1.Introduction

Type 2 diabetes mellitus (T2DM) is a collection of metabolic disorders characterized by persistently high blood glucose levels due to problems with insulin secretion and/or function [1]. It has emerged as the third most serious chronic non-communicable disease threatening global health, following cardiovascular diseases and cancer [2]. Diabetes inflicts chronic damage on various organs, including the eyes, heart, and kidneys, severely affecting patients' lives and quality of life, and placing a significant financial burden on families and individuals [3, 4].

Oral antidiabetic drugs are easy to take, which encourages strong adherence among T2DM patients. Effective management of T2DM involves a comprehensive approach that includes lifestyle modifications, medication, and regular blood glucose monitoring [5, 6]. Metformin, a biguanide, has been utilized as an oral hypoglycemic agent and is considered the first-line treatment for T2DM for over 50 years. However, as the disease progresses, due to insufficient dosing, poor adherence, or other conditions that affect glucose metabolism, metformin alone often fails to maintain satisfactory blood glucose levels [7]. In such situations, clinicians may enhance treatment by increasing the metformin dose or adding another T2DM medication to the regimen to achieve better glycemic control.

Metformin, a biguanide agent, acts primarily as an insulin sensitizer. Its primary clinical site of action is in the liver, improving hepatic insulin sensitivity and as a result, decreasing hepatic gluconeogenesis. Metformin may also increase both hepatic and splanchnic glucose utilization. Metformin also has a significant effect on peripheral insulin sensitivity, primarily in muscle and modestly at adipocyte by phosphorylation and activation of AMP-activated protein kinase [8]. Therefore, reductions in blood sugar with metformin occur in the setting of reduced insulin levels and theoretically reduced beta-cell work. There appears to be little or no direct mechanistic effect of the drug, however, on beta-cell function or apoptosis. Metformin is negligibly protein bound and is primarily renally cleared, proportionate to declines in creatinine clearance. Therefore, dosage reduction or therapeutic restriction may be required in the aging or renally impaired patient [9].

Troglitazone, the first thiazolidinedione approved for clinical use, was withdrawn from the market three years after its approval due to a high incidence of liver injury, including cases of acute liver failure. The thiazolidinediones (TZDs) currently available on the market, rosiglitazone and pioglitazone are potent agonists of the peroxisome proliferator-activated receptor (PPAR)-gamma. This receptor is a ligand-activated transcription factor crucial for glucose and lipid metabolism. Additionally, TZDs demonstrate modest activation of PPAR-alpha, another transcription factor involved in lipid metabolism and fat oxidation. PPAR-gamma receptors are highly expressed in the vasculature, adipocytes, and macrophages, indicating that thiazolidinediones may also significantly impact non-metabolic processes such as inflammation and atherogenesis.

The fixed-dose combination (FDC) of thiazolidinediones and metformin brings together two insulin sensitizers to improve insulin sensitivity [10]. This combination has been proposed as an alternative to increasing the dose of metformin for patients with T2DM who are not achieving adequate glycemic control [11]. The rationale behind combining these agents lies in their complementary mechanisms of action. Together, thiazolidinediones and metformin improve glycemic control by reducing insulin resistance and enhancing insulin sensitivity without causing hypoglycemia.

Additionally, it has been observed that increasing the number of medications can make it more challenging for patients to adhere to their treatment regimens. In contrast, using a single tablet containing fixed doses of both drugs, which have distinct mechanisms, can reduce the number of medications needed. This approach ensures better glycemic control and improves overall patient compliance. This paper aims to review and analyze the literature on the analytical methods for the determination of metformin combinations with thiazolidinediones in pharmaceutical dosage forms. To our knowledge, a review on this topic is not yet published.

2. Materials and methods

We conducted comprehensive literature searches in PubMed, Google Scholar, Web of Science; using the term “determination of metformin and glitazones combination or determination of metformin and thiazolidinediones combination”. The search covered the period from 2005 to the present.

3. Results and discussion

3. Analysis techniques

3.1 Official methods

For the determination of MET and PIO combination in tablets, the United States Pharmacopeia (USP) [12] employs high-performance liquid chromatography with an octacylsilane column (6.0 mm × 15 cm; 5 µm) and a mobile phase made of: 7.2 g/L of sodium dodecyl sulfate in a mixture of 0.05 M monobasic ammonium phosphate and acetonitrile (1:1). Metformin is detected at 255 nm, whereas pioglitazone is detected at 225 nm. The mobile phase is pumped at a flow rate of 1 mL/minute.

3.2 Separation methods

3.2.1 High-performance liquid chromatography (HPLC)

Reversed-phase high-performance liquid chromatography (RP-HPLC) is extensively used for quantifying MET and thiazolidinedione combinations in pharmaceutical dosage forms. Most established methods employ isocratic elution with reversed-phase columns (C₈ or C₁₈) and mobile phases composed of mixtures of organic solvents and buffers, with carefully adjusted pH levels.

Analyzing combination drug products containing two or more active pharmaceutical ingredients (APIs) presents several challenges for RP-HPLC. These challenges include difficulties with analyte retention, separating impurities, distinguishing excipients, maintaining peak shape, and ensuring column longevity [13].

The polarity and ionization potential of molecules play a crucial role in their retention on organic reverse-phase high-performance liquid chromatography (RP-HPLC) columns [14]. When active pharmaceutical ingredients (APIs) in a combination drug exhibit substantial differences in polarity, it often necessitates the development of multiple chromatographic methods. This method development and validation process can be both resource-intensive and time-consuming. A typical example of this challenge is observed in drug combinations that include metformin and thiazolidinediones. Metformin, being highly polar, contrasts significantly with thiazolidinediones, which have medium-to-low polarity, as reflected by their respective log P values in Table 1.

Table 1. Structure and Log P (octanol/water) of metformin, pioglitazone and rosiglitazone

Under reversed-phase conditions, due to the significant difference in polarity between the combined compounds, metformin is expected to elute very early, near or with the solvent front. In contrast, the retention of thiazolidinediones (PIO and ROS) varies depending on the pH of the mobile phase and the amount of the organic modifier used. To evaluate the effect of chromatographic conditions on the retention of metformin, we calculated the retention factor (k) according to the guidelines set by the United States Pharmacopeia (USP) [12]. Only methods [22, 23] satisfied the USP requirements (2 < k < 10), while in methods [18, 25-27, 29], metformin eluted slightly away from the solvent front. In the remaining methods [19, 20, 24, 28], metformin eluted very close to the dead time, resulting in a k value approaching zero.

The challenge of metformin's low affinity for reversed-phase packing material is further complicated by the presence of thiazolidinediones. At a retention time where metformin achieves adequate retention (i.e., k > 2), the thiazolidinediones elute at unacceptably long times. To address this issue, ion-pairing reagents such as sodium dodecyl sulfate and hexane sulfonic acid [23, 25] have been employed to enhance metformin's retention. El-Zaher et al. [28] successfully tackled the retention challenges of metformin and pioglitazone by using a cyanopropyl column, an approach similar to the one followed by Reid [30-32] and Gedawy et al. [33, 34] for the determination of combinations containing compounds of different polarities. This approach allowed for effective separation of metformin from thiazolidinediones and moved the metformin peak away from the solvent front. A description of the reported HPLC methods is given in Table 2.

Table 2. High performance liquid chromatographic methods used for the analysis of metformin and thiazolidinediones combinations.

3.2.2 Thin-layer high performance liquid chromatography (HPTLC)

The HPTLC separation of metformin (MET) and pioglitazone (PIO) is typically performed on silica gel 60F254 precoated plates using various chromatographic conditions. Dharmamoorthy et al. [35] employed a mobile phase of toluene, methanol, and triethylamine (6:4:0.1 v/v/v). The method had a working range of 50-300 µg/mL for MET and 1.5-9.0 µg/mL for PIO, with limits of detection (LOD) of 10 ng/spot and 0.5 ng/spot, respectively. Analytes were detected at 230 nm.

Khorshid et al. [36] used a mobile phase consisting of toluene, methanol, and acetic acid (5:5:0.5 v/v/v) to separate MET and PIO in the presence of the impurity melamine. The method's working range was 3-12 µg/band for MET, 3-20 µg/band for PIO, and 0.5-5 µg/band for melamine. Detection was performed at 240 nm.

Modi et al. used a mobile phase of butanol, 1,4-dioxane, and glacial acetic acid (5:3:2 v/v/v). The method exhibited a linear range of 2000-20000 ng/band for MET and 60-600 ng/band for PIO, with detection limits of 629.89 ng/band and 8.51 ng/band, respectively. Analytes were detected at 226 nm [37].

The selection of detection wavelengths was clearly justified and supported by the overlain spectra of the analytes in references [36] and [37]. However, in reference [35], the chosen detection wavelength is not consistent with the overlain spectra, which suggest 250 nm as the more suitable wavelength for detection.

The methods were thoroughly validated for specificity, linearity, limit of detection and quantitation, accuracy, precision, and robustness. They were successfully applied to the simultaneous determination of MET and PIO in tablets.

3.2.3 Capillary zone electrophoresis (CZE)

Metformin and pioglitazone were analyzed using capillary zone electrophoresis (CZE) with a 75 mmol/L phosphate buffer containing 30% acetonitrile (ACN) at pH 4.0. The conditions included a 10-second hydrodynamic injection time at 0.5 psi, a separation voltage of 25 kV, and a column temperature of 25°C. Detection was carried out at 210 nm. The method showed a linear range of 10-80 µg/mL for MET and 10-100 µg/mL for PIO, with limits of detection (LOD) of 0.091 µg/mL and 0.277 µg/mL, respectively [38]. The method's suitability was confirmed through validation for specificity, linearity, stability, accuracy, limit of detection and quantitation, precision, and robustness, as well as by comparing it for the analysis of the two drugs combination with a previously reported high-performance liquid chromatography method.

Yardımcı et al. [39] reported the determination of metformin (MET) and rosiglitazone (ROS) using capillary electrophoresis with a running buffer of 25 mM acetate at pH 4.0 and a fused-silica capillary column (80.5 cm × 75 µm i.d., effective length 72.0 cm). The applied voltage was +25.0 kV, and detection was performed at 203 nm. The method demonstrated a linear range of 1.0–12.0 µg/mL for ROS and 100.0–1200.0 µg/mL for MET, with a limit of detection (LOD) of approximately 0.5 µg/mL for both analytes. During optimization, various electrolyte parameters such as pH, buffer concentration and type, and the type and concentration of the organic modifier were studied. The effects of applied voltage, separation temperature, and injection time were also evaluated. The optimized method was then validated for specificity, linearity, stability, accuracy, limit of detection and quantitation, precision, and robustness. Their method was then successfully applied to the simultaneous determination of MET and ROS.

3.3 Spectrophotometric methods

Analytical chemists face considerable challenges when analyzing and controlling multicomponent formulations through direct spectrophotometry, primarily due to overlapping spectral bands. To address this, various spectrophotometric methods have been developed to resolve mixtures of compounds with overlapping spectra. The success of these techniques largely depends on the degree of spectral overlap and the number of components present in the mixture. When the overlap between analytes is minimal, methods involving simple mathematical manipulation of spectral data are employed [40]. However, for mixtures with extensive overlap, complete spectral analysis is used to determine all components [41]. The accuracy of these determinations can be enhanced by carefully selecting wavelength ranges, which helps reduce collinearity or spectral overlap [42]. Table 3 gives the details of the spectrophotometric methods used.