Preliminary investigation on the occurrence and health risk assessment of antibiotics in cultured tilapia retailed at a commercial outlet in Tema, Ghana

Abstract

In Ghana, Nile tilapia is one of the most commonly cultivated fish species. Bacterial infections, which mostly occur in intensive fish farming, are considered to be the most significant health issue facing these culture systems in Ghana's aquaculture industry. To prevent, and treat bacterial infections and promote fish growth, antimicrobials are often used, and in most cases at unregulated doses. However, this misuse and neglect of withdrawal durations for such antimicrobials may result in drug residues showing up in fish edible tissue, posing a risk to human consumers. To evaluate the risk to consumers, this study screened for antibiotic residues in popular tilapia fish sold at a retail outlet in Tema. Using ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC/MS/MS), the study analysed the levels of 12 antibiotics present in 24 tilapia samples sold at a retail outlet in Tema. Erythromycin, tetracycline, oxytetracycline, and amoxicillin were detected at varying levels, with frequencies of 20.8 %, 62.5 %, 58.3 %, and 54.2 %, respectively. The highest concentration of 3.521 ± 0.32 μg/kg was found for oxytetracycline, while erythromycin had the lowest concentration (0.276 ± 0.11 μg/kg) in the samples. According to the study, the levels of antibiotics detected in the sampled tilapia were lower than the maximum residue limits (MRL) recommended by the WHO. Additionally, both the hazard quotient (HQ) and hazard index (HI) values were less than one. Therefore, consuming retail farmed tilapia purchased from the commercial outlet in Tema metropolis was deemed to pose no significant risk to human health. However, regular monitoring of antibiotics and other contaminants is necessary to minimise their potential impacts on human health.

1. Introduction

Aquaculture systems, particularly intensive ones, are rapidly evolving to address the escalating demand for farmed fish, such as Nile tilapia (Oreochromis niloticus), and to mitigate production shortfalls. These aquaculture systems are gaining prominence, leading to various challenges such as physical stress, overstocking, and deteriorating water quality. Consequently, this phenomenon introduces aquatic microorganisms into the habitat of cultured fish, some of which can instigate diseases such as streptococci and motile Aeromonad septicemia as caused by bacteria, fungi, viruses, and parasites. Experts contend that infectious diseases substantially hinder aquaculture's expansion, significantly elevating morbidity, mortality, and production losses [1]. Developing countries, in particular, grapple with diseases that manifest more prominently and are accountable for half of the losses in aquaculture production [2]. These diseases are estimated to result in an annual economic loss of approximately 6 billion US dollars [2]. In the domain of finfish culture alone, the annual losses range from 1.05 to 9.58 billion US dollars [3]. These infectious diseases predominantly comprise viruses (22.6%), bacteria (54.9%), fungi (3.1%), and parasites (19.4%), all of which have a profound impact on aquatic organisms in aquaculture production [4,5].

Antibiotics, including sulphonamides, quinolones, macrolides, and tetracyclines, have been widespread in veterinary healthcare, human therapy, and as growth promoters in husbandry [1,6]. For instance, in the food animal industry, antimicrobials are incorporated into feed by either direct application or integration into a homogenised extruded diet [7]. Following consumption, approximately 30–90 percent of these added chemicals are excreted from organisms through urine or faeces, as they are generally poorly absorbed [8]. Inadequate removal and improper waste disposal result in the release of these excreted antibiotics into the aquatic environment through various channels, including residential wastewater effluents [9], concentrated animal feeding operation discharges [10], and agricultural field runoff. As a result, antibiotics are pervasive in multiple environmental components, encompassing sediment, surface water, wastewater, groundwater, drinking water, sludge, and silt [11]. This inevitably exposes aquatic organisms, starting with algae as the primary consumers in the food chain, to the ecotoxic effects of these residual antibiotic residues [12]. Earlier studies reported the occurrence of antibiotics in cultured fish from China [[13], [14], [15]], Argentina [16], Iran [17], and South Korea [18]. These studies reported various antibiotics in varying concentrations from different classes, such as tetracyclines, sulphonamides, quinolones, potentiators, pleuromutilin, penicillin, macrolides, lincosamides, cephalosporins, and amphenicols. Since antibiotics tend to induce genotoxicity, promote antibiotic resistance, and disrupt aquatic ecosystems, their usage in aquatic environments, especially in cultured tilapia, is not recommended [19,20]. Moreover, the presence of antibiotic residues in food poses a significant risk to human health. This risk primarily stems from the alteration of the human microbiome and the subsequent emergence and selection of resistant bacteria within the body [21]. These resistant bacteria are responsible for infections that are harder to treat, necessitating the use of limited, costlier, and often more toxic antibiotics [16]. Because of the harmful effects of antibiotics, they have been globally considered as emerging environmental pollutants. Macrolides, sulphonamides, β-lactams, tetracyclines and quinolones are among the most highly detected antibiotic classes in the environmental samples [22]. Thus, it is imperative to scrutinise the prevalent use and availability of antibiotics in aquatic environments.

Despite restrictions on the use of certain antibiotics in animal feed, such as tetracycline and sulfamethazine, since 1989, the quantities employed in 2013 reached a staggering 162,000 tonnes [23]. Tetracyclines (TETs), fluoroquinolones (FQs), and sulphonamides (SAs) constitute the primary antibiotics used in veterinary and human administration, comprising 14%, 15%, and 12% of total antibiotic consumption, respectively [24]. The widespread prevalence of antibiotics in aquaculture products can be attributed to their extensive usage and inadequate management practices [25,26].

Several nations have taken measures, including campaigns to restrict antimicrobial usage in animal-derived food production, to curb the proliferation of resistant microorganisms. Consequently, these nations are compelled to adhere to the regulations of importing countries to eliminate trade barriers [27]. In light of these factors, analysing veterinary drug residues in food through laboratory investigations becomes an integral component of a robust system to ensure food safety [28]. While aquaculture contributes approximately 46% of the total fish supply to meet the protein needs of the growing global population [29], the global use of antimicrobials in food animals, including aquaculture, has significantly increased, with an estimated 63,151 tonnes in 2010 and a projected 67% increase by 2030 [30]. Russia, China, Brazil, South Africa, and India are identified as having the highest global antimicrobial consumption [31]. Antibiotics have not consistently been used responsibly in aquaculture, and the oversight of their use has not provided sufficient assurance in preventing risks to humans [32]. While many studies have found no health risks associated with aquaculture fish and its products collected from diverse regions worldwide, continuous monitoring remains imperative to guarantee consumer safety, especially in developing countries like Ghana, where fish remains the cheapest and most consumed animal protein source.

In late 2018, Ghana experienced a widespread outbreak of infectious spleen and kidney necrosis virus (ISKNV) in tilapia farms, leading to substantial fish mortality in aquaculture systems, as documented by Ramírez‐Paredes et al. [33]. This outbreak reduced tilapia production from 76,000 to 52,000 in 2018 and 2019, as reported by Ragasa et al. [34]. Magna et al. [35] attributed the outbreak to the illicit importation of foreign tilapia strains, inadequate biosecurity measures, and seasonal water quality challenges. According to reports by Refs. [35,36], farmers could mitigate losses by intensifying fingerling production, administering various antibiotics to fish, and implementing autogenous vaccinations based on a comprehensive epidemiological investigation.

In 2018, the Ghanaian government initiated a programme called "Planting for Food and Jobs," with the aquaculture sector as one of its key focal points. This initiative has significantly expanded the aquaculture sector, which, in turn, may exacerbate the presence of residual chemicals in the most widely cultivated tilapia across various culture systems. Approximately 80% of the produce from these farms is retailed at the stationary commercial outlet in the Tema metropolis within the Greater Accra region of Ghana. However, there is a paucity of information concerning the levels of antibiotics in cultured tilapia sold at this prominent commercial outlet in Tema and the associated risks posed to consumers. The present investigation seeks to ascertain the levels of antibiotic contamination and assess the health risks associated with farmed tilapia retailed at the commercial outlet in the Tema metropolis, Greater Accra region of Ghana.

2. Materials and methods

2.1. Chemicals and reagents

Chemicals such as, NaCl, HCl and Na₂EDTA were purchased from Alfa Aesar, and HPLC grade methanol, acetone, and n-hexane from Fisher Scientific (Loughborough, UK). Laboratory-grade Trichloroacetic Acid (TCA) was procured from Lab Alley Essential Chemicals (USA). Cartridges (Oasis HLB SPE, 200 mg/6 mL) for solid phase extraction (SPE) were purchased from Waters Oasis Co. in Milford, USA. For sample extraction, pre-treatment, and instrumental analysis and clean-up, we employed DI water, acetone, citrate buffer, n-hexane, methanol, HCl and Na₂EDTA. All calibration standards and solutions for HPLC analysis were prepared with Ultrapure Milli-Q water, which was purchased from Millipore Bedford in the United States. Certified, high purity (>90.0%) reference standards of all pharmaceuticals, including enrofloxacin (ENR), florfenicol (FLO), sulfamethazine (SMZ), sulfadiazine (SDX), erythromycin (ERT), oxytetracycline (OTC), chloramphenicol (CPN), tetracycline (TET), trimethoprim (TRM), chlortetracycline (CTC), amoxicillin (AMX), and sulfamethoxazole (SMX) were obtained from Pharmacopoeia, United States.

The isotopically-tagged compounds employed as surrogate standards were: erythromycin-¹³C-d₃, sulfamethazine-d_4, enrofloxacin-d_5, trimethoprim-d₃, and flumequine-¹³C₃, purchased from Santa Cruz Biotechnology (Dallas, Texas, USA). The stock standard solution and each isotopically tagged standard were prepared to concentrations of 1000 mgL⁻¹ and 500 mgL⁻¹, respectively, and kept at −20 °C in the dark. Daily preparations of mixed standard solutions comprising all chemicals were made from suitably diluted stock solutions. The chemicals and solvents employed were all of analytical and HPLC quality. Ultrapure water was purified using the Milli-Q system at 18.2 MΩ cm⁻¹ for analysis.

2.2. Sample collection

Twenty-four (24) random samples of Nile Tilapia (Oreochromis niloticus), with an average size ranging from 250 to 500 g, were obtained at a tilapia retail outlet at the Tema metropolis from October to December 2022. The commercial outlet was chosen because it receives farmed tilapia from several farms in and around Ghana's Eastern, Volta, and Greater Accra regions. The fish samples were brought to the Ghana Standard Authority in ice boxes packed with ice, where they were kept at −20 °C until further examination in the laboratory.

2.3. Sample preparation and extraction

The fish extraction method was modified slightly from the previous investigation by Ref. [7]. A 10 g of pulverised fish muscles were homogenised with 5 mL of TCA (10%) for 1 min. The resulting mixture was immediately placed in 35 mL of citrate buffer (0.3 M, pH 4.0), and homogenised for 5 min. The homogenised mixture was emptied into Falcon tube, and centrifuged for 20 min at 5000 rpm. Following the same steps as the initial extraction, the sample residual was once again extracted using 35 mL of acetone. A 100-mL rotating flask was subsequently filled with acetone extract, which was then evaporated to dryness at 38 °C under vacuum. With the use of ultrasound, the dried acetone remnant was thoroughly dissolved in a 30 mL extract of citrate buffer, which was then moved into a 150 mL separation funnel, and then defatted using a liquid/liquid extraction method with n-hexane. The citrate buffer extract was combined with 30 mL of n-hexane and 1 g of NaCl. After manually shaking the mixture for 1 min, the mixture was allowed to stand for a few minutes to allow for complete separation. Centrifugation (5000 rpm, for 15 min) was employed to separate any emulsion that may have formed in the n-hexane and the citrate buffer extract. The SPE cartridge was filled with the citrate buffer extract, which was activated with 5 mL HCl +5 mL of DI water +5 mL methanol. The SPE was then rinsed with 5 mL of DI water and 5 mL of methanol in water (5%). The SPE cartridge was then air-dried at low pressure. 10 mL of methanol was used to elute the desired extract, and it was evaporated at 38 °C until completely dry. After being thoroughly mixed with 500 μL of HPLC-grade methanol, the dried residue was analysed using an ultra-performance liquid chromatography system/tandem mass spectrometry. The analytical procedure for the UHPLC-MS/MS is described on the supplementary document attached.

2.4. Analytical method validation

Following European Commission Regulation 2021/808, the validation of the method was conducted [37]. The assessment covered the following parameters: linearity, precision (including repeatability and reproducibility), decision limit (CCα), limit of detection (LOD), limit of quantification (LOQ), and accuracy determined by estimating trueness (recovery). Using a spiking blank matrix at 0.5, 1.0, and 1.5 times the MRL, the method validation and recoveries were ascertained.

To determine linearity, we established matrix-matched calibration curves. We applied six different concentration levels of fortification to the blank fish samples: 0.5, 10, 50, 100, 200, and 500 μg/kg. The results were analysed using the method of least squares, and the linearity was expressed through the coefficient of determination (R²), with a minimum requirement of R² ≥ 0.99, as recommended by the 2021/808/EC guidelines. The concentration of the target analyte at which, with an error probability of α (α = 1%), it can be concluded that a sample is non-compliant is known as the Decision limit (CCα). The value 1 - α indicates the statistical certainty in percentage that the allowed limit has been exceeded. CCα was computed following EC/2021/808.

Table 1 presents the decision limit (CCα) values for each substance being studied, where the values of CCα ranged from 79 to 228 μg/kg for the antibiotics. Considering the maximum residue levels (MRLs) set at 100 μg/kg, 200 μg/kg, and 50 μg/kg for the specified antibiotics in fin fish by the EU regulatory framework, it can be inferred that the method described is suitable for monitoring antibiotic residues in fish muscle samples.

Table 1.

MRL, LOD, LOQ, and the decision limit (CCα) for antibiotics in Tilapia.

Compound	MRL (μg/kg)	LOD (μg/kg)	LOQ (μg/kg)	CCα (μg/kg)
SMZ	100	0.018	3.162	110
SDX	100	0.040	2.132	113
SMX	100	0.059	2.702	106
OTC	100	0.098	1.683	108
TET	100	0.300	1.739	111
CTC	100	0.250	2.702	107
FLO	200	0.963	3.208	–
ERT	200	0.035	4.117	214
ENF	100	0.061	2.204	108
CPN	100	0.008	3.052	116
TRM	50	0.096	2.319	59
AMX	50	0.132	4.239	63

LOD -Limit of detection, LOQ- Limit of quantification, CCα- Decision limit.

The least spiked concentration that produced a signal-to-noise (S/N) ratio of 3:1 for LOD and 10:1 for LOQ was identified as the LOD and LOQ as shown in Table 1.

The method's accuracy was evaluated through recovery, intraday repeatability, and interday reproducibility. We utilized three spiking levels (0.5, 1.0, and 1.5 times the MRL) with a spiking blank matrix to determine the intraday (single day, n = 6) and interday (three consecutive days, n = 18) recovery values. Precision in terms of the coefficient of variation for repeatability (CV_r) and reproducibility (CV_R) was measured using internal standards and standards-spiked fish samples at the same concentration levels as recovery. Table 2 outlines the spiking levels. To assess reproducibility, three analysts conducted the process over three days, utilising different batches of reagents, solvents, and blank material on different days.

Table 2.

Recoveries and coefficient of variations spiked at 50 μg/kg, 100 μg/kg and 150 μg/kg.

	Spike Level 50 (μg/kg)			Spike Level 100 (μg/kg)			Spike Level 150 (μg/kg)
Analytes	R2	CVr (%)	CVR (%)	R2	CVr (%)	CVR (%)	R2	CVr (%)	CVR (%)
SMZ	0.9991	4.54	5.34	0.9964	8.61	8.84	0.9981	6.72	7.59
SDX	0.9980	5.19	5.20	0.9960	5.88	6.27	0.9989	4.91	6.94
SMX	0.9991	5.60	5.85	0.9994	5.96	6.21	0.9991	4.51	5.59
OTC	0.9984	4.98	5.02	0.9993	3.98	4.93	0.9989	2.29	4.07
TET	0.9987	6.43	7.15	0.9989	5.04	6.48	0.9983	4.17	6.68
CTC	0.9982	4.98	5.02	0.9982	4.24	4.11	0.9986	3.89	4.25
FLO	0.9990	5.19	5.38	0.9991	5.46	6.22	0.9990	5.58	5.50
ERT	0.9989	7.54	7.22	0.9989	4.00	3.86	0.9986	3.10	5.40
ENF	0.9995	7.67	8.08	0.9991	3.83	4.75	0.9992	3.25	5.73
CPN	0.9992	6.39	8.47	0.9982	4.24	4.11	0.9972	4.25	8.23
TRM	0.9996	6.07	6.43	0.9987	5.76	6.12	0.9992	3.18	3.82
AMX	0.9987	8.33	8.53	0.9984	5.62	6.11	0.9997	5.58	5.76

Coefficient of variation for repeatability (CV_r), coefficient of variation for reproducibility (CV_R).

The validation parameters (R² ≥ 0.99 and the coefficient of variation ≤15%) in Table 1, Table 2 show that the method is suitable for analysing contaminants and residues in tilapia fish, including antibiotics, and that it complies with the standards outlined in 2021/808/EC [37].

2.5. Human health risk assessment

2.5.1. Exposure assessment

Based on the average antibiotic detection concentrations in tilapia, exposure assessment was established. Additionally, the following formula was used to determine the estimated daily intake (EDI, in μg) of antibiotics from fish intake:

$$EDI=\frac{C\times IR}{BW}$$ (1)

where C (μg/kg) stands for the level of the antibiotics found in samples, EDI (μg/kg bw/d) represents the presumed daily intake of the antibiotic in fish, and BW (kg) indicates the consumer's body weight (70 kg, the mean adult weight). According to Ref. [38], Ghana has a fish per capita consumption rate of 26 kg/head/year. Therefore, the consumption rate (IR) was estimated as equivalent to 0.071 kg/person/day.

2.5.2. Risk characterization

The potential risk associated with a particular antibiotic is demonstrated by its hazard quotient (HQ), which is calculated as the ratio of the estimated daily intake (EDI) to the appropriate recommended daily intake (ADI). The hazard index (HI) is used to assess the cumulative health risks associated with all of the selected antibiotics. This index represents the combined risk of all antibiotics present in the samples and is calculated by adding up the hazard quotients (HQ) for each antibiotic. The following formulas are used for the calculations:

$$HQ=\frac{EDI}{ADI}$$ (2)

$$HI=\sum HQ$$ (3)

The Supplementary Table S1 displays the recommended daily intake (ADI) values for the antibiotics obtained from the literature. Typically, a hazard quotient (HQ) value of ≥1 indicates a significant health risk, while an HQ value of <1 suggests that the daily consumption dose is acceptable [39].

2.6. Statistical analysis

The information about antibiotics was recorded in Microsoft Office Excel (Microsoft Office 2021). The frequency of detecting antibiotics and statistical information such as the average, standard deviation, and range were calculated. Samples with antibiotic levels below the LOQ were not included in the frequency calculation. For estimating the EDI, antibiotic levels lower than the LOQ were replaced with values of LOD/2 [40].

3. Results and discussions

3.1. Antibiotics concentration in farmed tilapia

In all, four antibiotics (oxytetracycline, tetracycline, amoxicillin, and erythromycin) out of the 12 monitored antibiotics in the farmed biota samples sold at the retail outlet in this study were found at varying levels, as shown in Table 3. Again, it was observed that no residual levels of sulfamethoxazole (SMX) were quantified in all the investigated samples. These were grouped into different families such as tetracyclines (OTC, TET), β-lactams (AMX), macrolides (ERT) and sulphonamides (SMX). Their means in the farmed tilapia decreased in the following orders: TCs > β-LMs > MLs > SAs. With a mean concentration of 1.836 ± 0.18 μg/kg, the total amount of antibiotics discovered in the farmed tilapia varied from LOQ to 5.380 μg/kg.

Table 3.

Concentrations and detection frequencies of antibiotics in farmed tilapia retailed at a commercial outlet, Tema.

Antibiotics	Range (μg/kg)	Conc.± SD (μg/kg)	Detection frequency (%)
OTC	LOQ – 5.380	3.521 ± 0.32	58.3
TET	LOQ – 4.262	2.892 ± 0.18	62.5
AMX	LOQ – 1.537	0.655 ± 0.12	54.2
ERT	LOQ – 0.814	0.276 ± 0.11	20.8

LOQ – of quantification, SD – standard deviation, OXY – oxytetracycline, TET – tetracycline, AMX – amoxicillin, ERT – erythromycin.

Tetracyclines (TC), which accounted for 87.3% of the overall antibiotic load in the examined samples, were generally the most prevalent antibiotic class. The levels of oxytetracycline in farmed tilapia varied from LOQ to 5.380 μg/kg, with an average of 3.521 μg/kg, which was the highest among the two types of tetracycline antibiotics that were examined. Tetracycline had an average of 2.892 μg/kg and concentrations varied from LOQ to 4.262 μg/kg. It was the most commonly detected tetracycline antibiotic, with a detection frequency of 62.5%. Tetracyclines were classified as the most effective antibiotics in Africa [41], presumably as a result of their broad spectrum, low cost, and efficacy [42]. The findings of the current study showed that TCs (with high detection frequencies of 58.3% and 62.5%) were often used by farmers to monitor fish health and treat diseases, without adhering to the recommended withdrawal period for these antibiotics. Additionally, in some cases, the reason for high levels of tetracycline antibiotics in fish samples could be the use of substances containing antibiotics, like poultry manure, to improve plankton growth in aquaculture systems [43]. The maximum residual limits (MRL) of 100 μg/kg established by the European Commission [14] were not exceeded in any of the present study's tilapia samples that were contaminated with TCs.

[44] discovered that farmed fish contained oxytetracycline levels of up to 60 μg/kg when they were obtained from their natural habitat. According to their findings, the oxytetracycline residuals were below the maximum residue limit (MRL) established by the EU and the Korean Food Legislation. [45] conducted an investigation in Shahrekord, Iran, to evaluate the presence of oxytetracycline contaminants in 50 Rainbow trout flesh samples collected from markets. They discovered that before frying, only 6 % (3) of the samples had lower levels of oxytetracycline than the maximum residue limits established by Codex Alimentarius, while 24 % (12) of the samples had lower levels after frying. [17] conducted a study to determine the existence of tetracycline residues in 138 rainbow trout muscle samples from Iranian trout farms. They discovered that 63.1 % of the samples contained tetracycline residues when detected at quantifiable levels, with the highest sample containing residues ranging from 1.43 to 91.130 μg/kg below the oxytetracycline maximum residue limit set by the European Commission. [46] investigated the residual effects of tetracyclines (tetracycline, chlortetracycline, doxycycline, and oxytetracycline) using 70 fish samples from 70 different fish farms in Turkey's Mugla province. Tetracycline levels were not found to be exceeding the detection threshold. The studies conducted by Refs. [[47], [48], [49]] were consistent with the previously mentioned findings. They discovered oxytetracycline in tilapia fish samples, but the amounts were below the MRL. In contrast, [50] examined 193 Nile Tilapia samples from Brazil and did not find any tetracycline antibiotics. Additionally [51], evaluated 14 antibiotics, including quinolones and tetracyclines, and did not detect tetracyclines in 26 tilapia samples. Furthermore, reckless use of oxytetracycline in aquaculture can have harmful consequences if it leaves a residue in fish muscles. Many authors, including [[52], [53], [54]], contrary to the current study, observed significant oxytetracycline accumulation in farmed fish that exceeded the maximum residual limit established by the European Commission.

The macrolides (MLs) are commonly formulated in aquaculture feeds to enhance growth and for the treatment and prevention of diseases, as reported by Refs. [55,56]. In all samples, MLs had a low detection frequency of 20.8 %. Only ERT was found in 20.8 % of the samples out of the two MLs (enrofloxacin and erythromycin). It is important to note that the levels of erythromycin varied from LOQ to 0.814 μg/kg, with an average of 0.276 ± 0.11 μg/kg and levels below the MRL value of 100μg/kg. [57] reported that the primary antibiotic discovered in adult Fenneropenaeus penicillatus from Hailing Island was erythromycin, with levels ranging from 2498 to 15090 μg/kg. [58] found that cultured fish from various common aquacultures had ERT levels of 3.5–12 μg/kg. These values exceeded those that were seen in the current investigation. These findings suggested that the significant variation in amounts found in aquatic products could be attributed to a variety of potential ML sources, such as residues in sediments, disease treatment, feed additives and residues in water used for aquaculture ponds [59].

Amoxicillin was the only β-lactam group detected in the farmed tilapia samples, and its concentration ranges from LOQ to 1.537 μg/kg. It recorded an average value of 0.655 ± 0.12 μg/kg and accounted for about 54.2 % of the sampled fish. The study's findings indicate a high occurrence of amoxicillin, suggesting its widespread and indiscriminate use in the fish farms where these fish were bred. Levels of AMX for the present study were all below the MRL value of 50 μg/kg. According to Ref. [60], tilapia tissues collected from wet markets in Chittagong, Bangladesh, have a mean concentration of 683.2 μg/kg (with a range of 6.4–1525.2 μg/kg) of amoxicillin. In contrast to their study, the levels found in the present investigation were lower. This study's high incidence of amoxicillin residue might be the result of its widespread use due to its market accessibility and affordable pricing.

The data in Table 3 demonstrated that SA, including the SMX, was below the limit of quantification (<LOQ) in the samples from the retail centre. [43] found SMX in tilapia in El-Fayoum in another Egyptian study, with a mean concentration of 52.41 μg/kg. In a study conducted by Ref. [50] in Brazil, no SA molecules were detected in 193 samples of Nile Tilapia fish. According to Ref. [61], sulphonamide levels in fish in Argentina were detected to be high, varying from 0.7 μg/kg to 5.6 μg/kg. Meanwhile [62,63], discovered that sulfamethoxazole contamination in Pangasius catfish products exceeded the EU standard by 246 μg/kg in Thailand, along with enrofloxacin contamination. At both permissible aquaculture capacity (100 mg/kg/day) and environmental exposure concentrations (260 ng/L), sulfamethoxazole exposure changes the metabolism of nutrients and suppresses the inherent immune system in tilapia [64]. [64] reported that sulfamethoxazole exposure resulted in increased superoxide dismutase (SOD) transcription levels in tilapia's liver and intestine tissues. Additionally, sulfamethoxazole was found to enhance cytokine mRNA expression, such as TNF-α and IL-1β. Because of this, even at low doses and even if they have no effect on fish survival, some sulphonamides, such as SMX, can alter fish physiology and genetics.

3.2. Human health risk assessment

Human health concerns may result from consuming aquatic products polluted with antibiotics. [56] suggested that the potential risks to human health due to exposure to antibiotics through diet should be evaluated. To assess the potential health hazards associated with consuming contaminated fish, the hazard quotient (HQ) was calculated. The risk to human health associated with consuming residue-containing cultured tilapia is expressed by the hazard quotient, which also illustrates the severity of the hazardous effect.

Based on Table 4, the calculated EDI values for individual antibiotics ranged from 299 × 10⁻⁵ to 3.57 × 10⁻³ μg/kg bw/d. The highest EDI value was detected in TCs, reaching 3.57 × 10⁻³ μg/kg bw/d. The estimated daily intakes (EDI) for every antibiotic under study were found to be significantly lower than the acceptable daily intakes (ADI). Similarly, the estimated daily intake (EDI) derived from the average daily fish consumption in Egypt was determined to be lower than the acceptable daily intake values established by the World Health Organization [43]. Additionally, a recent study conducted in Argentina on fish revealed that the estimated dietary exposure to multiantibiotic residues in commercial fish did not surpass the acceptable daily intake [16]. Unlike the present study, the acceptable daily intake of antibiotics for individuals consuming fish in Changchun and Shenyang was determined to be lower than the EDI value for Yellow Sea fish (0.065 μg/kg bw/day) [66]. The HQ values for antibiotics also varied between 2.30 × 10⁻⁷ and 7.14 × 10⁻⁴. These hazard quotient values less than one demonstrated that the toxicological effects of fish residues found in fish did not significantly affect the health of fish consumers in the study area.

Table 4.

HQ of antibiotics based on the estimated daily exposure dose of Retailed farmed tilapia.

Antibiotics	EDI (μg/kg bw/d)	ADI (μg/kg bw/d)	HQ	MRL (μgkg−1)
SMX	$2.99\times {10}^{-5}$	130a	$2.30\times {10}^{-7}$	100
OTC	$3.57\times {10}^{-3}$	5c	$7.14\times {10}^{-4}$	100
AMX	$6.64\times {10}^{-4}$	2b	$3.32\times {10}^{-4}$	50
TET	$2.93\times {10}^{-3}$	5c	$5.86\times {10}^{-4}$	100
ERT	$2.80\times {10}^{-4}$	2c	$1.40\times {10}^{-4}$	100
$\mathit{H}\mathit{I}=\sum \mathit{H}\mathit{Q}=\mathbf{1.77}\times {\mathbf{10}}^{-\mathbf{3}}$

EDI-Estimated daily intake, ADI-Acceptable daily intake, HQ-Hazard quotient, HI-Hazard index, ADI according to: a). [65], b) [22]. c). [14], MRL according to [16].

Although the HQ approach for risk characterization focuses on the potential toxicity of a single stressor, such as antibiotics, real-life exposure situations do not align with these methodologies. Because, the human body is exposed to a variety of stressors every day from different sources, when stressors, such as antibiotics, are combined, they often have a synergistic impact that reduces the quantity of each chemical required to cause harm compared to the amount determined when studying the toxicity of each drug [67]. Since SA, ML, β-LM, and TC concentrations were found in Tilapia fish, according to Ref. [43], the current study also assessed the possible risk (hazard index (HI)) predicated on the aggregate exposure to those concentrations. The intake of farm-raised tilapia that has been contaminated with OTC, TET, AMX, SMX, and ERT poses no risk to the Ghanaian population, according to the Hazard Index (HI) for the antibiotic cocktail, which was less than one. Therefore, a more extensive monitoring programme and risk assessment are needed in order to determine the risks connected with food exposure for human health in the future.

4. Conclusions

The findings show that OTC, TET, AMX, and ERT were found in samples in various amounts and that the level of accumulation varied among tilapia fish. While SMX was not quantified. All antibiotics found in the muscle tissue of Oreochromis niloticus were below the FAO/WHO MRL, suggesting that fish from Tema's market centre may not ultimately trigger OTC, TET, AMX, ERT, or SMX-related health issues. The researchers demonstrated that no potential adverse effect on public health was observed from consuming tilapia samples with the measured levels of SA, MLs, β-LMs, and TCs based on current consumption trends. Since there is no discernible health risk from the combined effects of all antibiotics, their hazard index (HI) is less than one. Our study was limited by its brief duration and low-resource laboratory environment. Determining the full status of various antibiotic residues in different cultured fish samples throughout the year and quantifying the risks to consumer health associated with them will take a few years. Additionally, the sample size of the present research, which evaluates the health risk is constrained. To assess the risk and formulate recommendations with broad national ramifications, a more comprehensive analysis is needed. The findings of this study serve as a foundation for policymaking and further investigation, particularly in the context of risk analysis to safeguard public health. Therefore, expanding upon this study and developing a database that includes the concentrations of various antibiotic residues and their corresponding risk levels could enhance efforts to guarantee food safety and protect public health.

Funding statement

No funding was received to assist with the preparation of this manuscript.

Data availability

The datasets generated during the current study are available from the corresponding author on reasonable request.

CRediT authorship contribution statement

Emmanuel Kaboja Magna: Writing – review & editing, Writing – original draft, Investigation, Formal analysis, Data curation, Conceptualization. Francis Ofosu-Koranteng: Validation, Software, Methodology. Ruby Asmah: Supervision. Emmanuel Tetteh-Doku Mensah: Visualization, Supervision, Formal analysis. Ebenezer Koranteng Appiah: Visualization, Resources. Patrick Senam Fatsi: Writing – review & editing, Visualization, Project administration. Frank Adu-Nti: Resources, Project administration. Zenobia Castel Kpodo: Resources, Project administration, Data curation. Ishmael Lente: Visualization, Supervision.

Declaration of competing interest

We the authors have no relevant interest(s) to disclose.

Appendix A. Supplementary data

The following is the Supplementary data to this article: