Absorption kinetics of oral sotalol combined with cisapride and sublingual sotalol in healthy subjects

Abstract

Aims

To study the absorption kinetics of sotalol following administration of different formulations. A formulation which results in fast absorption might be useful in the episodic treatment of paroxysmal supraventricular tachycardia (SVT), atrial fibrillation (Afib) or atrial flutter (Afl).

Methods

In an open randomized crossover study seven healthy male volunteers were given an intravenous infusion of 20 mg sotalol, for assessing the absolute bioavailability, an oral solution containing 80 mg sotalol, an oral solution containing both 80 mg sotalol and 20 mg cisapride and an 80 mg sotalol tablet, which was taken sublingually.

Results

The addition of cisapride decreased the time at which maximum serum concentrations were reached (t_max) from 2.79 (1.85–4.34) h to 1.16 (0.68–2.30) h (P=0.009) [95% CI: -2.59, −0.55] and increased the absorption rate constant (k_a) from 0.49 (0.31–0.69) h⁻¹ to 1.26 (0.52–5.61) h⁻¹ (P=0.017). The absolute bioavailability of sotalol was reduced by cisapride from 1.00±0.15 to 0.70±0.26 (P=0.006), while maximum serum concentrations of both oral solutions were not significantly different. Compared with the sublingually administered tablet with a median t_maxof 2.12 (0.89–3.28) h, the sotalol/cisapride oral solution gave a smaller t_max (p=0.009) [95% CI: −1.64, −0.36]. The k_a of the sotalol/cisapride solution was significanty (P=0.010) larger than the k_a of 0.56 (0.33–0.75) h⁻¹ found after sublingual administration of the tablet.

Conclusions

The sotalol/cisapride oral solution might be suitable for the episodic treatment of SVT, Afib or Afl.

Methods

Subjects

Seven healthy male subjects, age 19 to 27 years (median 22 years), weighing 73 to 100 kg (median 79 kg) and body height 180 to 194 cm (median 187 cm), participated in this randomized crossover study, after giving written informed consent. The protocol was approved by the Hospital's Ethics Committee. All subjects were healthy as judged by medical history, routine laboratory tests and ECG recordings. On four separate occasions separated by a wash-out period of at least 1 week, different formulations of sotalol were administered after an overnight fast. The following formulations were given to all subjects in a randomized order:

Intravenous infusion of 20 mg sotalol, a tablet containing 80 mg sotalol, which was taken sublingually, an oral solution containing 80 mg sotalol and an oral solution containing 80 mg sotalol and 20 mg cisapride.

Dose selection

As a sublingual formulation the commercially available sotalol tablets were used. More rapid absorption could result in higher serum concentrations. For safety reasons the commercially available tablet with the smallest amount of sotalol was chosen. We choose to use equal dosages for the sublingual tablet and oral solutions. An intravenous dose of 20 mg sotalol was the smallest dose we considered sufficient for assessing absolute bioavailability.

Preparation and administration of sotalol dosage forms

Intravenous infusion

(±)-sotalol (Sotacor® Bristol-Myers Squibb) 2 ml intravenous injection (10 mg ml⁻¹) was added to 50 ml isotonic saline and administered over a 10 min period with an infusion pump (Asid Bonz PP 50-300).

Sublingual tablet

An 80 mg (±)-sotalol (Sotacor®) plain tablet was taken sublingually. The subjects were instructed to keep the tablet in position. They were asked to indicate the moment at which the tablet was completely disintegrated.

Sotalol oral solution

The oral solution was prepared by mixing 20 ml distilled water and 8 ml ±-sotalol (Sotacor®) intravenous injection (10 mg ml⁻¹).

Sotalol/cisapride oral solution

The oral solution was prepared by mixing, immediately before administration, 20 ml cisapride (Prepulsid® Janssen Pharmaceutica) suspension (1 mg ml⁻¹) and 8 ml ±-sotalol (Sotacor®) intravenous injection (10 mg ml⁻¹).

The subjects were instructed to take the total volume of the sotalol and sotalol/cisapride oral solution all at once followed by drinking 100 ml tap water.

Sampling procedure

Blood samples were taken from an indwelling catheter inserted in a forearm vein. After administration of the oral solutions and the sublingual tablet the following schedule for blood sampling was used: 0, 10, 20, 40, 60 min, 1.5, 2, 3, 4, 6, 8, 12, 24, 36, 48 h. After intravenous administration the sampling schedule was 0, 1,2, 8, 10, 30 min, 1, 1.5, 2, 4, 6, 8, 12, 24, 36, 48 h. The 24, 36, 48-h blood samples were obtained by separate venepunctures. Blood samples were centrifuged and serum samples stored in glass tubes at −20° C until analysis. Sotalol was stable in frozen serum samples for at least 2 months.

Drug analysis

Sotalol serum concentrations were determined by an ionpair reversed-phase h.p.l.c. method, with ultraviolet detection at 226 nm [17]. The method was modified from the method described by Kärkkäinen [18]. The lower limit of quantification of the assay was 0.04 mg l⁻¹ with an intraday coefficient of variation (CV) of 4.4% at 0.2 mg l⁻¹ and an interday CV of 2.6% at 0.2 mg l⁻¹.

Pharmacokinetic analysis

Sotalol serum concentrations were analysed by using the MWPharm® (version 3.03, MediWare, Heerenveen, The Netherlands) pharmacokinetic package. For each individual the maximum sotalol serum concentration (C_max) and the time after sotalol administration to reach this concentration (t_max) were calculated from the serum concentration vs time profiles by means of a non-linear least-squares iterative fitting procedure using a two-compartment model, assuming first order absorption kinetics. Absolute bioavailability was calculated as the dose corrected ratio of the area under the concentration vs time curve (AUC) from zero to infinity, with the intravenous injection as reference. Numerical integration was performed using the linear and logarithmic trapezoidal rule according to the SSD-criterion to assess the AUC up to the last measurable sotalol serum concentration [19]. AUC from the last data point to infinity was estimated from the concentration at the last data point divided by the terminal elimination constant, which was obtained from the terminal slope of the semilogarithmic concentration versus time curve. The cumulative absorption was calculated by numerical deconvolution using the data points up to 8 h after administration [20]. From the cumulative absorption profile the absorption rate constant (k_a) and the lag-time (t_lag) were calculated by fitting the cumulative absorption profile to the equation:

where A(t) is the cumulative amount absorbed at time t (t>t_lag) expressed as the fraction of the dose, and A_max is the maximum absorbable fraction of the dose [21]. AUC and the cumulative absorption were calculated by using KinBes® (version 1.37; MediWare, Heerenveen, The Netherlands). Serum concentrations in the absorption phase below the quantification limit of 0.04 mg l⁻¹ were set to zero.

Statistical analysis

All statistical calculations were performed by using NCSS® (version 5.0; J. L. Hintze, East Kaysville, Utah, USA). Pharmacokinetic parameters are expressed as means ±s.d., median values (ranges) or geometric means (ranges). Differences in pharmacokinetic parameters were examined by two-way analysis of variance (ANOVA) with subject and sotalol formulation as factors. Paired Student's t-tests were used for studying differences between formulations. For k_a and t_lag^elog-transformed data were used in two-way ANOVA and paired Student's t-tests. 95% confidence intervals (CI) for differences in pharmacokinetic parameters were calculated. The relationship between absolute bioavailability and k_a was examined using Spearman's Rank test. P values less than 0.05 were judged to be significant.

Sample size

The sample size calculation was based on the detection of a specified difference in t_max between formulations. A mean within subject standard deviation was calculated from published data and appeared to be 0.7 h [6, 8]. Detection of a mean difference in t_max of 1 h with 80% certainty at the P<0.05 level (α=0.05, β=0.8) would require six subjects. We therefore planned a study in which seven subjects were included.

Results

Mean (±s.e.mean) sotalol serum concentrations are shown in Figure 1 and Figure 2. Pharmacokinetic parameters are summarized in Table 1. Table 2 shows the 95% CI for differences in pharmacokinetic parameters. The relationship between absolute bioavailability and k_a is shown in Figure 3.

Figure 1.

Mean (±s.e.mean) sotalol serum concentrations of seven healthy subjects after the administration of 80 mg sotalol as an oral solution (open circles), as an oral solution also containing 20 mg cisapride (closed circles) and as a sublingual tablet (triangles)

Figure 2.

Mean (±s.e.mean) sotalol serum concentrations of seven healthy subjects after a 10 min intravenous infusion of 20 mg sotalol

Table 1.

Pharmacokinetic parameters of sotalol in seven healthy subjects after the administration of 80 mg sotalol as an oral solution, as an oral solution also containing 20 mg cisapride and as a sublingual tablet

Table 2.

95% confidence interval for the difference in pharmacokinetic parameters between three sotalol dosage forms containing 80 mg sotalol

Figure 3.

Relationship between absolute bioavailability (F) and absorption rate constant (k_a) obtained from seven healthy subjects after the administration of an oral solution containing both 80 mg sotalol and 20 mg cisapride. Spearman's rank correlation: r_s=−0.99, P<0.02

In most subjects the last measurable sotalol concentration for the intravenous injection was reached at 8 h after administration. To avoid the problem of extrapolation the cumulative absorption was calculated by numerical deconvolution using the data points up to 8 h. This is justified by the fact that visual inspection of the individual cumulative absorption profiles shows that it is unlikely that any significant absorption occurs after 8 h.

The addition of cisapride to an oral solution of sotalol decreased t_max from 2.79 (1.85–4.34) h to 1.16 (0.68–2.30) h (P=0.009) [95% CI: −2.59, −0.55]. When sotalol was taken with cisapride, k_a increased 2.6 fold from 0.49 (0.31–0.69) h⁻¹ to 1.26 (0.52–5.61) h⁻¹ (P=0.017). Cisapride gave a decrease of approximately 30% in absolute bioavailability from 1.00±0.15 to 0.70±0.26 (P=0.006), while C_max of both oral solutions being 0.58±0.15 mg l⁻¹ and 0.61±0.15 mg l⁻¹ were similar. The average percentage of AUC from zero to infinity obtained by extrapolation was 38.7% for the intravenous injection and 15.1% for the oral and sublingual dosage forms. Although the extrapolated percentage, especially of the intravenous preparation, is quite large, the calculated absolute bioavailability for the sotalol oral solution and sotalol tablet adminstered sublingually, are in good agreement with the asymptotic value of the cumulative absorption profile obtained by numerical deconvolution being 0.96±0.15 and 0.91±0.18 respectively. For the sotalol/cisapride oral solution the absolute bioavailability agrees reasonably well with the value obtained by numerical deconvolution being 0.60±0.26.

Absolute bioavailability and k_a were found to be negatively correlated at a significant level (r_s=−0.99, P=0.02).

After sublingual administration of the tablet it took 10.6±5.2 min until the tablet was completely disintegrated. Compared with the sublingually administered tablet with a t_max of 2.12 (0.89–3.28) h, the sotalol/cisapride oral solution gave a smaller t_max (P=0.009) [95% CI: −1.64, −0.36]. The k_a of the sotalol/cisapride oral solution is significantly (P=0.010) larger than the value found after sublingual administration of the tablet. The t_lag of both oral solutions being 0.30 (0.14–0.53) h (P=0.032) and 0.36 (0.17–0.54) h (P=0.029) were significantly smaller than the t_lag of 0.59 (0.31–1.35) h found for the sublingual tablet. There was no difference in absolute bioavailability and C_max between the sotalol oral solution and the sublingually taken sotalol tablet.

Discussion

Sotalol tablets given to healthy subjects resulted in peak serum levels 2–4 h after ingestion [6–8]. This is similar to the mean t_max of 2.87 h found for the sotalol oral solution in this study. These comparable times to reach peak serum levels suggest that dissolution is not the rate-limiting factor in the absorption process of sotalol.

The addition of cisapride to an oral solution of sotalol reduced t_max by 1.6 h and is associated with a 2.6 fold increase in k_a. Most drugs are absorbed from the small intestine. The contractions of the small intestine mix the drug with the intestinal contents, facilitate the access of the drug to the absorptive surface and affect the rate of transit through the small intestine. Therefore, drugs like cisapride, that increase the rate of gastric emptying and stimulate intestinal motility [10, 11] may enhance the absorption rate of co-administered drugs. Cisapride has proven to reduce t_max of other simultaneously administered drugs e.g. flecainide [12], ranitidine [22], diazepam [23], cimetidine [24]. However, this effect is not consistent. Cisapride appeared to have no effect on the t_max of paracetamol (acetaminophen) [25] and digoxin [26]. Healthy subjects received different formulations of cisapride in the studies in which the pharmacokinetic interaction between drugs and cisapride were investigated. Cisapride was administered either intravenously [23] or orally as an oral suspension [12] or as a tablet [22, 24, 25, 26].

When sotalol was combined with cisapride, the absolute bioavailability of sotalol was reduced by approximately 30%. From Figure 3 it appears that absolute bioavailability is negatively correlated with k_a. Therefore the most likely explanation for the reduction in absolute bioavailability is that the shorter gastrointestinal transit time decreases time available for absorption from the gastrointestinal tract. Another explanation might be a cisapride-induced increase in rate of elimination of sotalol. Such an interaction is unlikely since cisapride is extensively metabolized in the liver primarily by oxidative N-dealkylation and by aromatic hydroxylation [27] whereas sotalol is renally excreted unchanged to more than 70% within 72 h and no metabolites have been identified [8, 28]. Although sotalol and cisapride were mixed immediately before administration it is possible that after ingestion both compounds interact pharmaceutically e.g. form a complex resulting in a decreased bioavailability of sotalol. Cisapride also seems to decrease the bioavailability of cimetidine [24]. For the combination with ranitidine [22] and digoxin [26] a reduction in AUC was found.

In this study, the addition of cisapride increased the interindividual variability in absolute bioavailability of sotalol. However, this does not seem to be of clinical importance after a single dose administration, since the sotalol/cisapride and sotalol oral solution give comparable peak serum levels of 0.58±0.15 mg l⁻¹ and 0.61±0.15 mg l⁻¹ respectively and resulted in similar interindividual variability in C_max. In general, an increase in absorption rate, e.g. as a result of a modified formulation of a certain drug, results in a higher C_max, if bioavailability remains unchanged. The pharmacokinetic parameters per subject in case of the sotalol/cisapride oral solution (Figure 3) show that the higher k_a, the smaller absolute bioavailability of sotalol. In subjects with a high k_a, the expected increase in C_max is compensated by a reduction in absolute bioavailability.

As compared with the sotalol oral solution a faster absorption of sotalol was not achieved by the sublingual administration of a sotalol tablet. It should be emphasized that it was not possible to distinguish between sublingual and intestinal absorption. Intestinal absorption could occur as a result of direct swallowing of the drug and back diffusion of sotalol from the sublingual mucosa to the oral cavity followed by swallowing of sotalol [29]. A conventional sotalol tablet was used as a sublingual formulation. Since the tablet disintegrates within a mean time of 10 min and sotalol hydrochloride is very soluble in water, transmembrane passage rather than dissolution rate-limits absorption. It is unlikely that the use of a conventional tablet prevents sotalol from rapid sublingual absorption.

Intravenous sotalol at a dosage of 1 mg kg⁻¹ is effective in restoring sinus rhythm in patients with an episode of SVT [3]. In the current pharmacokinetic study in healthy subjects a 10 min infusion of a mean dosage of 0.24 mg kg⁻¹ gives at 20 min after the end of infusion a mean sotalol serum concentration of 0.29 mg l⁻¹. Since sotalol shows linear pharmacokinetics a 10 min intravenous infusion at a dosage of 1 mg kg⁻¹ would give serum concentrations of 1.2 mg l⁻¹ at 20 min after the end of infusion. The combination of cisapride and 80 mg sotalol resulted in a mean C_max of 0.58 mg l⁻¹. To obtain a concentration of approximately 1.2 mg l⁻¹ as C_max in case of the sotalol/cisapride oral solution the sotalol dosage should be increased to 160 mg if sotalol shows linear pharmacokinetics in this formulation. Whether this is the case should be further investigated. Since the volume of distribution is decreased in the elderly a smaller dosage might be effective in this category of patients [7].

The extent to which sotalol in patients with an episode of Afib or Afl is effective is still unclear. Sotalol at a dosage of 1.5 mg kg⁻¹ i.v. does not terminate AFib and Afl but does slow ventricular rate [3]. As compared with the treatment of SVT, a higher dosage may be needed to restore SR in patients with Afib or Afl. A reduction in ventricular reponse in patients with AFib and Afl is achieved by a dosage as low as 20 mg i.v. [30]. To obtain this effect an oral sotalol/cisapride solution containing 80 mg sotalol is expected to be appropriate.

In conclusion, the sotalol/cisapride oral solution might be suitable for the episodic treatment of SVT, Afib or Afl. Cisapride reduces the absolute bioavailability of sotalol and increases its interindividual variability. However, this does not seem to be of clinical importance in episodic treatment since C_max and its interindividual variability remains unchanged. However, the efficacy and safety of such a dosage form should be further investigated in a study including patients with an episode of the arrhythmia.