mg furosemide tablets were placed under the tongue and had to be kept there for 5 min. Before sublingual administration of the tablet, subjects had to drink 180 ml of water to ensure conditions similar to oral administration. During the disintegration time of 5 min (the period of sublingual administration), subjects were advised not to swallow. The exact procedure had been practised with every subject prior to tablet administration. Pilot studies had shown that the furosemide tablets used were completely disintegrating within 5 min under the tongue. Intravenously, furosemide was injected over 1–2 min directly into a cubital vein of the dominant arm. Before injection, subjects drank 180 ml water to ensure conditions similar to oral administration. In line with previous studies [5, 10], blood samples were collected from the venous catheter immediately before drug administration, followed by every 15 min for first 2 h, and then at 2.5, 3, 3.5, 4, 6 and 8 h after drug administration. The bladder had to be voided and urine was collected quantitatively immediately before and 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, 12 and 24 h after drug administration. Urine volumes were recorded and aliquots frozen until analysis. During the first 8 h after furosemide administration, urine output was isovolumetrically replaced by i.v. 0.45% NaCl solution. During the first 6 h after drug administration, the subjects were recumbent. Drinking or eating was not allowed for the first 6 h. Thereafter, a standardized meal and water were provided.

Furosemide concentrations in plasma and urine were measured as previously described using a high-performance liquid chromatography method [13]. Within- and between-assay coefficients of variation were <6.5% and <10%, respectively, and the lower limit of quantification was 12 ng ml⁻¹. Serum concentrations of sodium, potassium, chloride, phosphate, creatinine and urea, as well as the respective urine concentrations (except urea), were determined by routine methods.

Pharmacokinetic analysis

Maximal plasma concentration (C_max) and time to C_max (T_max) were determined by visual inspection of raw data by an investigator blinded to other results. Terminal elimination rate constant (λ_z) and half-life of furosemide (t_1/2), area under plasma concentration–time curve extrapolated to infinity (AUC_0–∞), systemic clearance over the bioavailable fraction (CL/F) and volume of distribution of the central compartment (V/F) were calculated by noncompartmental analysis using WinNonlin software (version 4.01; Pharsight Corp., Cary, NC, USA). The bioavailability (F) of furosemide was calculated as:

Pharmacodynamic analysis

Drug effects were assessed as the fractional and cumulative renal excretion of sodium and water.

Statistics

Results are presented as percentage or as mean (SD) unless indicated otherwise. Bioequivalence was assessed by calculating the 90% confidence interval (CI) of the log-transformed values for AUC and C_max. As the study aimed to compare oral with sublingual administration, only the differences between these two modes of administration were statistically analysed. Assuming a variability of 25%, a difference of approximately 30% could be detected with a statistical power of 80% at an α-level of 5%, when including 11 subjects. Pharmacodynamics were analysed by a general linear model for repeated measures over the first 4 h to test if the response to furosemide differed between the routes of administration. Other comparisons were done by two-tailed paired t-test or Wilcoxon test, as appropriate. A P-value of <0.05 was considered to be statistically significant. Statistical analyses were performed using SPSS 14.0 (SPSS Inc., Chicago, IL, USA).

Results

Pharmacokinetic data of furosemide are summarized in Table 1 and displayed in Figure 1. After i.v. administration, furosemide pharmacokinetics are characterized by the expected low volume of distribution, rapid elimination, and clearance in the range of the glomerular filtration rate. Compared with oral administration, sublingual administration resulted in significant increases in maximal plasma concentration of furosemide of 43%, AUC of 29% and bioavailability of 26%. The corresponding ratios of the log-transformed values (90% CI) were 1.06 (1.01, 1.11) for C_max, 1.03 (1.00, 1.06) for AUC and 1.06 (1.01, 1.11) for bioavailability. Half-life and time to peak did not differ between the two. Urinary excretion after 24 h was 67% of the dose administered intravenously, 37% orally and 45% sublingually (Figure 1b).

Table 1.

Pharmacokinetic parameters of 20 mg furosemide by route of administration

	i.v.	po	SL	P SL vs. PO
AUC(0-∞) (ng h−1 ml−1)	3038 (1505)	1203 (306)	1542 (505)	0.02
F (%)	100	47 (6)	59 (9)	0.001
λz (l h−1)	0.52 (0.07)	0.39 (0.09)	0.45 (0.07)	0.06
T1/2 (h)	1.3 (0.2)	1.9 (0.5)	1.6 (0.3)	0.11
Cmax (ng ml−1)	2861 (1372)	552 (221)	789 (242)	0.01
Tmax (h)	0.3 (0.0)	1.5 (0.8)	1.4 (0.6)	0.79
CL (ml min−1)	121 (61)	NA	NA
Vc (l)	14 (7)	NA	NA
Total urinary recovery (mg)	13.4 (1.1)	7.3 (1.8)	8.9 (2.2)	0.009
Fu (%)	100	54.3 ± 10.5	66.4 ± 14.2	0.01

Administrations were intravenous (i.v.), sublingual (SL) or oral (po). AUC, Area under the plasma concentration–time curve; F, absolute bioavailability; λ_z, elimination rate constant; T_1/2, terminal half-life; C_max, maximum plasma concentration; Tmax, time to reach C_max; CL, clearance; V_c, volume of distribution; F_u, relative urinary excretion. Results are presented as mean (SD). NA, Not available.

Figure 1.

Plasma concentration time-curve (A) and cumulative urinary excretion (B) of 20 mg furosemide, administered intravenously (IV), sublingually (SL), or orally (PO) (mean ± SEM). See Table 1 for pharmacokinetic constants. Intravenous, (○); oral, (); sublingual, ()

The excretion rate of sodium after sublingual administration differed from oral over the first 4 h (Figure 2, P < 0.05). The peak sodium excretion rate was higher [1.8 (0.6) mmol min⁻¹, P < 0.02] and slightly earlier [median 1 h, interquartile range (IQR) 0.5 h, P = 0.25] after sublingual than after oral administration [1.4 (0.6) mmol min⁻¹; 1.5 h, IQR 0.5 h]. The cumulative sodium excretion over the first 4 h was 167 (57) mmol and 140 (48), respectively (P = 0.1). Expectedly, the peak sodium excretion rate was much earlier and higher after i.v. administration, paralleling the urinary excretion of furosemide. The relation between the urinary excretion rates of furosemide and sodium is given in Figure 3. This relation was not different between the modes of application of furosemide. The excretion rate and the cumulative excretion of urine are shown in Figure 4. There was an early and high peak of the excretion rate after i.v. administration. After oral and sublingual administration, this peak appeared later and was broader. Neither the excretion rate (P = 0.1) nor the cumulative (P = 0.3) urine excretion after sublingual furosemide administration was significantly different from those after oral administration.

Figure 2.

Renal excretion rate (A) and cumulative excretion (B) of sodium after 20 mg of furosemide intravenously, sublingually or orally (mean ± SEM). Oral, (); sublingual, (); intravenous, (○)

Figure 3.

Relation between urinary excretion rates of furosemide and sodium after treatment with 20 mg furosemide intravenously, sublingually or orally. Intravenous, (○); oral, (); sublingual, ()

Figure 4.

Urine excretion rate (A) and cumulative urine excretion (B) after 20 mg of furosemide intravenously, sublingually or orally (mean ± SEM). Oral, (); sublingual, (); intravenous, (○)

The urinary excretion of potassium, chloride, phosphate and creatinine as well as the serum concentrations of sodium, potassium, chloride, phosphate, creatinine and urea were not different between the modes of furosemide administration (data not shown).

Discussion

Diuretics are a mainstay for treatment of volume overload in acute and chronic heart failure. However, efficacy of oral administration may be limited by delayed intestinal absorption, particularly in decompensated heart failure [3, 5, 6], often requiring i.v. application to achieve a sufficient diuretic effect. This study suggests that the sublingual administration of furosemide may overcome some of the shortcomings observed with oral administration.

Compared with i.v. administration, both oral and sublingual administration had a slower onset of action and a less accentuated early (first hour) effect, whereas the effect was more accentuated later on (1.5–4 h). The elimination half-life was longer than after i.v. administration, illustrating the slow absorption process of oral furosemide, leading to an apparent delay in the elimination [8, 10]. Nevertheless, our data in healthy subjects show that there are indeed differences in the pharmacokinetics and pharmacodynamics of oral and sublingual furosemide administration. After sublingual administration, the peak concentration was higher, as were the urinary excretion rates of furosemide and sodium between 1 and 3 h, and the total cumulative urinary excretion of furosemide. It may be speculated that these differences would be even larger in patients with decompensated heart failure, where intestinal furosemide absorption is further delayed [3, 5, 6].

As shown in Figure 3, there was no difference in the relation between the urinary excretion rates of furosemide and sodium. This finding indicates that the differences observed in renal sodium excretion between sublingual and oral application cannot be explained by the renal action of furosemide, but rather by the observed differences in its pharmacokinetics.

Considering the physicochemical properties of furosemide (XlogP 1.4, molecular weight 331), furosemide may be judged to be too hydrophilic and therefore unsuitable for sublingual absorption. However, not only lipophilicity matters for sublingual absorption, but also water solubility, necessitating a good balance between lipophilicity and hydrophilicity [14]. In this context, it is noteworthy that nitroglycerine, a substance with good buccal absorption [14], is more hydrophilic than furosemide (XlogP 1.0) and has a molecular weight in a similar range (227).

The same tablets were used for both sublingual and oral administration. Although pilot experiments have shown that the tablet disintegrates after 5 min, it is likely, suggested by the limited bioavailability, that buccal absorption of furosemide was not complete after sublingual administration and that a variable dose fraction was swallowed. Modifications of the galenical formulation (e.g. more rapid disintegration) could probably improve the pharmacokinetics and pharmacodynamics of furosemide administered sublingually. An additional possibility to optimize the diuretic effect of sublingual administration may be the choice of a more potent agent (e.g. torasemide, piretanide or bumetanide). As less active substance is needed to achieve pharmacodynamic effects for these agents, the fraction absorbed in the buccal cavity could be increased compared with furosemide.

In conclusion, sublingual and oral furosemide administered as the same galenical formulation differ in certain kinetic and dynamic properties. Sublingual administration of furosemide may offer therapeutic advantages over the oral route, particularly in patients with decompensated heart failure, but further studies are needed to define its clinical value.