1 Introduction

During the COVID-19 pandemic, pharmaceutical compounds (PhCs) were widely prescribed to provide immediate therapeutic responses. This was particularly crucial in the absence of vaccine alternatives during the last century's most significant global health crisis. Beyond the strain imposed on hospitals and health centers, the COVID-19 pandemic significantly reconfigured global healthcare systems (Shaker Ardakani et al., 2023). To treat COVID-19 necessitated the utilization of various PhCs, including antivirals, antibiotics, anti-inflammatories, and anticoagulants, to mitigate bacterial co-infections associated with COVID-19 (Li et al., 2023). The escalated use of these PhCs was driven by urgent clinical imperatives, public demand, and, in some instances, inaccurate information regarding their efficacy was disseminated.

The efficacy of these treatments exhibited variable outcomes throughout the pandemic. For instance, dexamethasone was utilized to reduce mortality in critically ill patients by attenuating cytokine storms (Martinez-Guerra et al., 2022; Noreen et al., 2021), resulting in a 170% increase in its sales in Mexico during 2020 (Durán-Álvarez et al., 2023). Concurrently, the demand for ivermectin in Mexico, often employed in conjunction with azithromycin and famotidine, surged by 652% due to public perception regarding its potential in COVID-19 prevention and treatment (Durán-Álvarez et al., 2023). Other PhCs, including chloroquine, hydroxychloroquine, and albendazole, were administered alongside antibiotics as part of a multi-faceted therapeutic approach to address parasitic and bacterial co-infections. Additionally, the consumption of over-the-counter analgesics such as paracetamol, diclofenac, and ibuprofen also increased significantly (Luis López-Miranda et al., 2022), with a reported rise of up to 210% in defined daily doses (DDD) in Poland (Kołecka et al., 2022).

The widespread prescription of PhCs has elicited significant environmental concerns. Following partial metabolization, PhCs enter wastewater treatment plants (WWTPs) through municipal sewage networks, where they are not entirely removed. Consequently, these compounds are released into the environment via effluent discharge or sludge disposal. Due to their structural complexity, PhCs exhibit stability and high mobility (Thilagam & Gopalakrishnan, 2022), persisting at concentrations ranging from nanograms per liter (ng L⁻1) to micrograms per liter (μg L⁻1) in surface water, groundwater, and wastewater (Dueñas-Moreno et al., 2024; Ronderos-Lara et al., 2021; Vázquez-Tapia et al., 2022). The escalating presence of these compounds in aquatic ecosystems poses an ecological risk, as the mechanisms of action and toxic effects of many PhCs on marine organisms remain incompletely characterized. Nevertheless, evidence suggests that certain PhCs can disrupt endocrine, reproductive, and developmental functions in these organisms, raising substantial concerns for both environmental conservation and public health (Drzymała & Kalka, 2020; Rehman et al., 2024; Riva et al., 2019).

While the presence of PhCs and personal care products (PCPs) in aquatic environments raises increasing concern, the risks associated with emerging contaminants, particularly PhCs used in COVID-19 treatment, remain inadequately characterized. The Nexapa River, situated in Izúcar de Matamoros, Puebla, receives municipal effluents that may contribute to PhC contamination, especially given the increasing drug consumption observed during the pandemic. Persistent pharmaceutical pollutants can potentially disrupt various trophic levels within aquatic ecosystems, inducing alterations in aquatic organisms'development, reproduction, and metabolism. Moreover, the potential for synergistic effects among multiple contaminants in water bodies is poorly understood, posing significant challenges for environmental risk assessment.

However, the limited information available regarding PhCs in the Nexapa River does not allow for the establishment of baseline contamination levels or the assessment of ecotoxicological risks associated with their use in the region before the pandemic period. Addressing this knowledge gap is critical. PhCs in the Nexapa River may affect water quality and public health, particularly for communities reliant on the river for agricultural or domestic activities. The accumulation of these compounds, potentially exacerbated by increased consumption of PhCs during the pandemic, highlights the urgent need for research to assess their potential environmental impact and provide strategies for managing emerging contaminants in water bodies receiving municipal discharges.

Therefore, this study developed and applied a solid-phase extraction (SPE) method to detect and quantify several PhCs in wastewater treatment plant (WWTP) effluents and the Nexapa River. Furthermore, a Quantitative Structure–Activity Relationships (QSAR) model was utilized to assess the potential environmental risks of these compounds on aquatic organisms across three trophic levels. This approach provides crucial information on the magnitude of pharmaceutical contamination and its possible ecological consequences. It will serve as a basis for future research and environmental management policies and for developing stricter regulations on wastewater treatment and proper disposal of pharmaceuticals.

2 Methods and Materials

2.1 Chemicals and Standards

The PhCs standards with high purity (> 98.5%) and were purchased from Sigma-Aldrich (St. Louis, MO, United States). Acetonitrile (99.9%) and acetic acid (99.8%) were obtained from Fermont (Monterrey, NL, Mexico), and ethanol (99.5%) was obtained from Karal (León, GTO, Mexico). The physicochemical properties of the target PhCs are shown in Table S1.

3 Study Area and Sample Collection

The study was conducted at the WWTP in Izucar de Matamoros, Puebla, Mexico (18°34′1"N, 98°28′27"W, 1245 m.a.s.l.). The area is characterized by a warm sub-humid climate with summer precipitation, a mean annual temperature of 22.8 °C (min. 15 °C; max. 36 °C), and an average rainfall of 800–1000 mm from May to October. Data were taken from the meteorological station 21,132, Izucar de Matamoros, Puebla (Coordinación general del servicio meteorológico nacional, 2024).

The WWTP was designed to comply with national standard NOM-001-SEMARNAT-1996 for pollutants in wastewater discharges in national waters. WWTP is configured to treat municipal wastewater and consists of two identical treatment trains operated in parallel with a total capacity of 5400 L min−1. The treatment process involves pretreatment to remove sand and grease, followed by a primary sedimentation tank with a hydraulic retention time of 120 min. The solids-free water is then directed to a biological filter to reduce biological oxygen demand (BOD), nitrogen, and phosphorus levels. Excess sludge is removed by a secondary clarifier and conveyed to a drying bed for disposal. The wastewater treated is disinfected by adding sodium hypochlorite to eliminate pathogens before being discharged into the Nexapa River for agricultural irrigation. Three strategic sampling points were established in the studied area (Figure S1): P1, In the Nexapa River upstream of the WWTP discharge point; P2, In the WWTP outlet stream; and P3, 300 m downstream of the WWTP discharge point. It is important to note that WWTP directly discharged municipal wastewater into the Nexapa River during the entire sampling campaign, meaning the wastewater did not undergo treatment at WWTP.

Sampling was conducted during four monitoring campaigns throughout January, May, August, and November 2022, covering dry and rainy seasons and transition periods. A total of twelve individual water samples were collected for analysis. At each sampling point, one liter of water was collected in pre-rinsed amber glass bottles for physicochemical analysis and solid phase extraction (SPE). All samples were stored at 4 °C until SPE and physicochemical analysis, which was carried out within 24 h.

Detailed physicochemical characteristics of the wastewater from this study are presented in Table S2.

4 Selection of Target PhCs

COVID-19 incidence data in Izucar de Matamoros during the sampling period was collected to contextualize the potential influence of the pandemic on PhCs contamination levels (Secretaría de Salud, 2023).

The selection of target PhCs was based on the following: (i) a literature review of pharmaceuticals used in the treatment of COVID-19 worldwide, (ii) a literature review covering the PhCs detected in wastewater worldwide, and (iii) the medical treatment protocols implemented during the health emergency in Mexico. The PhCs chosen for this study included acetaminophen (ACP), albendazole (ALB), chloroquine (CQ), dexamethasone (DEX), diclofenac (DF), hydroxychloroquine (HCQ), and ivermectin (IVM).

5 Sample Preparation and Extraction

Wastewater and spiked water samples were filtered through a cellulose filter to remove larger solid particles using a 0.45 μm PTEF MERC membrane filter. Concentrated target PhCs were extracted using hydrophilic-lipophilic-lipophilic-equilibrated (HLB) reverse-phase sorbent cartridges (Supel™ Swift HLB SPE cartridges 200 mg sorbent, 6 mL from Supelco, USA) connected to an SPE tube vacuum manifold. The HLB sorbents are microporous copolymers composed of divinylbenzene and vinylpyrrolidone. The cartridges were preconditioned with 3 mL of methanol and 3 mL of water before sample application. The pH of the sample was adjusted to 7 using hydrochloric acid or sodium hydroxide solution (0.1 M). A sample volume of 500 mL was loaded into the cartridge at a flow rate of 1 mL min−1 followed by a drying period of 10 min under vacuum to remove excess water. The retained analytes were eluted with two 5 mL aliquots of acetonitrile containing 0.1% (v/v) acetic acid. Subsequently, the filtrates were evaporated to dryness using a stream of nitrogen gas and redissolved in 500 µL of 1:1 (v/v) acetonitrile–water, reaching a 1000-fold concentration. The recovery of the SPE procedure was evaluated by adding 10 µg L−1 of the target analytes to the pre-filtered, pH-adjusted distilled water and calculated using Eq. 1.

$$Recovery \left(\%\right)=\frac{{C}_{1}}{{C}_{0}}\times 100\%$$
(1)

where C1 (µg L−1) is the concentration measured in the sample and C0 (µg L−1.) is the spiked concentration.

6 LC-DAD-MS Analysis and Method Validation

Detection and quantification of the target PhCs were analyzed using amplifying chromatography on an LC/DAD/MS instrument (Chromatograph series 1260 from Agilent Technologies, Santa Clara, Ca, United States). Separation was achieved with a Phenomenex Luna C18 column (4.6 × 250 mm, 5 µm), using mobile phases of water (A) and acetonitrile (B), both containing 0.1% (v/v) formic acid. The elution gradient started with 10% A for 2 min, changed linearly to 90% A from 2 to 22 min, and changed back to 10% A from 22 to 32 min, followed by a 10-min reconditioning period, resulting in a total time of 42 min. Detection was performed with a DAD at 244 nm and an ESI-QTOF-MS 6520 detector (Agilent) using positive and negative ionization modes with the following source parameters: fragmentor voltage 175 V, capillary voltage 3500 V, gas temperature 350 °C, N2 flow rate 11 L min−1, and nebulizer pressure of 60 psi. MS/MS analysis was done with a collision energy of 30 V. The linearity of the method was evaluated by constructing calibration curves for each PhC within a range of 0.01 to 10 mg L−1 in wastewater. OriginPRO 2023 software (V.10.0.0.154, Learning Edition) was used to fit the data to a linear model. The limits of detection (LOD) and limits of quantification (LOQ) were calculated using Eqs. 2 and 3.

$$LOD=\frac{3{S}_{0}}{m}$$
(2)
$$LOQ=\frac{10{S}_{0}}{m}$$
(3)

where S0 is the blank standard deviation, and m denotes the slope of the calibration curve for each PhC. The standard deviation of the blank was determined from ten replicates. The identification of PhCs was done with a high-resolution instrument with a difference of mass/charge (Δm/z) less than 5 ppm between the experimental and theoretical mass of the ion. Retention time comparison with PhC standards confirmed compound identification.

7 Ecological Risk Assessment

The aquatic risk assessment was estimated by calculating risk quotients (RQ) using Eq. 4 according to EU guidelines (European Commission, 2003).

$$RQ=\frac{MEC}{PNEC}$$
(4)

where MEC is the measured environmental concentration of each PhC (µg L−1), and PNEC is the predicted no-effect concentration (µg L−1). The PNEC values were calculated by dividing the half-maximal effective concentration (EC50) or the lethal concentration 50 (LC50), as reported in the literature (Table 1), by a risk assessment factor of 1000 for acute toxicity tests. The EC50 and LC50 values were obtained from literature sources and the ECOTOX database, prioritizing the lowest reported values (EPA, 2024).

Table 1 EC50 or LC50 values of PhCs from experimental studies used to calculate PNEC in algae, invertebrates, and fish
Full size table

For unreported data, the ECOSAR software was used (EPA, 2024). The criteria for interpretation of the RQ ratio were described by Sánchez-Bayo et al. (2002): RQ < 0.01 indicated"no risk", 0.01 ≤ RQ < 0.1 indicated"low risk", 0.1 ≤ RQ < 1 indicated"medium risk", and RQ ≥ 1 indicated"high risk".

8 Statistical Analysis

ANOVA was performed to evaluate the significance of treatment conditions during the SPE procedure. Tukey's test was used to estimate 95% confidence intervals for recovery percentages. All experimental trials were performed in triplicate. Statistical analysis was performed with Minitab® 19, Version 19.1.

9 Results and Discussion

9.1 Determination of LOD and LOQ and Validation of the SPE Method

The calibration curves for PhCs in deionized water fit the experimental data, as indicated by correlation coefficients (R2) greater than 0.990 in all cases (Figure S2). Linear ranges were established from 10 to 10,000 µg L−1 for ACP, ALB, CQ, DF, DEX; and from 100 to 10,000 µg L−1 for HCQ and IVM. Table 2 provides LOD and LOQ for each PhCs. ALB had the lowest LOD and LOQ values, at 2.46 µg L−1 and 8.17 µg L−1, respectively. In contrast, IVM had the highest LOD and LOQ values, reaching 58.02 µg L−1 and 193.42 µg L−1, respectively. The calculated detection limits for the remaining PhCs were similar to those previously reported for LC-DAD-MS studies, such as those conducted by Kumar Mehata et al. (Kumar Mehata et al., 2022).

Table 2 Calibration parameters, LOD, and LOQ calculated for target PhCs in wastewater. The linear range was from 0.01 to 10 mg L−1 of all PhCs
Full size table

In addition, distilled and tap water, containing 10 µg L−1 of each PhC, were used to determine the percent of recovery and confidence interval (CI) of the SPE method (Table 3). The two-factor ANOVA analysis (Table S3) indicated significant differences (α < 0.05) between the types of ambient water matrix used during the SPE extraction process. This finding highlights the importance of carefully considering the type of water used as a model when analyzing the presence of PhCs in aqueous matrices.

Table 3 Percent recovery and confidence intervals for the SPE method of target PhCs using distilled water, tap water, and wastewater
Full size table

9.2 Detection in Environmental Samples

At Point P1 (Nexapa River before the WWTP discharge), the presence of all compounds was detected, although in all cases, they were below the LOQ (Table 4). In general, the presence of all PhCs, except CQ (m/z = 320.183), at P2 and P3 was due to the discharge from WWTP. At these points, a signal attributable to HCQ (m/z = 336.1787) was observed below the limit of quantitation (LOQ). Lower but quantifiable concentrations were found at Point P3 (after WWTP) for ACP, ALB, DEX, and DF. These results indicate a contribution of the WWTP to the increase in PhCs concentrations at this discharge point in the Nexapa River, suggesting a direct release to the environment because of ineffective treatment by the WWTP. As already known, WWTPs are not designed to remove complex chemical compounds; in addition, in some cases, intermittent operation is associated with revamping and adapting the existing infrastructure.

Table 4 The concentration of PhCs at different sampling points surface (P1 and P3) and wastewater (P2)
Full size table

Comparable concentrations of IVM (5 to 20 µg L−1), DEX (3 µg L−1), and ALB (0.7 µg L−1) have been reported in WWTP effluents based on investigations conducted by (Chang et al., 2007; Charuaud et al., 2019; Mhuka et al., 2020). These results indicate consistency in the presence of these pharmaceutical compounds in the effluents of various WWTPs.

On the other hand, the highest concentrations were observed for DF and ACP, a common occurrence for these compounds in effluent studies. Raysyan et al. (Raysyan et al., 2021) reported DF concentrations of 3700 µg L−1, while (Mhuka et al., 2020) reported ACP concentrations of 6209 µg L−1. These results suggest that, compared to other pharmaceutical compounds, the concentrations of diclofenac and acetaminophen in WWTP effluents are relatively higher, which may indicate a relevant environmental impact and health risks associated with their presence in treated wastewater.

The detection of PhCs used in the symptomatic treatment of COVID-19 suggests that the presence of drugs in the wastewater of Izucar de Matamoros was influenced by the intensive use of these pharmaceuticals, coinciding with the period of incidence of active cases during the fourth (1123 accumulated cases) and fifth (663 accumulated cases) wave of COVID-19 in the region (Figure S3) (Secretaría de Salud, 2023). The high concentrations of PhCs match with reports from other studies that document an increase in the presence of these contaminants in water bodies due to the rise in their consumption during health emergencies. However, since these drugs have broader therapeutic applications, their detection in the analyzed samples cannot be attributed exclusively to the pandemic.

An additional challenge in interpreting these results is the limited or absent prior information on the presence of these compounds in the region, which prevents the establishment of a baseline of concentrations and accurately assessing the degree of increase due to the pandemic. In this sense, it's essential to implement continuous monitoring programs to generate historical data and evaluate the evolution of PhC contamination in local water bodies.

Many other compounds were detected by applying the HPLC mass separation and detection method; however, as they were not initially considered, their calibration curve for quantification was not available. Table 5 and Figure S4 show the compounds'identification parameters list.

Table 5 Detected Pharmaceutical compounds (PhCs), pesticides, and industrial chemicals in wastewater and the Nexapa River using LC-ESI-QTOF-MS/MS
Full size table

The pollutant loads (g day−1) in the WWTP effluent were calculated based on the WWTP flow rate of 5400 L min−1 (Table S4). All target PhCs are released into the river in gram quantities, with ACP reaching discharges of up to one kilogram. Using the Estimation Programs Interface (EPI) Suite™ (EPA, 2024b), selected PhCs'conversion degree was assessed for an equivalent WWTP configuration under optimal conditions. As shown in Table S4, the maximum conversion achieved was 56% for DF. These findings indicate that the PhCs studied are highly recalcitrant, suggesting their removal remains minimal even with WWTP operating at peak efficiency. Since the concentrations of CQ and HCQ were below the quantification level, the estimation of mass loading and the following associated environmental risk will focus exclusively on the chosen compounds.

As already known, WWTPs are designed to comply with local regulations by removing organic loadings such as BOD, chemical oxygen demand (COD), total suspended solids (TSS), nutrients, and pathogens (Abejón et al., 2015; Grandclément et al., 2017), and often fail to address recalcitrant compounds such PhCs. These design limitations explain the minimal reduction of PhCs loading despite their adequate operation, emphasizing the need for technological advancements in WWTPs to improve the removal or reduction of these pollutants in municipal wastewater.

9.3 Environmental Risk Assessment

After quantifying PhCs, the potential environmental risk to aquatic organisms was assessed at points P2 and P3. Three trophic levels -algae, invertebrates, and fish- were considered to address the complexity of the ecosystem.

Figure 1 summarizes the RQ values of the target PhCs. Higher values were observed in P2 than in P3, with the following risk order for the founded pharmaceutical compounds (PhCs): IVM > DFC > ACP > DEX > ALB, consistent with (Yang et al., 2022). Point 2 had four PhCs with high risk for the three trophic levels; in contrast with point 3, two of the PhCs showed high-risk RQ values.

Fig. 1
figure 1

RQ values of PhCs at three different trophic levels (algae, invertebrates, and fish) at the points P2(WWTP effluent) and P3 (after WWTP)

Full size image

In the model organisms studied, fish were found to be sensitive to ACP, DEX, ALB, while invertebrates were more sensitive to IVM, and algae showed higher sensitivity to DF. These results contrast with the study by (Kumari & Kumar, 2022), who reported that algae were the most sensitive organisms to the presence of these compounds.

IVM exhibited the highest RQ values among all compounds, both in the WWTP effluent (RQ = 1,082,500) and the Nexapa River (RQ = 16,316). The highest risk was observed for invertebrates, followed by algae and fish, suggesting that IVM may pose a higher environmental priority than CFD and ACP. The complete data is presented in Table S5 and were congruent with (Durán-Álvarez et al., 2023; Félix-Cañedo et al., 2013; Marques et al., 2023)) In our study, DXM was identified as the fourth highest-risk substance, in wastewater and surface water (RQ = 415.24 and 15.43, respectively), consistent with by (Durán-Álvarez et al., 2023), who found an RQ = 106.9–474.77 in wastewater.

The results of this study are of great relevance due to their impact on both the environment and human health. Although the detected levels are significantly lower than the EC50 and LC50 acute toxicity concentrations reported for each compound (Table 1), the potential risk certain pharmaceuticals pose to aquatic organisms, especially fish and invertebrates, is still a concern. This environmental concern is exacerbated as the waters of the Nexapa River are also used to irrigate crops such as vegetables, corn, and sugar cane.

PhCs can bioaccumulate in plants, leading to the biomagnification of their effects when consumed by animals and humans. For instance, DF has been widely demonstrated to undergo bioconcentration, bioaccumulation, and potential biomagnification in aquatic ecosystems (Zenker et al., 2014). Furthermore, these compounds can alter nutrient cycling (Mesa et al., 2017), impacting soil health and agricultural productivity (Dai et al., 2023; Gu et al., 2021). These micropollutants can exhibit synergistic effects as mixtures, increasing ecological risks to aquatic organisms (Drzymała & Kalka, 2020; Nguyen et al., 2019). This phenomenon, known as mixture toxicity, occurs when different substances affect the same molecular mechanism in a typical target cell. Certain PhCs can act as enhancer substances, exacerbating the adverse effect of a driver substance, as observed in the combination of ACP and DF on cyclooxygenase. Therefore, specific measures need to be implemented to mitigate the potential impact of these pharmaceutical compounds on local aquatic ecosystems. This proactive approach involves the development of effective treatment systems for the mitigation of PhCs, along with their integration into existing local regulations. In addition, preventive and corrective maintenance and updated WWTP programs are essential. These collective efforts are crucial to ensure the long-term protection of aquatic ecosystems from the increasing presence of PhCs.

A limitation of this study is its reliance on QSAR-derived EC50/LC50 values for the ecological risk assessment. While these models provide a valuable preliminary evaluation without experimental data, they are inherently associated with uncertainty. Factors such as the applicability domain of the QSARs, the accuracy of the input parameters, and the potential for extrapolation beyond the model's training data can influence the reliability of the predictions. Consequently, the risk estimates presented here should be interpreted as preliminary and indicative. This last suggests the need for future studies incorporating experimental toxicity testing to validate these QSAR-based predictions and provide a more robust ecological risk assessment.

10 Conclusions

The evaluation of the environmental concentrations detected in the effluents of the Izucar de Matamoros WWTP allows for estimating the mass load of PhCs associated with the treatment of COVID-19 in the Nexapa River. The RQ values presented in this study should be interpreted cautiously and recalculated when experimental EC50 and LC50 values become available to achieve a more accurate estimate. Our results emphasize the urgent need to implement wastewater treatment technologies that allow for the efficient removal of these compounds and regulatory measures and monitoring strategies for these compounds to mitigate the ecological risks and public health problems associated with PhCS pollution. The European Union model could be adopted based on a watch list of pollutants that must be analyzed to ultimately have a list of priority substances to control and identify the sources of emissions. Priority pollutants must be included in the official Mexican standard. However, it is essential to recognize that the results of this study are limited to the specific conditions under which the samples were collected, which may not fully reflect the temporal and spatial variability of PhCs concentrations in the Nexapa River. Uncontrolled environmental and operational factors could influence the results, such as variations in wastewater flow, interactions with other contaminants, or adsorption and desorption processes in sludge. These limitations should be considered when interpreting the data and assessing the potential risk to aquatic ecosystems. Therefore, it is recommended that future studies expand the sampling scope and consider these variables to provide a more complete and accurate assessment of the environmental impact of the analyzed pharmaceuticals. In addition, due to the reliance on QSAR-derived toxicity data, the conclusions presented here should be viewed as preliminary. Future studies should also prioritize experimental validation of the predicted toxicity to refine the risk assessment and reduce uncertainty. Finally, interdisciplinary collaboration and innovative technological approaches will be essential to effectively address this complex environmental challenge and ensure the long-term sustainability of aquatic ecosystems.