Sex-related in vitro/in vivo and PK/PD correlations after oral single dose furosemide administration

Background: The goal of this study was to develop an in vivo-in vitro (IVIV) correlation, both in men and women, which allows constructing a model to predict bioequivalence assessments for drugs with narrow absorption windows. Besides, pharmacokinetic and pharmacodynamic equivalences were also investigated. Furosemide was chosen as a prototype. Methods: Twelve healthy Caucasian volunteers (8 women and 4 men) participated in a relative bioavailability study. Two oral formulations [Lasix® (Reference, R) and Furosemide EFA® (Test, T)] were administered under fasting conditions. Urinary excretion of unchanged drug (PK), and of chloride, sodium and potassium (PD) wasmonitored throughout time. PK and PD parameters were calculated from each respective excretion rate versus time curve. In vitro dissolution testing of both formulations was carried out using the USP apparatus 2 and 4 with fixed and variable dissolution media. Results: T and R could be considered bioequivalent since the 90% confidence intervals for the T/R ratio of geometric means for the area under the urinary drug excretion rate versus time curve and for the maximum excretion rate were within the 0.80-1.25 bioequivalence interval. However, T had faster initial absorption and higher levels in women, while R displayed such characteristics in men. Closer IVIV correlations in women were obtained when apparatus 4 with variable biorelevant dissolution media were used [going from fasting state simulated gastric fluid to fasting state simulated intestinal fluid]. Since R had faster disintegration time than T, a shorter stay of R under gastric conditions was required in order to obtain a good IVIV correlation in men. Saluretic effect displayed a typical clockwise hysteresis loop for the PKPD correlation assessed through chloride-versus-furosemide urinary excretion rates. Even though a higher amount of furosemide was excreted with the urine in men, differences in the excretion of electrolytes between sexes were almost negligible. Conclusions: Sex-differences in the gastrointestinal transit of formulations, under fasting conditions, determined the extent and the rate of furosemide absorption. The prolongation of the absorption process by mean of slowing the gastric emptying would make the formulation more effective. The USP-4 apparatus with variable dissolution media was able to discriminate the formulations even between sexes, becoming a promissory in vitro dissolution testing to predict bioequivalence.


Introduction
Furosemide is a loop diuretic that is commonly used in the treatment of edematous states associated with cardiac, renal, hepatic failure and in the treatment of uncontrolled hypertension with abnor- mal renal function. Its oral bioavailability is low (67%) even if a solution of 40 mg is given to healthy male subjects [1]. In a study carried out with women and men [2], absolute bioavailability under fasting state was reduced to 51% revealing a sex-related influence on the oral absorption of furosemide. Sustained-release formulations rendered even lower drug absorption [3,4]. Switching from intravenous bolus to oral slow release dosage forms, more than three quarters of the administered dose were not absorbed. Therefore, drug dissolution might play a significant role in furosemide absorption as it was previously reported [5,6]. Further extensive information about furosemide oral absorption and an appropriate background for this work can be found in Magallanes et al., 2015 [7].

Materials and methods
Equipment GFL 2008 distillation apparatus (Germany) to provide distilled water, andOakton ph6 pH-meter (IL, United States) to adjust the pH of dissolution media were used.
To perform in vitro dissolution testing the following equipment was used. USP apparatus 2 (paddle method): Distek® dissolution system 2100C equipment, configured with a model 89092EO Agilent peristaltic pump, and coupled with spectrophotometer Agilent 8453. A software ChemStation® CPU (Agilent) controlled all the dissolution testing. USP apparatus 4: a Sotax CE 7 smart USP-4 flow through dissolution system connected to a Sotax CP-7 35 Piston Pump (Sotax, Switzerland), and collected samples assayed by a spectrophotometer UV-Vis Thermo Spectronic, (model HeliosGamma, Thermo Scientific, UK).
In vitro disintegration times of formulations were measured with the USP disintegration apparatus (U.S. Pharmacopeial Convention, 2014) provided by local vendor (Tecnolab®), without disks. Individual disintegration times were manually recorded once dosage forms were completely broken down.
Furosemide urine concentrations were measured by high performance liquid chromatography coupled with ultraviolet detector (HPLC-UV) with Dionex® Ultimate 3000 system equipped with a Phenomenex® Luna C18 reverse phase column (5 μm, 150x4.6 mm) kept at 40°C.

In vitro disintegration at pH 2.5
Six units of each product were tested for disintegration according to the United States Pharmacopoeia (USP30-NF35, physicochemical methods <701>) using the simulated gastric fluid without enzyme (aqueous solution of HCl and KCl, pH 2.5).

In vitro dissolution with USP-2 apparatus (pH 4.6)
Dissolution of six units of each formulation was performed in medium acetate buffer pH 4.6, at 37±0.5°C, volume 900 mL; stirring speed 50 rpm, sampling times: 5, 10, 15, 20, 30, 40, 60, and 90 min. Quantification of furosemide concentrations in the dissolution samples was carried out by UV-absorption at 277 nm wavelength based on calibration curves in the dissolution medium.

In vitro dissolution with USP-4 apparatus
Dissolution studies of the T and R formulations were performed 3 doi: 10.7243/2050-120X-5-2 with the flow through system operating in the open mode at 37±0.5°C. A 5mm-size glass bead was positioned in the tip of the Ø 22.6mm cell, the cone was filled with 1mm-size glass beads and on the top of the cell a Whatman® glass fiber filter (GF/F: 0.7μm pore size) was placed. Formulations were placed on the tablet holder.

Dissolution at pH 4.6
The assay was carried out in medium pH 4.6 acetate buffer at 37±0.5°C. Flow rate was set at 16 mL/min in order to allow the formulation to be in contact through 60 minutes with approximately the same volume of solvent as in the USP-2 assay (960 mL vs 900 mL). Samples were collected every 15 minutes and quantified as it was indicated under the dissolution with USP-2 pH 4.6 section.

Dissolution in variable biorelevant media from-1.6-to-6.5 pHs
Dissolution of furosemide from the formulations was monitored in triplicate using biorelevant media in order to simulate the in vivo dissolution of both formulations under fasted state [8,9]. Sequential change of both the dissolution medium and the flow rate was performed: from FaSSGF pH 1.6 at 8 mL/min for 60 min to FaSSIF-V2 pH 6.5 at 4 mL/min for 90 min. Samples were collected every 15 minutes. Furosemide concentration in the dissolution sample was quantified by an HPLC method adapted from Sora et al., 2010 [10], using methanol:formic acid 0.1% (v/v) 60:40 as mobile phase, 0.8 mL/min as flow rate, 25°C as run temperature, 20µL as injection volume, and 233nm as detection wavelength.

Subjects and in vivo study design
Twelve Caucasian healthy volunteers (eight women and four men) between 21 and 37 years old with mean body weight (SD) of 64 (5) and 77 (21) kg, respectively, were enrolled. The study was carried out with one tablet (40 mg) of the formulation T, or the formulation R, administered under fasting conditions with 200 mL of water, in a randomized, two-period, two-sequence (TR and RT), balanced crossover design. A one-week washout period was kept between both administrations. The study protocol and informed consent form were designed according to the ethical guidelines for human clinical research and were approved by the Institutional Ethics Review Committee of the Faculty of Chemistry (Uruguay). Written informed consent was obtained from all subjects before their entry in the study. The study was performed in the Bioavailability and Bioequivalence Centre for Medicine Evaluation, situated in ''Dr. Juan J. Crottogini'' Hospital (Montevideo, Uruguay).
Volunteers came to the Centre the first day of each week with an eight-hour overnight fasting period. Urine samples were collected before dosing and at 20, 40, 60, 80, 100, 120, 140, 160, 180, 240, 300, 360, 420, 480, 600, and 720 min after dosing. The volume of urine was registered and two aliquots, one for electrolytes determination (chloride, sodium, potassium) and the other for furosemide quantification, were stored until analysis, refrigerated (2-8°C) and frozen (-20°C) respectively. Standardized meals (lunch, tea, and dinner) were provided at 5, 8, and 12 h after dose administration. From 40 to 180 min post-dose, and following each urine sampling, 70 mL of rehydration fluid, prepared after water dilution of Rehidron®, was given to the subjects. From 3 h post-dose, volunteers were rehydrated with 100 mL of fluid every hour after each urine sample collection. The fluid contained 15 mmol/L sodium, 4 mmol/L potassium, 13 mmol/L chloride, 15 mmol/L glucose, 2 mmol/L citrate.

Chemical analysis in urine
Quantification of furosemide in urine was carried out by HPLC, based on the method described by Abou-Auda et al., 1998 [11] with minor modifications. Briefly, the assay was a liquid-liquid extraction of furosemide (with n-hexane : ethyl acetate [50:50]) from the urinary matrix (250 µL) by the addition of 250µL of phosphoric acid 2.5M and 50 µL of a methanolic solution of internal standard (phenytoin, 40 µg/mL). Ten seconds of gentle agitation by vortex, separation of the organic phase, and evaporation until dryness under stream of nitrogen. Reconstitution of the residue was performed with mobile phase (buffer phosphate 0.02M pH 3.5 : acetonitrile [65:35]). A volume of 20 µL was injected onto the reversed-phase column and eluates were monitored at 230 nm. Retention times of analytes were: 5.5 min (furosemide) and 7.7 min (phenytoin). Linearity was assessed between 0.5 to 20µg/mL of furosemide in urine. The lower limit of quantification was 0.5µg/mL, since intra-and-inter-day precision was below 20%, in terms of coefficient of variation, and accuracy. Otherwise, precision and accuracy was comprised between±15% and within the 85-115% interval, respectively.
Chloride, sodium, and potassium electrolytes were measured accordingly with the instructions given by the manufacturers (Roche/Hitachi, package insert).

Pharmacokinetic (furosemide) and pharmacodynamic (electrolytes) analysis
The volume of urine was multiplied by furosemide and ions concentrations in urinary samples in order to obtain the amount of analyte (ΔA) in each interval of time (ΔT). By summing all ΔA, the total amount excreted in the urine of each analyte was obtained for the 12-hour study period (E 0-12 ). Urinary excretion rates for each analyte (ER) were calculated dividing ΔA by ΔT and the result was assigned to the middle of the ΔT interval. Urinary excretion rates of electrolytes were recorded as the pharmacodynamic responses (PDs) of the formulation given, and urinary excretion rate of furosemide as the corresponding pharmacokinetic response (PK). Non-compartmental PK/PD analysis for each analyte was performed over each ER versus time curve (Microsoft Office Excel 2007 software). The maximum ER (ER MAX ) and the time-to-peak (T MAX ) were recorded from experimental data for each volunteer and analyte. The area under the ER-time curve from zero to infinite (AUC INF ), 4 doi: 10.7243/2050-120X-5-2 for furosemide, was calculated using the trapezoidal rule until the last experimental time (AUC 11-hour ), and extrapolated to infinite adding the term ER 11 /k EL , being k EL the first order elimination rate constant calculated from the slope of the terminal log-linear ER-time regression of data. AUC INF corresponds to the total amount of furosemide excreted in urine (E INF) . Furosemide half-life (T 1/2 ) was calculated as Ln(2)/k EL .

Statistical analysis
ER versus time curves for both electrolytes and furosemide aregraphed as mean values±standard error (SE). However, electrolyte versus furosemide ERs are presented as mean values±standard error of the electrolyte only, since it is set to assess if the same pharmacokinetic response could be associated with different pharmacodynamic response. If there was a significant difference (p<0.05) between two excretion rates of the electrolyte, then a significant hysteresis for the PK/PD relationship was assessed.
Pharmacokinetic parameters for furosemide (ER MAX , E 0-12 , and E INF ), in logarithmic scale, were processed by analysis of variance (ANOVA, Microsoft Office Excel 2007 software) considering subjects, sequences, periods and treatments as variation sources. Coefficient of variation (CV) of the ANOVA was calculated according to the equation (1): With being the residual variance of the ANOVA performed on the log-transformed parameters.
A T-Wilcoxon test was used to evaluate the T-R difference for T MAX , since a non-normal distribution was assumed. Average bioequivalence between T and R, considering all the subjects, was declared if the 90% confidence intervals (90% CI) for the T/R ratio of the geometric means for each parameter were within the range of 0.80-1.25, and T MAX did not differ significantly.

In vitro/in vivo and pharmacokinetic/pharmacodynamic correlations
Since only two formulations were assayed, no mathematical correlation will be performed between the in vitro dissolution and in vivo pharmacokinetics of furosemide. Just visual comparisons between percentages dissolved in vitro and percentages of doses recovered in urine from both formulations, in both male and female subjects, will be carried out.
Diuresis and/or electrolyte ER versus furosemide ER will be correlated in order to assessPK/PD relationship in both sexes.

In vitro studies
Dissolutions carried out with the USP-2 apparatus revealed that formulation Tremaine at the bottom of the vessels up to its total dissolution, even though it dissolved faster than R. Reference disintegrated immediately it was plunged into the dissolution medium (pH 4.6). Disintegration in simulated gastric fluid (pH 2.2) was significantly different between the (Equation 1) two formulations, with the formulation T having a higher disintegration time than the formulation R(180±10vs.22±2 sec).
Dissolutions performance of the two formulations under all conditions tested are presented in Figure 1. Dissolution of furosemide from T is faster and more complete than from R after 1h in the acetate buffer with USP-2 and USP-4 apparatus. In biorelevant conditions with USP-4 apparatus, a negligible dissolution is observed in simulated fasted state gastric conditions for both formulations. Furosemide dissolves slightly faster from T in the first part of the simulated fasted state intestinal conditions, having a complete dissolution at the end of the experiment from both formulations.

In vivo pharmacokinetic and pharmacodynamic study
The mean ER of furosemide throughout time for both formulations in all the subjects, in men, and in women are presented in Figure 2. Table 1 summarizes the mean pharmacokinetic parameters obtained.
Parameters for all electrolytes are summarized in Table 2. Since there was a highly significant linear correlation between ER of urine (diuresis) and ER of chloride (data not shown), this was selected as the PD response of formulations to be presented graphically (Figure 3). A clockwise hysteresis loop was evident in both sexes, being significant in women.   cant difference between formulations was obtained for T MAX . Considering the sex of individuals, E 0-12 , E INF and ER MAX differed significantly (p<0.01 in both formulations, mean and SE values in Table 1). Half-life calculated in men (average between formulations=2.07 h) was lower than that in women (average between formulations=2.24 h), but without a statistical significance.

Sex-related differences in the gastrointestinal tract
Regarding the bioequivalence between formulations, T could be assessed as bioequivalent with R. The extent of absorption in the total subjects (men and women) was similar between both products as revealed by the 90% CI of E INF T/R geometric mean ratio being between 0.80 and 1.25. Similarly, the absorption rate of formulations was assessed as bioequivalent considering the T/R geometric mean ratio of ER MAX (see Table 3) and T MAX did not differed significantly. Figure 2 (upper panel) supports this average bioequivalence judgment.
However, when only women were considered, bioequivalence could not be assessed because both E INF and ER MAX 90% CI of T/R ratios moved to the right and could not be included  within the bioequivalence interval. Even though no confidence intervals were constructed for male subjects, since only four individuals were enrolled, a faster beginning of drug absorption and a higher maximum excretion rate of furosemide from R, with similar-to-the-T amount of drug excreted (see T-R relative values of T MAX , ER MAX and E INF in Table 1) was shown. This is contrary to the performance observed in women, and put in evidence that sexes might discriminate these formulations differently.   . In the case of men, and for acidic drugs like furosemide, as their higher strength of gastric contraction promotes shorter intervals between each gastric discharge, the faster the formulation disintegrates, the faster its intestinal dissolution starts. Women, with longer intervals of gastric discharge, are unable to differentiate the passage of formulations from the stomach to the intestine even though they could have some differences in their disintegration times. Therefore, a good correlation between the in vitro dissolution and the in vivo absorption of furosemide would be expected in women. Figure 4 displays similar in vitro/in vivo shapes for the cumulative T-R differences in dissolution and in urinary excretion of furosemide (upper and lower panel, respectively).

T MAX (h) a ER MAX (mmol/h) E 0-12 (mmol)
The use of sequential change of biorelevant media and physiological relevant flow rates with the flow through cell apparatus led to accurate predictions of the in vivo absorption of furosemide after oral administration of the two formulations (Figure 1). Clearly, not the apparatus, but the condition of the in vitro drug dissolution was the main cause for predicting accurately the in vivo absorption of formulations. The use of a compendial dissolution medium (acetate buffer pH of 4.6) with the paddle and the flow through cell apparatus was over discriminative. A change in the dissolution conditions passing from gastric to intestinal environment resulted more predictive and realistic for woman subjects instead. Results obtained in men could be reproduced in vitro by simply shortening the gastric residence of the formulation R (Figure 5). In this way, the passage of the R granules to the duodenum, where furosemide dissolution is more effective and its absorption more efficient, was faster (Figure 2, middle panel).
In terms of the amount absorbed in both sexes, it can be seen that a similar amount of furosemide is absorbed from both formulations in man, whereas the fraction of dose absorbed from the two formulations is different in women. It is known that women have a more alkaline intestinal environment than men [14]. This finding explained the sex-related amount of ketoprofen absorbed from extended release formulations and the recirculation rate of its stereoisomers [15]. According to this, it could be hypothesized that women present a shorter absorption window for furosemide (pKa 3.8, [16]) in comparison with men (vertical line in Figures 4 and 5 represent the limit of time above which the absorption would be negligible, in women and men respectively). For this reason, furosemide could be absorbed to the same extent from both formulations in men. Conversely, since the absorption of the drug drastically diminishes short time after its appearance into the intestinal tract of women, this sex could not allow R (slower dissolution rate) to be absorbed to the same extent as T (faster dissolution rate).

Sex-related differences in drug absorption and kidney blood perfusion
Diuretic (saluretic) effect of furosemide is the result of its concentration at the action sites in the ascending limb of the loop of Henle [17]. Arterial blood stream delivers the drug to the kidneys (glomerulus, proximal convoluted tubules and loop of Henle, among others sections of the nephron). One portion of the diffused amount in the renal parenchyma filters through the glomerular membrane. The other part enters the cells of the proximal tubule, of the ascending limb of loop of Henle that contain the saluretic action sites, and the other cells of the nephron. One fraction of this diffused amount returns to 8 doi: 10.7243/2050-120X-5-2 the venous blood stream, while the complementary fraction pursues its renal excretion and metabolism (mainly from the proximal convoluted tubules). Venous blood collects the drug at the concentration it has in the interstitial space of the tissue. In the case of kidneys, this drug concentration is the one it had in the lumen of renal tubules before the current cycle of the circulatory system, mainly from the proximal tubule and the descending limb of the loop of Henle where reabsorption is maximal. In this scenario, and while drug absorption is proceeding, furosemide plasma concentration in the venous blood coming from the kidney would have lower value than in the arterial blood that reaches the organ. There will be a progressive reduction of the arterial-venous concentration difference throughout the absorption process up to the end of drug entrance. After this, the drug will only pursue its disposition throughout the body. Each time the absorption resumes, a new increase in the arterial-venous difference occurs. Theoretical and experimental research dealt with this physiological phenomenon and with its consequence on drug action [18,19].
According to this, furosemide would act before it could be excreted, since its arrival at the action sites from the capillaries took lesser time than its arrival at the end of the luminal space of renal tubules from the glomerulus. For this reason, it is not surprising to observe a clockwise hysteresis loop for the relationships between chloride and furosemide ERs (see Figure 3). This resembles what was observed when drug effect, or arterial drug concentration, versus venous drug concentration was graphed for nicotine [19]. Accordingly, furosemide ERs would surrogate their venous plasma concentrations.
This finding was traditionally interpreted in the literature as an acute tolerance to the action of furosemide following single dose administration, mainly when extended release drug formulation or sustained intravenous administration were given [3,4,[20][21][22][23]. As it was informed, no hysteresis or a counter-clockwise hysteresis was found after an intravenous bolus dose, revealing, this fact, that there would be a delay between drug concentration in venous blood and drug concentration at the action site (biophase), probably due to the extremely fast drug absorption, and thereafter absorption duration, in comparison with drug diffusion towards the tissue. However, when oral formulations were administered [3,4,20], drug release from the dosage form was slow enough to display such sequential steps in the arterial-biophasevenous concentration. The slower the absorption process is, the wider the clockwise hysteresis loop becomes.
Our results clearly show a difference in the absorption time between sexes. As it can be seen in Figure 3 (upper panel), men displayed a more pronounced hysteresis loop after R than after T administration, even though both PK/PD loops have not reached significance (probably due to the low number of subjects). Once the formulation T passed to the duodenum, its rapid dissolution practically avoided the hysteresis loop. Conversely, repetitive passages of granules from R (tablet fragments) allowed the hysteresis loop to appear. Women slower gastric emptying increased the width of the hysteresis loop significantly (p<0.01), independently of the formulation given (Figure 3 lower panel). Therefore, furosemide absorption kinetics in women was governed by the gastric emptying, and not by the formulation.
As Figure 3 shows women and men do not have the same slope for the PK/PD relationship (see both graphs superimposed in the upper panel of Figure 6). Men had a slower slope. It has been reported [24] that men have a higher percentage of cardiac output destined to the kidneys (19% vs 17%). So, a higher percentage of molecules in men (12%) are driven to the kidneys every blood circulation cycle, and then, their parenchyma would be additionally loaded with furosemide to an extent of 12% in relation to female subjects [25]. If the furosemide ERs were reduced by a factor of 1/1.12, drug excretion rate of men would be adapted to that of women. In other words, if plasma drug concentrations and kidney/blood concentration ratios were similar between women and men, 9 doi: 10.7243/2050-120X-5-2 both PK/PD graph should have identical slope provided no difference in their pharmacodynamics would exist. However, even though this transformation was performed sex differences in the PK/PD slope persisted.
Up to now, we assumed that the pharmacodynamics of furosemide did not affect its pharmacokinetics, as if both PK and PD variables were independent. However, it should be born in mind that a higher renal blood flow fraction means not only higher drug concentration at the action site [26-28] but also at the luminal space of tubules, from where urinary excretion of furosemide will take place as a consequence of its own diuretic effect. So, a double contribution of such higher renal cardiac output should be considered to transform the furosemide ERs in men. Then, a straightening of 25% was applied on the furosemide ER axis (values multiplied by a factor of 0.80 [=1/1.25]), and the result can be seen in the lower panel of Figure 6. Now, a good PK/PD correlation was achieved regardless the sex of the individuals. This transformation simply means that men would have a 25% higher excretion renal clearance than women.
However, it should be kept in mind that the intensity of furosemide action is not linearly related with its concentration, and then, the assumption of a constant factor of 0.80 for the axis straightening should not be entirely appropriate. Renal excretion of furosemide might not be constant, as it is influenced by its diuretic effect. No data about sex differences in the excretion renal clearance (CL excr ) of furosemide was published in the literature, but it seems reasonable to admit a higher renal clearance for men taking into account similar findings for other drugs or endogenous substances highly excreted in urine [29,30].

Actual sex-related differences in furosemide oral bioavailability
Previous paragraph leads us to consider whether the urinary excretion of furosemide was reliable enough to assess oral bioavailability difference between sexes. Regarding the values of E INF from Table 1, a difference between 16.7 and 9.49 mg for the Reference seems too high in favor of men.
Taking into account that drug plasma concentration (C) and its urinary excretion rate (ER) are related by the equation 2, ER MAX , E 0-12 , and E INF data for men, shown in Table 1, should be divided by a factor of 1.25 (man-to-women ratio of excretion renal clearances) in order to be compared with those for women.
(Equation 2) These transformed (for men) and non-transformed (for women) values (i.e., 13.4 mg [=16.7/1.25] and 9.49 mg for E INF , respectively) would still maintain sex differences in bioavailability and other clearances apart from the excretion renal one. Assuming that only the bioavailability remained as source of variation, a women-to-men percentage for the absorbed dose would be 75%. So, a supplementary loss of 25% of the administered dose might be expected for women in relation with men, which could be caused by their higher intestinal pH and their shorter absorption window.Our estimation is now more in agreement with Waller et al., (1982) [1] and Hammarlund et al., (1984) [2] findings.

Conclusion
Men and women have dissimilar furosemide oral bioavailability probably due to their differences in the intestinal pH, which in turn produces different absorption windows. The particular intestinal environment that women have was able to differentiate the extent of drug absorption of two immediate-release formulations marketed in Uruguay on the basis of their dissimilar drug dissolution. Conversely, in men drug was absorbed at the same extent because of the longer absorption window available for furosemide. Due to their faster gastric emptying, men were able to differentiate the beginning of drug absorption from both tablets, maybe because of their dissimilar disintegration times during the stay in the stomach. Even though both tablets could have been discriminated by women and men, they are assessed as average bioequivalent considering all the subjects. Minimal differences between formulations were detected both in doi: 10.7243/2050-120X-5-2 in vivo and in vitro assays, arriving to a good in vivo-in vitro correlation when the USP-4 apparatus under biorelevant conditions was used.
Similarly, the pharmacodynamics of both formulations was equivalent. Once again, different PKPD correlation related with the sex of individuals could be seen. Such sex differences might be explained on the basis of their different renal cardiac output distribution, which influenced their renal excretion as well. Because its diuretic effect, the pharmacodynamics of furosemide interacts with its own pharmacokinetics by increasing its renal excretion. Finally, the excretion rate of chloride is correlated with the excretion rate of furosemide, showing a clockwise hysteresis loop that relates to the rate of drug absorption. In the case of immediate-release dosage forms, the gastric emptying of individuals would control the width of the hysteresis loop more than the formulations themselves, as women and men showed.