*2.7. In Vitro Dissolution*

Fasted state-simulated intestinal fluid (FaSSIF) was prepared according to the recommended recipe from the supplying company [45].

To determine the equilibrium solubility of the drug in SGF, we used phosphate buffer, pH 6.8, and FaSSIF, a saturation shake-flask method as follows: An excess amount (100 mg) of nifedipine was added to 5 mL of each buffer. The samples were protected from light and stirred in a water bath at 37 ◦C for 96 h. Samples were then centrifuged at 2000 RCF for 30 min and analyzed with HPLC. Each measurement was performed in triplicate.

For in vitro dissolution testing, aliquots of formulations were filled in hydroxypropyl methylcellulose capsules. The dosage of all formulations were selected to achieve a 5 mg/L concentration of nifedipine in the dissolution vessel. The dissolution experiments were carried out in a USP II apparatus (Sotax, Aesch, Switzerland) in 900 mL of medium [44] with stainless steel sinkers to prevent capsule floatation. The temperature was set to 37.1 ± 0.5 ◦C and the paddle speed was 100 rpm. Nifedipine concentration was measured by UV–VIS (Ultraspec 3100 pro, Amersham Biosciences, United Kingdom) at a 236 nm wavelength with 1-min software-controlled intervals (DissoControl v1.0, University of Basel, Basel, Switzerland). In parallel, to reduce errors introduced by possible phospholipid structures (i.e., lipid vesicles), for quantification of the dissolved nifedipine, we made manual samplings. Manual sampling was done after 5, 10, 15, 30, 60, 90, 120, and 150 min. To eliminate any undissolved nifedipine crystals in the dissolution aliquotes, we centrifuged an amount of 2 mL of the samples at 37 ◦C and 2000 RCF for 15 min. A volume of 500 µL of the supernatant was diluted with an equal volume of methanol to dissolve phospholipid vesicles. Removed volumes were replaced with fresh dissolution medium.

Assuming sink condition in the dissolution vessels, we fitted the dissolution profiles to Equation (1):

$$\mathcal{C}\_t = \mathcal{C}\_d - a e^{-kt} \tag{1}$$

where *C<sup>t</sup>* (*mg*/*L*) is the concentration of nifedipine that is dissolved in the medium at time *t* (*s*), *C<sup>d</sup>* (*mg*/*L*) is the concentration of the nifedipine after the dissolution reached a plateau, *a* is the scaling constant, and *k* is the dissolution rate constant. In order to obtain the dissolution rate constant value for all the tested formulations, we fitted the experimental dissolution curves to Equation (1). The fittings were done in Wolfram Mathematica software, version 12 (Wolfram, Oxfrodshire, UK, 2019). To measure the goodness of fit, we used the standard errors for parameter estimates from the default Non-linear Model Fit function of the Mathematica software [46]. Two-way ANOVA and Bonferroni post-hoc test were performed (OriginPro 2018, OriginLab, Northampton, MA, USA) on the obtained dissolution rate constants with significance levels of 0.05 and 0.005, respectively.

The dissolution efficiency (*DE*) of the formulations in the first 120 min of the in vitro dissolution was calculated according to Equation (2) [47]:

$$DE = \frac{\int\_{0}^{120} D\_{120} dt}{D\_{\text{max}} 120} \times 100 = \frac{AUC\_{0-120}}{D\_{\text{max}}} \times 100\tag{2}$$

where *D*<sup>120</sup> is the amount of drug dissolved at 120 min (%), *Dmax* is the maximum drug dissolved at the end of the dissolution (%), and *AUC*0–120 is the area under the dissolution curve between the times of 0 and 120 min.
