Exploring the multiple effects of nifedipine and captopril administration in spontaneously hypertensive rats through pharmacokinetic‐pharmacodynamic analyses

Abstract

This study assessed the pharmacokinetics (PKs) and pharmacodynamics (PDs) of two antihypertensive drugs, nifedipine and captopril, by exploring their main (blood pressure [BP]) and secondary effects (heart rate [HR] and QT interval [QT]) in spontaneously hypertensive rats. This study aimed to assess the relationship between PKs and PDs. Using these PD parameters, BP, HR, and QT during coadministration were estimated. The coadministration of nifedipine and captopril resulted in an increase in nifedipine's total body clearance (CL_tot) and a reduction in its mean residence time (MRT) with an increase in the terminal elimination half‐life (t_1/2) and volume of distribution at steady state (V_dss) of captopril. However, no significant PK interactions were observed. During monotherapy, BP reduced rapidly following nifedipine infusion. Subsequently, despite the increase in nifedipine plasma concentration, BP recovered, likely because of homeostasis. Similar results were observed with coadministration. Subsequently, BP demonstrated a sustained reduction that was greater than or equal to the additive effect estimated from each PK. Captopril exhibited a minimal effect on HR, except for a transient increase observed immediately after starting infusion, consistent with observations during coadministration. Subsequently, the HR reduction was nearly equal to that calculated from the nifedipine PK. QT prolongation was more rapid with captopril than with nifedipine. Although QT prolongation during the initial 60 min of coadministration was approximately the sum of both effects, the recovery period to baseline QT was faster than that in the simulation.

Abbreviations

ACE

angiotensin‐converting enzyme

AIC

Akaike's Information Criterion

ARB

angiotensin II receptor blockers

AUC

area under the plasma concentration‐time curve

AUMC

area under the first moment curve

BP

blood pressure

BPB

2, 4′‐dibromoacetophenone

C_last

the last measured plasma concentration at time t

CL_tot

total body clearance

C_max

maximum plasma drug concentration

CYP

cytochrome P450

D

administered dose

ECG

electrocardiography

EC₅₀

drug concentration at half‐maximum effect

EHS

endogenous hypertensive substances

E_max

maximum drug effect

ESI

electrospray ionization

E₀

basic effect before drug administration

HPLC

high‐performance liquid chromatography

HR

heart rate

iv

intravenous

K_EHSin

k_EHSout

zero‐ and first‐order rate constants related to the formation and degradation of EHS

k_e

terminal elimination rate constant

k_e0

first‐order rate constant of the effect compartment

k₀

infusion rate

k₁₂

k₂₁, and k₁₀

first‐order rate constants for the pharmacokinetic process

LC–MS/MS

liquid chromatography–tandem mass spectrometry

MRM

multiple reaction monitoring

MRT

mean residence time

PD

pharmacodynamics

PEG

Polyethylene glycol 400

PK

pharmacokinetics

QT

QT interval

QTc

corrected QT by the RR duration

SHR

spontaneously hypertensive rat

TDM

therapeutic drug monitoring

T_inf

infusion period

t_1/2

terminal elimination half‐life

V_dss

volume of distribution at steady state

V₁

distribution volume of the central compartment

α

slope of a linear equation of a feedback system

γ

Hill constant

1. INTRODUCTION

Generally, the plasma concentration of a drug correlates with its effects and adverse effects. Therefore, therapeutic drug monitoring (TDM) aims to achieve effective drug concentrations in the plasma and design a treatment plan for clinical practice. Consequently, scientific descriptions of drug dosage, pharmacokinetics (PKs), and effects and/or adverse effects can be quantitatively assessed using a PK‐pharmacodynamic (PD) (PK‐PD) modeling techniques. ¹ , ² , ³ , ⁴ Previously, we assessed the PK and PD of nifedipine and propranolol using rats as experimental animals. ⁵ Blood pressure (BP) reduction was assessed as a main effect, whereas heart rate (HR) and QT interval (QT) were assessed as adverse effects. The differences in these characteristics between the two drugs were explained through PK‐PD analysis.

Hypertension is becoming increasingly common in Japanese adults, with cardiovascular diseases causing significant clinical, social, and economic challenges. The 2019 hypertension treatment guidelines ⁶ indicated Ca²⁺ channel blockers, Angiotensin II receptor blockers (ARBs), angiotensin‐converting enzyme (ACE) inhibitors, and diuretics as first‐line drugs for hypertension treatment, depending on the symptoms. However, <40% of patients with hypertension achieved antihypertensive therapy with monotherapy, with a common combined therapy. ⁷ , ⁸ , ⁹ When antihypertensive goals fail, two or three drugs from various pharmacological classes are administered in combination. ACE inhibitors with Ca²⁺ channel blockers are recommended. A more effective therapy can be obtained by the coadministration of antihypertensive drugs with distinct pharmacological mechanisms. ¹⁰ , ¹¹ In contrast, the adverse effects of these drugs may be enhanced by their coadministration.

This study simultaneously assessed the PK and main/secondary effects of two antihypertensive drugs with distinct pharmacological mechanisms—nifedipine, a potent Ca²⁺ channel blocker, and captopril, an orally active ACE inhibitor. Spontaneously hypertensive rats (SHRs) were used as a disease model to assess the relationship between the PK and PD of these drugs. BP as a hemodynamic effect and HR and QT as secondary effects were assessed alongside PK analysis. These drugs are often coadministered in clinical practice. Nifedipine is widely used in the treatment of cardiovascular disorders and metabolized by the hepatic metabolizing enzyme cytochrome P450 (CYP) 3A4, which has various interactions. ¹² Its PK interacts with grapefruit juice and anti‐fungal drugs. ¹³ Therefore, its clinical hypotensive effects exhibit inter‐ or intra‐individual variations. Captopril exhibits favorable tolerability and long‐term efficacy as a first‐line therapy in patients with mild or moderate essential hypertension. ¹⁴ The values of CL_tot and t_1/2 of captopril and nifedipine are approximately 1–2 h and 0.5–0.8 L/h/kg, respectively. ¹⁵ , ¹⁶ Adding to our previous investigation, ⁵ both drugs have been used clinically for a long time, and it is thought that their effects appear at the same time after administration. The PK‐PD models were constructed to elucidate the time‐dependent hemodynamic alterations associated with plasma drug concentrations. Based on these results during monotherapy, simulations were performed for the PK‐PD in combination. The results were compared with the actual measurements to assess the PK‐PD of coadministered drugs.

2. MATERIALS AND METHODS

2.1. Materials

Nifedipine and captopril were purchased from Nacalai Tesque, Inc. (Kyoto, Japan) and FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan), respectively. Acetonitrile (LC–MS grade) and polyethylene glycol 400 (PEG) were obtained from FUJIFILM Wako Pure Chemical Corporation and Nacalai Tesque Inc., respectively. All other reagents were of analytical grade and obtained commercially.

Nifedipine and captopril standard stock solutions (1.0 mg/mL) were prepared by dissolving in methanol. These solutions were used to prepare the respective calibration curve standards.

2.2. Animal experiments

Male SHRs (SHR/Izm; Shimizu Laboratory Co., Ltd., Kyoto, Japan), weighing 215–315 g, were used. The rats were housed under controlled environmental conditions and fed commercial feed pellets with free access to water. Each experiment involved 3–10 rats. Under anesthesia through intraperitoneal injection of urethane (1.0 g/kg), a polyethylene catheter (inner diameter [ID] 0.5 mm, outer diameter [OD] 0.8 mm; Natsume, Tokyo, Japan) was surgically implanted in the left carotid artery to monitor BP using BP Amps Model ML117 (AD Instruments Japan Inc., Aichi, Japan), and HR and QT were monitored through electrocardiography (ECG; Bio Amp Model ML132, both AD Instruments Japan Inc.) continuously from −15 to 180 min following the initiation of drug infusion. These data were loaded onto LabChart Pro v8.1.16 using PowerLab 4/26 Model ML846 (both AD Instruments Japan Inc.).

Intravenous (IV) infusion test solutions of nifedipine and captopril were prepared using 0.5 mL PEG per rat. The monotherapy doses of nifedipine and captopril were 1.0 and 15.0 mg/kg body weight, respectively. In contrast, during coadministration, the doses of nifedipine and captopril were reduced to 0.5 and 5.0 mg/kg. These drugs exhibit similar pharmacological effects, and the combined doses were determined to achieve similar effects when administered individually. The test solution was infused for 30 min at a constant rate of 0.5 mL/30 min into the right femoral vein using an infusion pump (KDS100; LMS Co., Ltd. Tokyo, Japan). In parallel with the PD measurements, blood samples (120 μL) were collected from the right jugular vein using a heparinized syringe (1.0 mL) from 5 to 300 min following infusion initiation. All blood samples were centrifuged to obtain plasma fractions, which were frozen immediately and stored in a freezer at −30°C until further analysis. Drug concentrations in rat plasma were measured using liquid chromatography–tandem mass spectrometry (LC–MS/MS) assay method, as described below.

2.3. Quantitative analysis

The extraction procedures for nifedipine and captopril from rat plasma were as follows.

In an ice‐cold glass extraction tube (15 mL), 50 μL of 1.0 mg/mL 2,4′‐dibromoacetophenone (BPB) was added, followed by plasma sample (50 μL), 1.0 N HCl (1.0 mL), and CH₂Cl₂:Et₂O (3:7) mixture (2 mL). BPB was added to stabilize captopril in the plasma and obtain captopril‐BPB for quantification. ¹⁷ , ¹⁸ The tubes were shaken for 20 min and subsequently centrifuged (3000 rpm, 15 min). The organic extracts were separated by freezing the aqueous layer and decanting the organic liquid into a clean glass tube. The organic liquid was evaporated using a SPD1010 SpeedVac concentrator system (Thermo Electron Corporation, Yokohama, Japan). The resulting residue was reconstituted by adding 50 μL of high‐performance liquid chromatography (HPLC) mobile phase (acetonitrile: 0.1% formic acid solution = 90:10), of which a 30 μL aliquot was injected into the LC–MS/MS system described below. Ten calibration curve samples were prepared per assay by adding known amounts of nifedipine or captopril (0.25–2000 ng) to rat plasma. Subsequently, the samples were extracted and analyzed simultaneously with the test samples using the same method. The correlation coefficient of the calibration curve is not <0.995.

Nifedipine and captopril plasma concentrations were determined using an LCMS 8050 LC–MS/MS system (Shimadzu Co., Kyoto, Japan) equipped with a Prominence HPLC system (Shimadzu). The HPLC system consisted of two LC‐20 AD pumps and an SIL‐20 AC automatic sample injector, both operating at a flow rate of 0.2 mL/min. The analytical column (Cosmosil 5C₁₈‐MS‐II, 50 × 2.0 mm ID; Nacalai Tesque, Inc.) was maintained at 40°C using a CTO‐20A column oven (Shimadzu). The data were loaded onto the Lab Solutions ver. 5.91 analytical software (Shimadzu) by connecting to a Cubic Meter (CBM)‐20A (Shimadzu) communication bus module. Detections were performed in the multiple reaction monitoring (MRM) mode of the parent and selected using product ions acting in positive mode. Nifedipine and captopril were monitored using the following mass transitions with the electrospray ionization (ESI) method: nifedipine, m/z 348 → 316; captopril‐BPB, m/z; 416 → 255. The calibration curves were linear within the range of 5.0–40 000 ng/mL.

2.4. PK analysis

PK parameters were derived through a moment analysis of individual data using WinNonlin professional version 8.3.3.33 (Pharsight Corporation, Mountain View, CA, USA). The terminal elimination rate constant (k_e) of the plasma concentration‐time curve was determined using the linear regression of three minimal data points from the terminal portion of the plot. The area under the plasma concentration‐time curve (AUC) following IV infusion was calculated using the linear trapezoidal rule up to the last measured plasma concentration (C_last) and extrapolated to infinity using a correction term. Terminal elimination half‐life (t_1/2) was determined by dividing ln2 by k_e. The area under the first moment curve (AUMC) following drug administration was calculated using the linear trapezoidal rule up to C_last, with the addition of a correction term following the last measured point to infinity. The total plasma clearance (CL_tot) and mean residence time (MRT) were determined by dividing the administered dose (D) by AUC and AUMC/AUC‐T_inf/2, respectively, where T_inf represents the infusion period (30 min). The volume of distribution at steady state (V_dss) was calculated using CL_tot/k_e. The maximum plasma concentration (C_max) was obtained from measured values.

General 2‐compartmental PK analysis was applied to the data using WinNonlin software. Figure 1 illustrates this model, which is based on the following assumptions: first, the rate constants (k₁₂, k₂₁, and k₁₀) for the PK process, followed first‐order kinetics, and the infusion rate (k₀) of nifedipine or captopril followed zero‐order kinetics. Indices 1 and 2 represent the central and peripheral PK compartments, respectively. V₁ represents the distribution volume of the central compartment.

FIGURE 1.

Schematic representation of the pharmacokinetics‐pharmacodynamics (PK‐PD) models for blood pressure (BP), heart rate (HR), and QT interval (QT) alterations induced by nifedipine and captopril. (A) The model for BP alteration induced by nifedipine. (B) The model for other alterations induced by nifedipine and captopril.

PK model equations,

$$\frac{{\mathit{dC}}_1}{\mathit{dt}}=\frac{k_0}{V_1}-\left(k_{12}+k_{10}\right)\times C_1+k_{21}\times C_2$$

$$\begin{array}{l}{k}_0=\frac{D}{T_{\inf }}\left(t\le 30\min \right)\\ k_0=0\begin{array}{cc}& \end{array}\left(t>30\min \right)\end{array}$$

$$\frac{{\mathit{dC}}_2}{\mathit{dt}}=k_{12}\times C_1-k_{21}\times C_2$$

at t = 0, C₁ = 0, and C₂ = 0.

2.5. PK‐PD analysis

The PK‐PD models illustrated in Figure 1 include k_e0 as the first‐order rate constant of the effect compartment (Index 3). E_max and EC₅₀ represent the maximum drug effect and the drug concentration at half‐maximum effect, respectively. Gamma (γ) and E₀ represent the Hill constant and basic effect before drug administration, respectively.

The effect of nifedipine on BP was analyzed using model (A) illustrated in Figure 1, which incorporates an effect compartment with the BP homeostasis mechanism, as previously reported. ⁵ In Figure 1A, Index 4 represents the compartment corresponding to the endogenous hypertensive substance (EHS). According to the E_max model, nifedipine induces a rapid reduction in BP. However, despite sustained high plasma nifedipine levels, BP increases through a homeostatic feedback mechanism. Therefore, the alteration in the BP effect is defined as follows:

$$\text{Effect}\mathrm{on}\mathrm{BP}=\text{hypotensive effect}+\text{hypertensive effect}$$

The hypertensive effect is represented by a precursor‐dependent indirect PD response model. ¹⁹ K_EHSin and k_EHSout represent the zero‐ and first‐order rate constants associated with the formation and degradation of the EHS, respectively. The feedback system was assumed to have a linear relationship (slope = α).

$$\frac{d{C}_3}{\mathit{dt}}=k_{e0}\times C_1-k_{e0}\times C_3$$

$$\frac{d{C}_4}{\mathit{dt}}=K_{\mathit{EHS},\mathit{\text{in}}}\times \left(1+\frac{E_0-E}{E_0}\times \alpha \right)-k_{\mathit{EHS},\mathit{\text{out}}}\times C_4$$

$$\frac{\mathit{dE}}{\mathit{dt}}=-\frac{\left(E_0-E_{\mathit{max},1}\right)\times E{C}_{50,1}}{{\left(E{C}_{50,1}+C_3\right)}^2}\times \frac{d{C}_3}{\mathit{dt}}+\frac{\left(E_{\mathit{max},2}-E_0\right)\times E{C}_{50,2}}{{\left(E{C}_{50,2}+C_4\right)}^2}\times \frac{d{C}_4}{\mathit{dt}}$$

at t = 0, C₃ = 0, C₄ = 0, and E = E₀.

Indices 1 and 2 in E_max and EC₅₀ express the PD parameters of the hypotensive and hypertensive effects, respectively.

The PK‐PD analyses for the HR and QT of nifedipine and the effects of captopril were performed using an ordinary sigmoid E_max model with an effect compartment, as illustrated in Figure 1B.

For HR,

$$\frac{d{C}_3}{\mathit{dt}}=k_{e0}\times C_1-k_{e0}\times C_3$$

$$E=E_0-\frac{E_{\mathit{max}}\bullet C_3^{\gamma }}{E{C_{50}}^{\gamma }+C_3^{\gamma }}$$

at t = 0, C₃ = 0 and E = E₀.

For QT,

$$\frac{d{C}_3}{\mathit{dt}}=k_{e0}\times C_1-k_{e0}\times C_3$$

$$E=E_0+\frac{E_{\mathit{max}}\bullet C_3^{\gamma }}{E{C_{50}}^{\gamma }+C_3^{\gamma }}$$

at t = 0, C₃ = 0 and E = E₀.

PK parameters were used to predict PD parameters by fitting the data to the PK‐PD models using WinNonlin based on the mean plasma time courses. The effects of PD before drug administration exhibited no significant differences among nifedipine, captopril, and coadministration in both groups.

2.6. PK‐PD simulation

The PD parameters were obtained in a single administration group, and the PD models were used to stimulate BP, HR, and QT following coadministration. The results indicated that the PKs of both drugs were nearly linear within these dosage ranges. Therefore, PK parameter values for coadministration were used in the simulation study.

For BP and HR,

$$E_{\mathit{mix}}=E_0-\left(\left(E_0-E_{\mathit{\text{Nifedipine}}}\right)+\left(E_0-E_{\mathit{\text{Captopril}}}\right)\right)$$

For QT,

$$E_{\mathit{mix}}=E_0+\left(\left(E_{\mathit{\text{Nifedipine}}}-E_0\right)+\left(E_{\mathit{\text{captopril}}}-E_0\right)\right)$$

2.7. Sensitivity check

A sensitivity check was performed to confirm the sensitivity of PK and PD parameters obtained using the model analysis. The simulation results obtained using the calculated parameters of the model analysis were compared with the simulation results obtained by reducing each parameter to 50% or increasing it to 150% of the calculated value.

2.8. Statistics

Statistical significance was assessed using two‐sided Student's t‐tests, with statistical significance set at p < .05.

3. RESULTS

Figure 2 illustrates the plasma concentration‐time curves following 30‐min IV infusions of nifedipine (A) and captopril (B) during monotherapy and coadministration. The nifedipine and captopril doses during monotherapy and coadministration were 1.0 and 0.5 and 15 and 5 mg/kg, respectively. Plasma concentrations of both drugs increased during infusion, followed by 2‐compartmental degradation from rat plasma following infusion. Dashed curves were calculated using the dose during coadministration and monotherapy PK parameters. The plasma nifedipine concentration profiles were reduced during coadministration. The PK parameters are summarized in Table 1. For nifedipine, the AUC per dose during monotherapy and coadministration was 237 ± 85 and 127 ± 47 ng⋅min/mL, respectively. This resulted in an increase in CL_tot and a corresponding reduction in the MRT of nifedipine compared to monotherapy. For captopril, the AUC per dose during monotherapy and coadministration was 115 ± 66 and 82.2 ± 81.3 ng⋅min/mL, respectively. Moreover, the t_1/2 and V_dss of captopril increased following coadministration. However, none of the parameters listed in Table 1 exhibited significant differences between the monotherapy and coadministration groups.

FIGURE 2.

Fitted and observed plasma concentrations of nifedipine (A) and captopril (B) following IV infusion to spontaneously hypertensive rats (SHRs) during monotherapy (○) and coadministration with each other (●). Fitted curves are obtained according to a 2‐compartment pharmacokinetics (PK) model. Each point represents the mean + standard deviation (SD) of the data obtained from 3 to 5 experiments. Symbols. (A) ●; 0.5 mg/kg ○; 1.0 mg/kg. (B) ●; 5.0 mg/kg (○); 15 mg/kg. Lines. Solid: Fitting curves at 1.0 mg/kg dose (A) and 15 mg/kg dose (B) during monotherapy. Dashed: Calculated curves of 0.5 mg/kg from PK parameters of 1.0 mg/kg dose (A) and 5.0 mg/kg dose from PK parameters of 15 mg/kg dose (B). Bold solid: Fitting curves at 0.5 mg/kg dose (A) and 5.0 mg/kg dose (B).

TABLE 1.

Pharmacokinetic (PK) parameters of nifedipine and captopril following IV infusion to spontaneously hypertensive rats (SHRs) calculated through moment analyses during monotherapy and coadministration.

	Parameters	Nifedipine						Captopril
		Monotherapy			Combination			Monotherapy			Combination
		Mean		SD	Mean		SD	Mean		SD	Mean		SD
Dose	(mg/kg)		1.0			0.5			15.0			5.0
AUC	(μg × min/mL)	237	±	85	63.4	±	23.4	1730	±	411	411	±	407
t1/2	(min)	69.0	±	41.9	53.8	±	17.1	81.6	±	58.4	190	±	157
CLtot	(mL/min)	1.35	±	0.57	2.24	±	1.30	4.21	±	4.25	4.92	±	2.57
Vdss	(mL)	162	±	171	161	±	61	395	±	334	1565	±	1695
MRT	(min)	82.8	±	60.8	39.0	±	3.6	69.8	±	26.7	98.9	±	70.5

Note: Values are expressed as the mean ± standard deviation (SD) from 3 to 6 experiments.

Abbreviations: AUC, area under the plasma concentration‐time curve; CL_tot, total body clearance; MRT, mean residence time; t_1/2, terminal elimination half‐life, V_dss, volume of distribution at steady state.

Subsequently, PK parameters were calculated by fitting the plasma concentration profiles to a 2‐compartment PK model for PK‐PD analysis of both drugs. The plasma concentration‐time curves resulting from the monotherapy and coadministration of both drugs are illustrated in Figure 2. The C_max of nifedipine was overestimated during monotherapy (30 min infusion), whereas for captopril, it was underestimated. However, the fitting curves obtained using the model analysis approximately reflected the measured values. The calculated PK parameters are listed in Table 2. Although data were not demonstrated, both drugs were better fitted to the 2‐compartment model than to the 1‐compartment model. During coadministration, the V₁ of nifedipine was lower than that during monotherapy, and the values of k₁₀, k₁₂, and k₂₁ increased. The differences in PK parameters between monotherapy and coadministration were lower for captopril than for nifedipine. Akaike's Information Criterion (AIC) values of both drugs during coadministration were smaller than those during monotherapy. Additionally, the plasma concentration‐time curves following coadministration calculated from the PK parameters during monotherapy are illustrated in Figure 2 (dashed lines). Comparing the plasma concentration profiles of both drugs during the coadministration (closed circle) and simulation, the nifedipine plasma concentration during coadministration was slightly lower than that during the simulation. However, for captopril, the plasma concentration profiles were similar.

TABLE 2.

Pharmacokinetic (PK) parameters of nifedipine and captopril to spontaneously hypertensive rats (SHRs) calculated using a 2‐compartment PK model demonstrated in Figure 1 during monotherapy and coadministration.

Parameters		Nifedipine		Captopril
		Monotherapy	Combination	Monotherapy	Combination
V1	(mL)	54.9	26.7	54.2	41.7
k10	(min−1)	0.0218	0.0768	0.0521	0.0842
k12	(min−1)	0.0121	0.132	0.153	0.0631
k21	(min−1)	0.0165	0.0764	0.0694	0.0529
AIC		185.3	145.0	218.2	188.6

Note: Each value is calculated using the mean data obtained from 3 to 6 experiments; Doses of nifedipine are 1.0 and 0.5 mg/kg during monotherapy and coadministration, respectively; Doses of captopril are 15.0 and 5.0 mg/kg during monotherapy and coadministration, respectively; V₁: distribution volume of the central compartment, k₁₂, k₂₁, and k₁₀: first‐order rate constants for the pharmacokinetic process illustrated in Figure 1.

Based on the monotherapy PK results for both drugs, PK‐PD analyses were performed on the PDs of the drugs. In this study, we assessed BP as a drug effect, and HR and QT as secondary effects. Prior to drug administration, there were no significant differences in the BP, HR, and QT values among the three groups. The PD profiles during monotherapy are illustrated in Figure 3. The relationships between plasma drug concentrations and PDs at monotherapy are also indicated in Figure S1 (solid lines). Figure 3A and S1A indicate the results of BP. Our previous findings demonstrated that nifedipine exerted a potent antihypertensive effect, initially reducing BP following nifedipine infusion (E_max = 70.9 ± 8.6 mmHg), followed by a biological homeostatic function that results in increasing BP, despite the increase in plasma nifedipine concentration. Accordingly, a PK‐PD model considering this mechanism of nifedipine was reported ⁵ (Figure 1A), which was used in this study. In contrast, captopril did not induce similar hypotensive effects. This was milder than that of nifedipine and corresponded to the plasma concentration profile. Therefore, the common sigmoid E_max model with an effect compartment (Figure 1B) was used. The PD parameters of BP calculated through PK‐PD analyses are listed in Table 3. Figure 3B and S1B show the results of HR, which reduced gradually following nifedipine administration and reduced following a transient elevation (measured E_max = 325 ± 35 beats/min) after captopril infusion. Both nifedipine and captopril similarly reduced the HR after 60–90 min (approximately 250 beats/min). Figure 3C and S1(C) illustrate the results of QT. The extent of QT prolongation was similar for both drugs after approximately 60 min, which was similar to that of HR. However, nifedipine exhibited gradual QT prolongation, whereas captopril induced QT prolongation more rapidly. The common sigmoid E_max model (Figure 1B) was used to determine the effects of both drugs on HR and QT. PK‐PD analyses were conducted using the PK parameters listed in Table 2. The PD parameters calculated using PK‐PD analyses are listed in Table 3. Additionally, the difference in k_e0 values between the two drugs demonstrates the characteristics of their effect on QT. In addition, QT was corrected by the duration of the R‐R interval (QTc) because QT is influenced by the HR (Figure 3D and S1D). For QT correction, the standard Basett's equation (QTc = QT/(RR)^1/2) was used as reviewed in Goldenberg et al. ²⁰ The PK‐PD and QT analyses were performed after nifedipine and captopril administration. The calculated PD parameters of nifedipine were as follows: k_e0 = 0.0130 min⁻¹, EC₅₀ = 522 ng/mL, E₀ = 169 msec, E_max = 29.5 msec, and γ = 1.92. However, the PD parameters could not be calculated because of the transient increase of HR; thus, the quantitative relationship between captopril plasma concentration and QTc was not obtained.

FIGURE 3.

Pharmacodynamic (PD) responses associated with time following IV infusion of nifedipine and captopril to spontaneously hypertensive rats (SHRs). (A), blood pressure (BP); (B), heart rate (HR); (C), QT interval (QT) and (D), corrected QT interval by the RR duration (QTc). The nifedipine and captopril doses are 1.0 and 15 mg/kg, respectively. Symbols and bars indicate the observed mean value and its standard deviation (SD), respectively. (n = 8–10). Symbols. ●; nifedipine ○; captopril. Solid and dashed lines represent fitted PD effects of nifedipine and captopril, respectively.

TABLE 3.

Pharmacodynamic (PD) parameters of nifedipine and captopril in spontaneously hypertensive rats (SHRs) calculated using the pharmacokinetic (PK)‐PD models illustrated in Figure 1.

	BP			HR			QT
	Parameter			Parameter			Parameter
Nifedipine	ke0	(min−1)	0.135	ke0	(min−1)	0.006589	ke0	(min−1)	0.0117
	KEHS,in	(unit/min)	0.0217	EC50	(ng/mL)	115	EC50	(ng/mL)	3227
	kEHS,out	(min−1)	0.0119	E0	(beat/min)	306	E0	(msec)	75.8
	EC50,1	(ng/mL)	86.3	Emax	(beat/min)	63.3	Emax	(msec)	76.3
	EC50,2	(unit)	1450	γ		0.931	γ		0.857
	α	(mmHg/min)	1.17	AIC		268.9	AIC		191.1
	E0	(mmHg)	96.8
	Emax,1	(mmHg)	67.1
	Emax,2	(mmHg)	49 338
	AIC		151.4
Captopril	ke0	(min−1)	0.0400	ke0	(min−1)	1.21 × 10−5	ke0	(min−1)	4.05 × 10−5
	EC50	(ng/mL)	7885	EC50	(ng/mL)	7.6	EC50	(ng/mL)	4873
	E0	(mmHg)	105	E0	(beat/min)	312.8	E0	(msec)	81.2
	Emax	(mmHg)	41.8	Emax	(beat/min)	80.4	Emax	(msec)	41.9
	γ		0.662	γ		1.4	γ		0.1347
	AIC		213.2	AIC		299	AIC		220.0

Note: Each value is calculated using the mean data obtained from 8 to 10 experiments; Doses of nifedipine and captopril are 1.0 and 15.0 mg/kg, respectively.

Abbreviation: EHS; endogenous hypertensive substances; K_e0, first‐order rate constant of the effect compartment, K_EHS, _in and k_EHS, _out, zero‐ and first‐order rate constants associated with the formation and degradation of EHS; EC₅₀, drug concentration at half‐maximum effect; α, a slope of a linear equation, E₀, basic effect before drug administration; E_max, maximum drug effect; γ, Hill constant.

Subsequently, the PD profiles were assessed when the drugs were coadministered. The results are illustrated in Figure 4. Figure 4A–C demonstrate the BP, HR, and QT time profiles, respectively. In addition, the relationships between plasma drug concentrations and PDs at coadministration are also shown in Figure S1 (dashed lines). The PD profiles of nifedipine and captopril, simulated by PK‐PD analyses from the PK data following coadministration (Table 2) and calculated PD parameters (Table 3) are indicated by solid and dashed lines, respectively. The bold solid lines indicate the additive effects of nifedipine and captopril. Nifedipine exhibited a quicker and stronger antihypertensive effect than that of captopril monotherapy. As illustrated in Figure 4A, the alteration in BP during coadministration was nearly equal to that produced by nifedipine. Additionally, the maximum hypotensive effect (E_max = 75.0 ± 4.8 mmHg) was approximately 10 min following infusion initiation, despite the plasma concentration still increasing. However, after recovering from rapid BP reduction, a sustained reduction in BP was observed that was equal to or greater than the additive effect predicted from their PK. HR increased transiently (E_max = 356 ± 28 beats/min) following infusion initiation, similar to that with captopril monotherapy, and subsequently reduced (Figure 4B). The extent of this reduction was similar to that observed in the simulated profile for nifedipine. Captopril exhibited minimal effects on HR at this dose. The extent of QT prolongation (Figure 4C) was similar to the sum of the simulated effects of the plasma concentration profiles of nifedipine and captopril. However, the recovery period to baseline was faster than that of additive‐simulated QTs. Unfortunately, a quantitative relationship between the QTc value and plasma concentration of captopril could not be obtained after administration of captopril alone. Therefore, the QTc value during the concomitant use of both drugs could not be estimated.

FIGURE 4.

Pharmacodynamic (PD) responses associated with time following IV infusion of nifedipine and captopril to spontaneously hypertensive rats (SHRs). (A) Blood pressure (BP); (B) heart rate (HR); and (C) QT interval (QT). Nifedipine and captopril doses are 0.5 and 5.0 mg/kg, respectively. Closed circles and bars indicate the observed mean value and its standard deviation (SD), respectively. (n = 4–5). The solid and dashed lines represent the simulated PD effects of nifedipine and captopril, respectively, calculated from pharmacokinetics (PK) following coadministration. The bold solid lines indicate the effect as an additive of nifedipine and captopril.

4. DISCUSSION

Multiple‐drug combinations are used in clinical practice; however, possible interactions may occur, resulting in adverse effects and toxicity, enhancing or reducing effects, or inefficacy. Herein, we assessed the effects of coadministered antihypertensive drugs with different pharmacological mechanisms. Similar effects to those of the primary drug or additive/synergistic effects are possible. Therefore, simultaneous examination of the PK and several effects of the two antihypertensive agents with various pharmacological mechanisms was conducted in SHRs to assess the quantitative relationship between their PK and PD.

Nifedipine is a widely used antihypertensive drug that is highly selective for vascular smooth muscle. ²¹ In vivo studies indicate that ≥95% of nifedipine in rat plasma binds to proteins. The hepatic metabolism eliminates nifedipine from the body and predominantly metabolized by the microsomal CYP3A4 to its primary pyridine metabolite, dehydronifedipine. ¹² , ²² , ²³ , ²⁴ Therefore, nifedipine interacts with numerous drugs, ²⁵ and its PK varies depending on the pathology. ²⁶

For captopril, 30%–40% of the unaltered form is excreted in urine. Various metabolites of captopril, primarily its disulfide metabolite, constitute 5%–20% of the total drug following IV administration in rats. ²⁷ In healthy fasting individuals, approximately 70% of the oral captopril dose is absorbed, and the absolute bioavailability is approximately 60%. ²⁸

The obtained PK parameters of nifedipine and captopril agree with those of previously published studies in rats. ²⁶ , ²⁹ , ³⁰ The fitted plasma concentration‐time profiles generated using the PK parameters accurately reflected the experimentally observed values. Sensitivity analyses of the calculated PK parameters during monotherapy were performed to confirm their effects on plasma concentration profiles (Figure S2). Although k₁₂ and k₂₁ exhibited low sensitivity to the plasma concentration profiles, specifically for nifedipine, other parameters were effective in their profiles. k₁₂ and k₂₁ represent the transition process, and their values are smaller than k₁₀ for nifedipine. The PK interactions of a drug affect its disposition throughout absorption, distribution, metabolism, or elimination. In this study, the PK of nifedipine and captopril exhibited no significant alterations in PK parameters. Coadministration reduced the plasma concentration profiles of both drugs, followed by an increase in CL_tot and a reduction in the MRT of nifedipine. Additionally, these effects resulted in an increase in V_dss and a prolongation in the t_1/2 of captopril. However, these differences were not statistically significant. Given that the PK characteristics of the two drugs differed, a significant interaction was unlikely.

PD alterations resulting from PK interactions between antihypertensives lead to additive/synergistic effects or antagonism in BP reduction, or other adverse effects, depending on the agents. Antihypertensives used in patients with various pathological conditions and combination treatments increase the possibility of significant drug–drug interactions. In this study, each PD parameter in the PK‐PD analysis was calculated when nifedipine or captopril was administered individually. Sensitivity analyses of PD parameters were conducted to confirm their effect on PD profiles during monotherapy (Figures S3 and S4). The sensitivity of k_e0s to the BP of nifedipine and QT of captopril was low. Additionally, the EC₅₀ of captopril QT was low. The drug concentration at half‐maximum effect of E_max model 1 (EC_50,1) of nifedipine BP, and k_e0 and EC₅₀ of nifedipine HR exhibited low sensitivity to their profiles. Nonetheless, other parameters affected their PD profiles. Additionally, the transient increase in HR during individual captopril administration was observed during coadministration. Captopril mildly affected the HR values calculated from the PK profiles. The effect of both drugs on HR is considered a secondary effect, and the HR reduction profile was less than that of the additive effect simulated using both PKs. Nifedipine is known to cause tachycardia. ³¹ , ³² However, in this study, HR tended to decrease with the administration of nifedipine. Studies using pentobarbital‐anesthetized rats and isolated‐perfused rat hearts reported the reduction of HR by nifedipine administration. ³³ , ³⁴ Possibly, the reflex tachycardia induced by nifedipine observed in conscious animals was suppressed by anesthesia. HR slightly increased in the early stage after IV administration of captopril to SHR and normotensive rabbits ³⁵ ; furthermore, chronically higher captopril dosing mice had a trend for increased HR compared to that of control mice, ³⁶ although the cause is unclear.

Subsequently, the PD during coadministration was calculated by adding the PD calculated when both drugs were administered individually. However, a slightly different effect was observed. The effect on BP immediately following coadministration was similar to that of nifedipine alone. When nifedipine was administered individually, BP reduced rapidly, reaching its minimum at approximately 10 min following administration. Subsequently, BP began to recover despite continued infusion of the drug, likely because of homeostasis. Similar results were observed with coadministration; however, BP demonstrated a sustained reduction that was equal to or greater than the additive effect estimated from each PK following the recovery period from a rapid BP reduction. In contrast, the hypotensive effect of captopril was milder than that of nifedipine, and the increase and decrease were similar to those of the plasma concentration profile.

QT prolongation occurred more rapidly with captopril than that with nifedipine, reaching a similar levels after approximately 60 min. When coadministered, the QT prolongation was similar to the additive effect. The effects of both drugs on QT are anticipated as secondary effects. The recovery period from this effect during coadministration was faster than that anticipated from the plasma profiles of both drugs. In certain cases, drugs with similar pharmacological effects may exhibit smaller effects than those simulated from their PK profiles, with PD profiles resembling those of more potent drugs. This is possible because the physiology of the cardiovascular system tolerates rapid effects/adverse effects. ³⁷

In healthy humans, nifedipine exhibits a high protein‐binding ratio of 92–98%. ³⁸ In addition to albumin, alpha 1‐acid glycoprotein is crucial in the plasma binding of nifedipine. ³⁹ Nifedipine is completely metabolized in the liver to numerous inactive metabolites, ⁴⁰ , ⁴¹ with CYP3A4 being the primary isoform involved, ¹² which is also found in rats and dogs. ⁴² Although over 90% of the nifedipine dose is absorbed from the gastrointestinal tract following oral administration, a significant portion (30%–40%) is subjected to first‐pass metabolism. ⁴¹ The protein‐binding ratio of captopril in human plasma is approximately 23%–30%, which is lower than that of nifedipine. ⁴³ In humans, captopril is eliminated through the kidneys and metabolism. ⁴⁴ Numerous inactive metabolites are produced in the liver, with the cysteine‐mixed disulfide form constituting the primary metabolite. ⁴⁵ Approximately 40% (cumulative urinary excretion ratio) of the administered drug is excreted unaltered in the urine for 24 h, whereas various metabolites constitute approximately 26%. ⁴⁶

The coadministration effect of nifedipine and captopril in patients with essential hypertension was assessed. ⁴⁷ While nifedipine‐induced reduction in BP was significantly greater than that induced by captopril, with each drug reducing the mean BP during monotherapy, coadministration significantly reduced BP. However, the combined dose of drugs was similar to the individual dose. In this study, it was observed that the coadministered dose was approximately 1/2 to 1/3 of the individual doses. Although SHR was used in this experiment, the dose of each drug in combination was reduced to ensure that the BP reduction during combination therapy was similar to that during monotherapy. There was no PK interaction when an ARB (telmisartan) was combined with a diuretic (hydrochlorothiazide) in SHR. ⁴⁸ However, a synergistic antihypertensive PD interaction was observed between these drugs that affected BP. This effect was simulated from the PK of telmisartan and hydrochlorothiazide using a PK‐PD model following prolonged coadministration. The interactions on PK affect its absorption, distribution, metabolism, and elimination. In contrast, the interactions on PD result in synergistic/antagonistic effects of BP‐lowering or other adverse effects of drugs, depending on the agents.

Herein, urethane, a long‐lasting and stable anesthetic with minimal cardiovascular and respiratory system depression, ⁴⁹ , ⁵⁰ , ⁵¹ , ⁵² was used to monitor the ECG profile. The produced anesthesia was sufficient to allow surgical procedures in small rodents. Pharmacokinetic studies must be conducted on anesthetized animals because stressful conditions must be avoided. However, urethane inhibits the hepatic metabolism of CYP3A4. Nifedipine is metabolized by CYP3A4; thus, it can inhibit nifedipine hepatic metabolism, ⁵³ , ⁵⁴ representing a potential confounding factor here. The PK of voriconazole, a triazole antifungal agent metabolized by hepatic CYP3A4, was investigated in awake and urethane‐anesthetized rats ⁵⁵ without observing statistical differences in the t_1/2 and CL_tot of voriconazole. However, the PK of nifedipine may be affected by urethane, making it necessary to predict pharmacological effects based on the PKs under awake conditions. When pharmacokinetic interactions occur, the PDs of each drug at coadministration can be predicted using the PK parameters. The baseline value of BP was lower than that of SHRs in waking conditions. ⁵⁶ Mean BP of Wistar rats and SHRs were 72.6 and 97.8 mmHg, respectively, under anesthesia. ⁵ Other PD parameters showed no significant difference between SHR and Wistar rats. In this study, urethane decreased the BP of rats, which in anesthetized animals cannot fully reflect that of the awake condition. However, basic knowledge of the relationship between the drug concentration profile and the associated PD at monotherapy and coadministration can be obtained.

5. CONCLUSION

We constructed mechanism‐ and evidence‐based PK‐PD models for two antihypertensive drugs with different pharmacological mechanisms. Additionally, we assessed the relationship between PK and the several effects of these drugs at coadministration using PK‐PD analysis. The combined effects predicted from the PK profiles differed for each effect and over time, which appeared as an effect of the more potent drug or an additive effect of both drugs. When HR and QT were considered secondary effects, the results demonstrated that the secondary effects were less than those calculated as the additive effect of both drugs. Additionally, an in vivo homeostasis mechanism is involved in the time‐profile of these effects.

AUTHOR CONTRIBUTIONS

Kiriyama: Participated in research design and conducted experiments. Kiriyama and Kimura: Contributed analytic tools. Kiriyama and Yamashita: Performed data analysis. Kiriyama and Yamashita: Drafted or contributed to the writing of the manuscript.

CONFLICT OF INTEREST STATEMENT

None of the authors report a conflict of interest concerning this article.

ETHICS STATEMENT

All animal experiments adhered to the Guidelines for Animal Experiments of Doshisha Women's College of Liberal Arts, which comply with the Fundamental Guidelines for Proper Conduct of Animal Experiments and Related Activities in academic research institutions (Ministry of Education, Culture, Sports, Science and Technology, Notice No. 71) and Guidelines for Proper Conduct of Animal Experiments (Science Council of Japan).

Supporting information

Figure S1.

Figure S2.

Figure S3.

Figure S4.

Kiriyama A, Kimura S, Yamashita S. Exploring the multiple effects of nifedipine and captopril administration in spontaneously hypertensive rats through pharmacokinetic‐pharmacodynamic analyses. Pharmacol Res Perspect. 2024;12:e1249. doi: 10.1002/prp2.1249

DATA AVAILABILITY STATEMENT

The authors declare that all the data supporting the findings of this study are contained within the paper.