All chemicals and reagents used are of chromatographic purity. Polyvinylpyrrolidone reference standards and a ready calibration set of polyethylene glycol/polyethylene oxide were obtained from Sigma Aldrich. Tablet forms containing metformin hydrochloride 500 mg and pioglitazone hydrochloride equivalent to pioglitazone 15 mg are produced in the R&D laboratory of PT. Kalbe Farma, Tbk, and the reference products are commercially available. An Agilent HPLC 1260 Infinity II LC system, consisting of the following components: 1260 Infinity Quaternary Pump, 1260 Infinity Degasser, and refractive index detector, was used. The separation was performed on an Ultrahydrogel 120 column (300 × 7.8 mm, particle size 6 μm, pore size 120 Å) under size exclusion chromatographic conditions. The mobile phase is an 80:20% v/v mixture of 0.1M sodium nitrate in purified water: acetonitrile. The mobile phase was filtered through a 0.45 μm PVDF membrane filter and degassed using an ultrasonic bath for about 15 minutes. Sample solutions were also filtered using a 0.45 μm PVDF membrane filter. The working conditions are as follows: flow rate of 0.6 mL/min, column temperature of 30 °C, injection volume of 20 μL, and total retention time of 30 minutes.

A stock solution of PVP 30 was prepared at a 25 mg/mL concentration in purified water. The stock solution was diluted to produce working standard solutions with a 2.5 mg/mL concentration. Calibration standards were prepared by diluting the stock solution with purified water to obtain a concentration range from 0.25 mg/mL to 3.75 mg/mL. Ready calibration Standard polyethylene glycol/polyethylene oxide was directly injected into the system without further dilution.

A bulk mixture without PVP was prepared per tablet formula using metformin HCl and pioglitazone HCl. Common excipients used in tablet formulations were added to this mixture, including glycerine, croscarmellose sodium, microcrystalline cellulose, magnesium stearate, and purified water.

Pharmaceutical tablets were prepared using the wet granulation method, consisting of metformin HCl, pioglitazone HCl, and all excipients, compressed with an average weight of 585 mg per tablet.

Ten tablets, each containing metformin hydrochloride 500 mg and pioglitazone hydrochloride equivalent to pioglitazone 15 mg, were weighed and then powdered. The amount equivalent to the average weight of 1 tablet was then transferred to a 10 mL volumetric flask and dissolved with about 8 mL of purified water. The samples were allowed to dissolve in water for 3 hours and then made up with purified water. After that, the solution was filtered through a PVDF 0.45 μm membrane filter and analyzed.

The HPLC-SEC method for quantifying PVP 30 in pharmaceutical tablets was validated in terms of system suitability, specificity, linearity, limit of detection (LOD), and limit of quantification (LOQ), accuracy, precision, and robustness by the International Conference on Harmonization (ICH) guidelines.

The system suitability test was evaluated by injecting a working standard solution into the chromatographic system for six replicate injections. System fit characteristics such as tailing factor, retention time, resolution, theoretical plates, and analyte peak responses were assessed.

The objective of the specificity testing was to demonstrate that the main peaks of the analyte could be differentiated from those of the sample matrix and any associated peaks. A placebo solution was also injected into the chromatographic system to ensure no interference from the sample matrix. The spiked substance (PVP 30) was used in the placebo samples to evaluate the specificity of detecting and differentiating PVP 30 from other substances.

Linearity was assessed by examining calibration standards (ranging from 0.25 mg/mL to 3.75 mg/mL) at 7 different concentrations in 3 replicates. Calibration curves were generated by linear regression, plotting the PVP 30 concentrations (x) as the abscissa and the peak response (y) as the ordinate.

The detection or quantification limits are the minimum concentrations at which the analyte can be reliably detected or quantitated. A signal-to-noise ratio of 3:1 is generally considered acceptable for estimating the detection limit. A ratio of at least 10:1 is considered acceptable for the quantification limit. The detection limit was defined using the equation LOD = 3.3 σ/S, and the limit of quantitation with the equation LOQ = 10 σ/S, where σ is the standard deviation of the responses, and S is the slope of the calibration curve.

The method’s accuracy was measured at three concentrations of the sample solutions (70%, 100%, and 130%) using nine determinations (three replicates of each concentration). Three different levels of standard solutions (1.75, 2.50, and 3.25 mg/mL) were spiked into the placebo. Percent recoveries were calculated by comparing the measured amount of those standards with the amount added.

The method’s precision was assessed in terms of repeatability and intermediate precision. Repeatability was evaluated through six determinations at 100% of the test concentration, while intermediate precision was investigated using authentic homogeneous samples performed on three different days and by three different analysts. Precisions were determined by calculating the relative standard deviation percentage (% RSD) on the same day and on three different days, respectively.

The method’s robustness was assessed by intentionally modifying the chromatographic conditions, such as the percentage of the organic phase, flow rate, and column temperature, by slightly increasing or decreasing them. Each parameter was changed, one at a time. System suitability and sample runs were conducted using unchanged and modified method parameters. Furthermore, standard and sample solutions were stored at room temperature and analyzed at initial, 24, 48, and 72-hour intervals to assess sample stability. The solutions were tested against a freshly prepared standard at each time point.

The primary objective of chromatographic optimization is to achieve elution with an acceptable retention time, peak symmetry, high selectivity, and accuracy of the PVP peak in tablet formulations. The parameters assessed and optimized in this study include the column type, detector type, percentage of organic modifier, aqueous phase concentration, mobile phase additives, and flow rate. Several system conditions were examined for the optimization of chromatographic conditions, including reverse-phase and size-exclusion chromatography systems with ultraviolet (UV) detectors and refractive index (RI) detectors.

The studies employed a reverse-phase system with a C18 column and a mobile phase consisting of propanol and 0.01% tetrahydrofuran in ratios of 80:20, 50:50, and 20:80. The UV detection wavelength was set to 243 nm. Ultrahydogel 120 and 1000 columns were used in SEC studies, with mobile phases of 0.1M sodium nitrate and acetonitrile in ratios of 80:20, 85:15, 90:10, and 95:5. An RI detector was employed in this system. Other salts, such as sodium chloride and ammonium acetate, were also examined. The flow rate and column temperature were studied within the limits specified in the column’s instruction manual.

In this study, several types of PVP were injected into the chromatographic system, and the separation of the target analytes was observed. PVP-10, PVP-25, PVP-30, PVP-40, and PVP-90 were chosen considering that these types are commonly used in tablet formulations, as referenced in the FDA Inactive Ingredient Search for Approved Drug Products. The optimal conditions were obtained with the SEC system using the Ultrahydrogel 120 column, an RI detector, and a mobile phase of sodium nitrate: acetonitrile (80:20) with a 0.6 mL/minute flow rate. The injection volume and column temperature were set at 20 μL and 30 °C, respectively.

Calibrations were performed using narrow polyethylene glycol/oxide (PEG/PEO) standards to identify the type of PVP that might be used in the reference tablet. PEG/PEO-ready calibration standards, obtained from Sigma Aldrich and Polymer Laboratories with molecular weights ranging from 238 to 1,180,000 Da, were used to create calibration curves. These standards were directly injected into the system. The oxides cover the higher molecular weight range, while the glycols cover the lower molecular weight range. Calibration curves plotting retention volume against the logarithm of molecular weight were created. Subsequently, the retention volume of PVP detected in the tablet solutions was plotted on these curves to estimate the molecular weight of the PVP. In this study, an SEC column of Ultrahydrogel 1000 was coupled with Ultrahydrogel 120 to facilitate separating a wide range of molecular weights in the PEG/PEO standard calibration. Coupling columns of different pore sizes gives SEC a tremendous dynamic range, separating within a single experiment across orders of magnitude in molar mass (Striegel 2022).

PEG/PEO calibration curves and chromatograms were presented in Figs 2, 3 and Tables 1, 2. The molecules’ size and shape mainly govern the separation process in size-exclusion chromatography. As the sample is injected into the column, it undergoes a size-based separation, with larger molecules eluting first and smaller molecules eluting later (Soni 2023).

Figure 2. 

PEG/PEO calibration curves.

Figure 3. 

Chromatogram of PEG/PEO standards. a. Standard Black (PEO-1180000, PEO-99000, and PEG-6530); b. Standard Blue (PEO-504000, PEO-42700, and PEG-599); c. Standard Yellow (PEO-217000, PEO-18600, PEG-2130, and PEG-238).

Based on the estimated molecular weights obtained from calibration plotting to the PEG/PEO narrow calibration standards, the PVP used in PioMix and ActMet Met Tablets has molecular weights of approximately 13,678 and 17,365, respectively. These molecular weights are close to those of PVP-30 and PVP-40. Based on this finding, further quantitative determination of PVP in the tablet formulations was conducted using PVP-30 as the reference standard, considering that PVP-30 is the most common PVP used in tablet formulations.

The system suitability tests represent an integral part of the analytical procedure and are used to ensure adequate performance of the chromatographic system (United States Pharmacopeia 2023). Under optimized chromatographic conditions, system suitability parameters, such as column efficiency expressed by the number of theoretical plates (N), tailing factor (T), resolution (R), and repeatability (RSD of peak response and retention time), were evaluated and listed in Table 3. The calculated parameters were satisfactory and suggested the method’s high efficiency for determining PVP 30.

From the experiments performed, a complete separation of PVP 30 from the diluent and sample matrix was obtained, as shown in Fig. 4. The PVP peak has a clear baseline separation and is well separated from other components in the placebo. The retention time is sufficient for the analyte separation, and there is no overlap between peaks obtained from resolution data, indicating a reliable system.

Figure 4. 

Chromatogram specificity of PVP 30. a. Diluent; b. PVP 30 standard solution; c. Placebo solution; d. Placebo spiked with PVP 30 solution, and e. Formulation sample solution.

Linearity was determined for a standard of PVP 30 over a concentration range of 0.25 mg/mL to 3.75 mg/mL. The calibration curve was created using seven points covering seven different test compound concentrations within the specified concentration range. Linear regression was employed to process the calibration data. Then, the limit of detection and quantification was defined using the equation: LOD = 3.3 σ/S and LOQ = 10 σ/S, where σ is the standard deviation of the responses, and S is the slope of the calibration curve.

Figure 5. 

Calibration plot of PVP 30 standard solution.

The linearity parameters, such as the correlation coefficient, intercept, slope, and standard deviation of the residuals for the calibration data and concentration range, are presented in Table 4.

Recovery studies were conducted to verify the accuracy of the developed method. To pre-analyze the tablet solution, a specific concentration of PVP standard (70%, 100%, and 130%) was added, and then its recovery was determined. The statistical validation of the recovery studies is presented in Table 5.

The results of accuracy studies met the acceptance criteria, with the average recovery from each concentration level falling within the range of 98.0%–102.0% and a total average recovery of 100.4%.

The method’s repeatability was evaluated through six determinations at 100% of the test concentration. Intermediate precision was investigated using authentic homogeneous samples, performed on three days by three analysts. The percentage of relative standard deviation (% RSD) was calculated to determine the method’s precision. The results are presented in Tables 6, 7.

The precision testing results were within the acceptance criteria of % RSD, which was not more than 2.0%. Moreover, the results of the accuracy and precision studies indicated that the method operates accurately and precisely within acceptable parameters.

The robustness of an analytical procedure is a measure of its capacity to meet the expected performance criteria during normal use. Robustness is tested by deliberate variations of analytical procedure parameters (Committee for Medicinal Products for Human Use 2022). The robustness test used varying column temperature, flow rate, and mobile phase composition. The column temperature was varied at 20 °C, 30 °C, and 40 °C. The flow rate of the mobile phase was varied at 0.5 mL/min, 0.6 mL/min, and 0.7 mL/min.

Meanwhile, the mobile phase composition of 0.1M sodium nitrate and acetonitrile was varied at compositions of 85:15, 80:20, and 90:10. Evaluation was carried out on system suitability parameters and % recovery for each variation. The variation in column temperature, flow rate, and mobile phase composition had little effect on chromatogram quality parameters such as retention time, resolution, and theoretical number of plates, as mentioned in Table 6. % RSD area and % recovery were within limits. As a result, the proposed method can be used for quantitative analysis of PVP 30 in tablet formulations containing metformin HCl and pioglitazone HCl.

The stability of standard and sample solutions was assessed to ensure their stability during analysis. Testing was conducted at room temperature (below 30 °C) over 72 hours, with observations made at 0, 24, 48, and 72 hours. The standard PVP-30, 100% accuracy, and PioMix tablet sample solutions were prepared and analyzed at specified intervals. Results showed that all solutions remained stable throughout the testing period, with deviations in peak response and PVP-30 content less than 2.0%. This indicates that standard and sample solutions can be stored at room temperature for 72 hours without significant degradation. The results are presented in Table 9.

Based on the comparison of the suitability parameters with the ICH Q2(R2) guideline range for these parameters, it is shown that the method is suitable for the analysis of PVP in tablet formulations.

The validated method was subsequently applied to determine PVP content in two different pharmaceutical formulations: PioMix tablets produced in the R&D laboratory of PT. Kalbe Farma and ActMet tablets are commercially available. PVP 30 was detected at its designated retention time. The findings were presented as a percentage of the tablet weight, as detailed in Table 11.

Povidone, or PVP, is a binder in wet granulation and is generally used as a solution. Although use levels in the literature are reported as 2% to 5%, higher levels of up to 10% may have to be used in challenging, poorly compactable formulation (Handbook of Pharmaceutical Granulation Technology 2021). The obtained PVP 30 content was 3.91% and 4.12% by weight for PioMix and ActMet tablets, respectively.

The novelty of the optimized methodology lies in the absence of a specific established method for determining the type and quantity of PVP in tablets containing metformin HCl and pioglitazone HCl. The advantages of the optimized method include its ability to quantify a wide range of PVP concentrations with simple preparation and reliable accuracy and precision. However, a disadvantage of the method is its lengthy runtime per injection, approximately 30 minutes per injection.