Introduction

Contaminants of Emerging Concern (CECs) are biologically active and potentially toxic molecules of recent or prolonged use in which its isolated and combined effects to the aquatic ecosystem are still unknown (Deere et al. 2020). CECs include pesticides, fragrances, plasticizers, hormones, flame retardants, nanoparticles, siloxanes, among others (López-Pacheco et al. 2019); however, the main group of CECs are pharmaceutical and personal care products (PPCPs) such as antibiotics, anti-inflammatory, central nervous system stimulators, ß-blockers, lipid regulators, anticonvulsant, X-ray contrast media, insect repellents, antimicrobials, preservatives and sunscreen UV filters (Kovalakova et al. 2020; Liu et al. 2020b).

Research carried out in the last 10 years had detected the presence of CECs in Brazilian aquatic environments. Arsand et al. (2020) studied the occurrence of 40 antibiotics of different classes in surface water from Dilúvio River during a 2-year period and its association with the presence of antibiotic resistance genes. Roveri et al. (2020) screened and quantified 23 pharmaceutical compounds (including illicit drugs), at two sampling points near the diffusers of the Guarujá submarine outfall, State of São Paulo, Brazil, where caffeine, diclofenac, valsartan, benzoylecgonine and cocaine were the main compounds detected. Santos et al. (2020) monitored pharmaceutical compounds during 1 year in four Brazilian water sources, aiming to understand the factors that influence their occurrence and removal in conventional drinking water treatment plants (DWTPs) and to assess the environmental and human health risks. Starling et al. (2019) published the first review about occurrence, control and fate of CECs in environmental compartments in Brazil, in which data gathered indicated that caffeine, acetaminophen, atenolol, ibuprofen, cephalexin and bisphenol A occur in μg L−1 range in streams near urban areas. However, most of the published studies have not carried out an environmental risk assessment; therefore, the risks caused in aquatic organisms are still unknown.

Brazil is the fifth largest country in extension and the sixth most populous in the world, with an area of 8,515,767 km2 and about 211 million inhabitants. It is divided into five geographic regions, which present considerable differences in relation to economic, demographic and sanitation characteristics. The southeast region houses the largest industrial park in the country and more than a third of the Brazilian population, about 89 million inhabitants. While in the other regions, dwell about 57, 30, 18 and 17 million inhabitants in northeast, south, north and midwest, respectively (IBGE 2020). According to the latest Statistical Yearbook of the Pharmaceutical Market published in Brazil, there are 1794 different active pharmaceutical ingredients registered, totaling 6587 pharmaceutical presentations on sale. Among the most commercialized pharmaceuticals products are acetaminophen, atenolol, ibuprofen and sodium diclofenac in association with caffeine (BRASIL, 2018).

The environmental occurrence patterns of pharmaceuticals could be estimated establishing a relationship between human drug use and its presence in the aquatic environment. However, available data is still inadequate to establish the country-specific data for pharmaceutical consumption due to the uncertainty related to the consumption estimates and related to those pharmaceuticals sold as unregulated “over the counter drugs”. Besides, the consumption patterns of pharmaceuticals can be influenced by socioeconomic conditions, seasonal changes and on the basis of location/region, affecting the establishment of the relation of PPCP consumption patterns with environmental occurrence (Patel et al. 2019).

Determination of CECs in the environment increased in the last decade due to advancement of analytical chemistry and sample preparation techniques that allowed the quantification of concentrations between μg and ng L−1 (López-Pacheco et al. 2019). However, in addition to occurrence studies, a more careful view must be given to the environmental risks caused by CECs to aquatic organisms. Chemicals with low detection frequency but with high ecological risk should be carefully analyzed. More approaches are needed to provide a rigorous scientific basis for identifying the environmentally hazardous compounds. Therefore, the aim of this study was to conduct a detailed review of CECs (pharmaceuticals and personal care products, illicit drugs, plasticizers and hormones) in Brazilian surface water and to perform environmental risk assessment (ERA) using concentrations reported in the last 10 years (2010 to 2020), generating technical support for risk management of CECs in Brazil.

Material and methods

Review and data collection

Occurrence and quantification data of CECs were obtained from previous studies performed in the Brazilian aquatic ecosystem. Peer-reviewed publications published between 2010 and 2020 using the keywords “emerging contaminants”, “PPCPs”, “surface water” and “Brazil” were reviewed. These studies reported data on 73 CECs, including pharmaceuticals, hormones, plasticizers and personal care products (PCPs) detected in surface waters.

Ecotoxicity data of chemicals for algae, crustaceans and fish were obtained from the ECOTOX database of United States Environmental Protection Agency (USEPA) and previous studies. Effect concentration 50 (EC50) and lethal concentration 50 (LC50) were selected as endpoints for each species for growth, reproduction or lethality. The lowest EC50 or LC50 was selected when there were more than one value reported for the same endpoint. Toxicity data for at least three trophic levels for each chemical was used.

Environmental risk assessment

Environmental Risk Assessment (ERA) of CECs in Brazilian aquatic environment was conducted according to the Technical Guidance Document on risk assessment from European Commission (Eur, E. C 2003) for ecological multiple level using the Risk Quotient (RQ) (Eq. 1).

$$ RQ=\frac{MEC}{PNEC} $$
(1)

In Eq. 1, MEC is the maximum environmental concentration measured on surface water samples (ng L−1), and PNEC is the predicted no-effect concentration to aquatic organism (ng L−1). Ecotoxicological data are presented in Supplementary Material—Table S1, in addition to being available on the US EPA ECOTOX database (2020) (https://comptox.epa.gov/dashboard/). In determining of PNEC values for each chemical and species, an appropriated assessment factor (AF) was used to account for any uncertainty associated with the available data. AFs were selected according to Technical Guidance Document from the European Commission (AF 10, 50, 100 or 1000, depending on the toxicity values used) (Eur, E. C 2003). Environmental risk was divided into four classes: RQ < 0.01, insignificant risk, RQ ≤ 0.1, low risk and no adverse effects are expected, 0.1 ≤ RQ ≤ 1.0, moderate risk, therefore possible adverse effects should be taken into consideration and RQ ≥ 1.0, high risk, thus it is probable that adverse effects may occur (Sharma et al. 2019).

Prioritization of CECs in Brazilian aquatic environments

Based on the frequency of detection of CECs in Brazilian surface water, concentrations and the environmental risk involved, chemicals were selected to compose the priority ranking of CECs in Brazil, considering studies from the last decade. Those with frequency of detection ≥ 50% were selected, in addition to those with moderate to high environmental risk. Chemicals with high environmental risk reflect the impact of urbanization and industrialization on the environment. Less populated and less socioeconomically developed regions (e.g. Brazilian North, 18,195,973 population) showed a smaller number of studies; however, lower concentrations are observed for most of the CECs detected when compared to large urban centers (e.g. Brazilian Southeast, 89,452,960 population). Therefore, despite there are few studies in Brazil when compared to other countries (China and Spain), this study helps in identification of priority compounds.

Results and discussion

Distribution and frequency of detection of CECs in Brazilian surface water

Studies on the CEC occurrence, mainly PPCPs, in surface water in Brazil have become more frequent in the last decade. Caldas et al. (2013), Thomas et al. (2014), Campanha et al. (2015), Pereira et al. (2016), López-Doval et al. (2017), Barros et al. (2018), Sousa et al. (2018), Reis et al. (2019), Roveri et al. (2020), among others, studied PPCPs in Brazilian waters, including compounds from anti-inflammatories, anticonvulsant, ß-blockers, plasticizers, hormones, bactericides, preservative and solar UV-sunscreen classes. Locatelli et al. (2011), Jank et al. (2014), Monteiro et al. (2018) and Arsand et al. (2020) showed the presence of antibiotics of different classes in waters from Atibaia, Dilúvio and Guandu Rivers, respectively. While Montagner et al. (2014), Galinaro et al. (2015), Silva et al. (2015), Santos et al. (2016) and Ide et al. (2017) screened PCPs in water bodies from Southeast Brazil.

Among thirty-five peer-reviewed publications about occurrence of CECs in Brazilian surface water, twenty were carried out in the southeast region (in São Paulo, Rio de Janeiro and Minas Gerais States), twelve in the south region (in Rio Grande do Sul and Paraná States), one in North (Amazonas State), one in Midwest (Mato Grosso do Sul State) and one in Northeast (Maranhão State). Table 1 shows the CECs found in each study, concentration levels, its maximum environmental concentration measured (MEC) and frequencies of detection (Fd). From all reviewed studies, 73 CECs were identified, of which nine drugs (losartan, metformin, nimesulide, ibuprofen, atenolol, acetaminophen, diclofenac, caffeine and albendazole) make up the ranking of the 20 most commercialized pharmaceutical substances in Brazil (BRASIL, 2018). São Paulo was the state with the highest number of different CECs detected, 41 in total.

Table 1 Sampling location, detected chemicals, range of concentration in ng L−1 (mean concentration), frequency of detection, Fd in % (sample number, n) and maximum measured environmental concentration in ng L−1 (MEC) in Brazilian surface water
Full size table

Caffeine deserves attention, since it was found in 41% of the reviewed papers, including studies in all Brazilian regions, reaching concentrations from 7 to 129,585 ng L−1 and mean concentrations from 80 to 14,955 ng L−1. Caffeine mean concentrations in some studies were higher than concentrations reported in Uruguay (200 ng L−1) (Griffero et al. 2019) and India (743 ng L−1) (Sharma et al. 2019) and lower than that found in Ecuador (248,686 ng L−1) (Voloshenko-Rossin et al. 2015). Caffeine is an emerging contaminant consumed by the population in coffee, tea, drinks and medicines, considered an indicator of human contamination that has been widely detected in aquatic systems worldwide (Dafouz et al. 2018). Rigueto et al. (2020) showed that the removal of caffeine in WWTPs using different treatment methods occurs from 46 to 98%; therefore, high concentrations of this compound in surface waters may indicate untreated sewage disposal.

Evaluating by class, from 73 CECs detected in the reviewed studies, 25% belong to the antibiotics class, and 21% were NSAIDs (Figure 1). Among antibiotics, sulfamethoxazole and trimethoprim presented frequencies of detection of 18% and 15%, respectively (Table 2). Sulfamethoxazole was quantified in concentrations from 0.6 to 572 ng L−1 reaching mean concentrations from 1 to 458 ng L−1. Some studies in Brazil quantified concentrations higher than that found in China (26 ng L−1) (Li et al. 2018) and in India (8.5 ng L−1) (Sharma et al. 2019) and other, lower than reported in Kenya (1214 ng L−1) (Kairigo et al. 2020) and in Mexico (173–1143 ng L−1) (Rivera-Jaimes et al. 2018). Trimethoprim, commonly used in combination with sulfamethoxazole, was determined in concentrations from 1 to 484 ng L−1 and mean concentrations ranging from 3 to 85 ng L−1, which is lower than found in Colombia (210–3580 ng L−1) (Bedoya-Ríos et al. 2018).

Fig. 1
figure 1

Frequencies of detection of therapeutic classes in Brazilian aquatic environment, considering thirty-five studies available on literature

Full size image
Table 2 Frequencies of detection (Fd %) of individual emerging contaminants in Brazilian aquatic environment, considering thirty-five studies available on literature
Full size table

In NSAID class, diclofenac (41%), acetaminophen (32%), ibuprofen (24%) and naproxen (15%) were the most detected (Table 2). Similar to caffeine, diclofenac also deserves attention since it was detected in 41% of studies, and it has been reported in several countries, having been included in the list of priority substances in the EU Water Framework Directive (Eur, E. C 2013). Diclofenac was quantified in concentrations from 4 to 2626 ng L−1 and mean concentrations from 20 to 1161 ng L−1. The highest mean concentrations quantified in Brazilian studies are higher than those found in China (67 ng L−1) (Dai et al. 2015) and lower than those quantified in South Africa (600–8174 ng L−1) (Agunbiade and Moodley 2016). In Brazil, diclofenac, together with nimesulide, ibuprofen and acetaminophen, is the most commercialized NSAID. In 2018, between 25 and 50 million presentations of diclofenac and nimesulide were sold, while for ibuprofen and acetaminophen, between 50 and 100 million presentations were sold (BRASIL, 2018).

Acetaminophen presented concentrations from 1.2 to 30,421 ng L−1 reaching mean concentrations from 1.3 to 6860 ng L−1, while ibuprofen, from 3 to 2710 and mean concentrations from 33 to 1472 ng L−1. Mean concentrations in some studies are higher than those reported in Ganges River, India (2 and 23 ng L−1) (Sharma et al. 2019) and Jiulong River, China (2 and 69 ng L−1) (Lin et al. 2016). South American countries have reported high concentrations of these compounds. In Mexico, concentrations from 354 to 14,900 (1634) ng L−1 of acetaminophen, from 284 to 2835 (520) ng L−1 for ibuprofen and from 258 to 2470 (740) ng L−1 for diclofenac were detected (Rivera-Jaimes et al. 2018). In Colombia, Pemberthy et al. (2020) quantified 460 ng L−1 for ibuprofen and 310 ng L−1 for diclofenac in Gulf Urabá. Brazil is among the ten largest consumers of medicines in the world. According to Aitken and Kleinrock (2015), over 50% of the world population will consume in 2020 more than 1 dose per person per day of medicines, up from one third of the world in 2005, driven by India, China, Brazil and Indonesia.

Preservatives, hormones, plasticizers, UV-sunscreen and ß-blockers are 6%, each, from 73 ECs detected in this study (Figure 1). In the preservative group, ethylparaben (29%), methylparaben (18%) and propylparaben (15%) presented the highest frequencies of detection. Ethylparaben was detected in concentrations from 2 to 29,800 ng L−1, while methyl and propylparaben were reported in mean concentrations from 8 to 170,870 ng L−1 and from 13 to 243 ng L−1, respectively. Methyl and propylparaben were quantified in some studies in concentrations higher than those reported in Jiulong River, China (21 and 16 ng L−1) (Sun et al. 2016) and Pakistan (6 and 3 ng L−1) (Ashfaq et al. 2019).

Hormones (17α-ethynylestradiol, 17ß-estradiol, estrone and estriol) were detected mainly in the southeast region. 17ß-estradiol concentrations ranged from 0.3 to 6808 ng L−1 and 17α-ethynylestradiol from 6 to 4390 ng L−1 with mean concentrations from 1.8 to 2515 and from 15 to 1957 ng L−1 of 17ß-estradiol and 17α-ethynylestradiol, respectively. Some studies reported 17ß-estradiol and 17α-ethynylestradiol concentrations lower than those reported in Uruguay (2350 and 11,600 ng L−1, respectively) (Griffero et al. 2019), while other, higher than those quantified in Malaysia (31 and 8 ng L−1) (Ismail et al. 2019). Estrone occurred in 18% of the studies, in concentrations ranging from < 0.1 (Monjolinho River) to 940 ng L−1 (Iguaçu River) (Campanha et al. 2015; Ide et al. 2017), while estriol was quantified in concentrations from 1 to 1398 ng L−1 in Atibaia River (Montagner et al. 2019). In developed countries, as in the USA and Spain, relatively low concentrations are reported for hormones. Deere et al. (2020) reported 54, 89 and 726 ng L−1 for 17α-ethynylestradiol, 17ß-estradiol and estrone, respectively, in Minnesota. While Gorga et al. (2015) found up to 2, 8, 7 and 6 ng L−1 for 17α-ethynylestradiol, 17ß-estradiol, estrone and estriol in Iberian rivers. The presence of hormones in Brazilian surface waters is mainly related to the use of oral and injectable contraceptives by Brazilian women. About 65% of women aged 15 to 49 reported using a modern contraceptive method (Farias et al. 2016). Steroid estrogens gained scientific attention in the last decades due their potential to cause undesirable ecological effects at very low concentrations (0.1–0.5 ng L−1), being considered as prospective endocrine disruptors (Ilyas and Van Hullebusch 2020).

On plasticizers group, bisphenol-A, dibutylphthalate, diethylphthalate and dioctylphthalate were detected. Bisphenol-A was quantified in 21% of studies, in concentrations from 2 to 39,860 ng L−1 and mean concentrations from 20 to 15,890 ng L−1. This chemical has exhibited genotoxicity, reproductive toxicity, endocrine disrupting effects, cytotoxicity and neurotoxicity in these concentration levels (Liu et al. 2020a). Global bisphenol-A production had exceeded 4.6 million tons in 2012 and is expected to grow at an annual rate of 4.6% from 2013 to 2019. These data highlight the concern with the presence of this compound in the environment, since it has proven adverse effects to living beings and it has been detected in many biological matrices including human amniotic fluid, blood, breast milk, placenta, sweat and urine (Wang et al. 2021).

On pharmaceutical ß-blockers group, atenolol, metoprolol and propranolol were detected. Atenolol presented frequency of detection of 15%, being found in the southeast region in concentration level from 0.1 to 8199 ng L−1. This compound has not been detected in other Brazilian regions. This is one of the most used ß-blocker to control blood pressure in Brazil, being marketed between 50 and 100 million presentations in 2018 (BRASIL, 2018).

On illicit drugs group, only cocaine and its metabolite benzoylecgonine were studied. These compounds were detected in 18% of the reviewed studies, in concentrations ranging from 0.3 to 5896 ng L−1 for cocaine and from 0.3 to 3586 ng L−1 for benzoylecgonine in the north region (Thomas et al. 2014). In the southeast region, concentrations from 13 to 537 ng L−1 for cocaine (Pereira et al. 2016) and from 3 to 179 ng L−1 for benzoylecgonine (López-Doval et al. 2017) were quantified. North American countries have shown the presence of these compounds in its surface water; Deere et al. (2020) quantified up to 259 ng L−1 for cocaine and 61 ng L−1 for benzoylecgonine in the USA, while Comtois-Marotte et al. (2017) found up to 4 ng L−1 for cocaine and up to 10 ng L−1 for benzoylecgonine, in Canada. Fontes et al. (2019) showed that cocaine and their metabolites are widespread in aquatic ecosystems in levels able to trigger sub-lethal effects to non-target organisms, besides to concentrate in seafood, presenting risks to human health and the environment.

Carbamazepine, an anticonvulsant drug, was detected in Brazilian surface water with concentrations from 0.1 to 659 ng L−1 and mean concentration up to 207,131 and 36 ng L−1 in North, Southeast and Northeast regions. Some Brazilian studies reported carbamazepine concentrations lower than those found in other Latin American countries (195,943 ng L−1 in Ecuador and 4300 ng L−1 in Colombia) and other, higher than European countries (37 ng L−1 in Italy, 29 ng L−1 in Spain and 25 ng L−1 in German) (Carmona et al. 2017; Kötke et al. 2019; Feo et al. 2020). Studies suggest that rivers from developed countries contain less PPCPs than developing countries, primarily attributable to better medical regulations rather than availability of wastewater treatment facilities (Kumar et al. 2019).

Other chemical detected in Brazilian surface waters was triclosan, an antimicrobial agent widely used in personal care products such as soaps, skin creams, toothpaste and deodorants. In Brazil, the National Health Surveillance Agency (ANVISA) allows concentration of up to 0.3% of triclosan in personal care products as a preservative (BRASIL 2012). Triclosan was reported in 24% of the reviewed studies in concentrations from 2 to 789 ng L−1, reaching mean concentration up to 253 ng L−1 in Upper Tibagi River (Reichert et al. 2020) and 69 ng L−1 in Jundiaí River (Sousa et al. 2014). These concentrations are higher than those quantified in Nigeria (59 ng L−1) (Inam et al. 2015) and in China (22 ng L−1) (Sun et al. 2016) and lower than those found in Colombia (295 ng L−1) (Pemberthy et al. 2020) and in the USA (1830 ng L−1) (Deere et al. 2020).

Therefore, evaluating the concentrations of CECs quantified in Brazilian surface waters, a high level is observed for some compounds when compared to developed countries such as the USA and European countries. However, when comparing the levels found in Brazil with other South American countries, lower concentrations are observed for most compounds. Caffeine, diclofenac, acetaminophen, ethylparaben, ibuprofen, triclosan and bisphenol-A deserve attention, since among the ECs reviewed in this study, they had the highest frequencies of detection.

Ecological risk of CECs in Brazilian surface water

Using the MEC values reported in studies carried out in the last decade in Brazil, an environmental risk assessment multiple-level ecological was performed. In Figure 2, risk quotients (RQ) for chemicals found in each Brazilian region are presented. Southeast seemed to be the region most impacted by CECs in Brazil; however, few studies have been conducted in the North, Northeast and Midwest regions, and it is not possible to have a complete outline of the real situation of ECs in the country. Analyzing the three trophic levels, daphnia was the most sensitive organism being subject to high risk mainly for acetaminophen (950), caffeine (431), 17ß-estradiol (340), ibuprofen (309), dibutylphthalate (292), diclofenac (262) and methylparaben (106). Therefore, it is probable that adverse effects may occur to aquatic biota exposed. In addition to these ones, 17α-ethynylestradiol (42), diethylphthalate (14), sertraline (2.5), triclosan (2) and bisphenol-A (2) presented high environmental risk in the southeast region.

Fig. 2
figure 2

Risk Quotients (RQ) for maximum concentration measured in surface water from Brazilian southeast (a), northeast (b), south (c), north (d) and midwest (e). The dashed lines represent the separation of the levels of environmental risk. Values below the dashed lines represent low risk, between the lines, moderate risk and above, high environmental risk

Full size image

High environmental risk resulting from high concentrations of CECs quantified in the southeast region may be related to the greater socioeconomic development of this region. The large industrial centers in Brazil are concentrated in the southeast of the country, with emphasis on the São Paulo State, which is home to large chemical and pharmaceutical industries. In addition, the southeast region stands out for containing the cities with the highest demographic densities in the country, which also contributes to environmental contamination by CECs, due to generating a greater volume of domestic sewage when compared to less populous cities.

Considering the vast geographical area of Brazil, the differences in the population density and the wide variation in the climate conditions that influences the pattern of the PPCP consumption, the discussion was done considering the spatial distribution. However, considering the highest RQ found in the Brazilian regions, a comparison with other studies around the world was carried out after.

In the south region, acetaminophen, 17ß-estradiol, 17α-ethynylestradiol, caffeine, estrone and triclosan presented high environmental risk (RQ > 1.0), while methylparaben, gemfibrozil and sulfamethoxazole present moderate risk (0.1 < RQ < 1). Other chemicals (acetylsalicylic acid, bisphenol A, butylparaben, ciprofloxacin, ethylparaben, norfloxacin, propylparaben and trimethoprim) showed low or insignificant risk in the region. In the North Brazil, high risk was estimated to diclofenac and sertraline, moderate risk to carbamazepine and insignificant risk to metoprolol and propranolol. In the northeast region, high risk was calculated to acetaminophen (algae and Daphnia), caffeine (Daphnia and fish), diclofenac (fish) and ibuprofen (Daphnia). Sulfamethoxazole showed moderate risk for algae, while carbamazepine, mebendazole and methylparaben, low risk for the three trophic levels. In the midwest region, high risk was found only to caffeine (Daphnia and fish), while moderate risk was detected to 17α-ethynylestradiol (Daphnia). Triclosan present low risk and bisfenol A, insignificant risk.

Similarly to the RQs found in Brazil, low or medium RQ in surface waters was found for atenolol and naproxen in Pakistan (Ashfaq et al. 2019), carbamazepine, metoprolol, norfloxacin and trimethoprim in China (Zheng et al. 2020), ciprofloxacin in India (Singh and Suthar 2021), mebendazole in China (Chen et al. 2021) and India (Singh and Suthar 2021), propranolol in Sri Lanka (Guruge et al. 2019) and trimethoprim in Vietnam (Ngo et al. 2020).

Related to the four parabens investigated (ethylparaben, methylparaben, propylparaben and butylparaben), only methylparaben have shown high risk in Brazil. In the surface water of the Yangtze River in China, the same preservatives were investigated and minimal or low risk was found (Liu et al. 2015).

17ß-estradiol and 17α-ethynylestradiol showed high risk in this study. In surface water samples obtained from sampling points in Mexico, the RQs of 17ß-estradiol ranged from 0.07 to 0.94 and of 17α-ethynylestradiol from 0.21 to 2.56, presenting medium to high risk (Calderón-Moreno et al. 2019). In China, medium to high risk was also found for these compounds (Zhong et al. 2021). Estrone showed medium to high risk in this study. Similar results were found in surface waters from China (Zhong et al. 2021).

Acetaminophen showed high RQ in Brazil and in China (Zheng et al. 2020) but showed medium RQ in Pakistan (Ashfaq et al. 2019) and low RQ in India (Singh and Suthar 2021). Ibuprofen has been shown to have different RQ according to the studies. Similarly, to this study, high RQ was found in Pakistan (Ashfaq et al. 2019) and Sri Lanka (Guruge et al. 2019), but in China and India, low RQ was found (Zheng et al. 2020; Singh and Suthar 2021). Diclofenac showed high RQ in Brazil and China (Zheng et al. 2020) and low to medium in Vietnam, Pakistan and Sri Lanka (Ashfaq et al. 2019; Guruge et al. 2019; Ngo et al. 2020).

Similarly to the high RQ found in this study, high risk was also found for bisphenol A in surface waters from Mexico (Calderón-Moreno et al. 2019), and for caffeine in China (Zheng et al. 2020) and Pakistan (Ashfaq et al. 2019). Dibuthylphthalate and diethylphthalate were detected in Brazilian surface waters in concentrations that represent high risk (RQ > 1.0). In a study carried out in Uganda, high RQ was found for dibuthylphthalate and low risk for diethylphthalate (Nantaba et al. 2021). Benzophenone-3 showed high risk in Brazil, but in surface waters from Romania, low risk was found (Chiriac et al. 2021) and medium risk in surface waters from Shanghai, China (Wu et al. 2017).

Gemfibrozil presented high RQ in this study and low in Pakistan (Ashfaq et al. 2019) and Sri Lanka (Guruge et al. 2019) surface waters. Sertraline had shown high RQ in some studies in Brazilian waters, and in Turkey surface waters low RQ was calculated (Guzel et al. 2019). For sulfamethoxazole, low to medium RQ was found in this study and in China (Zheng et al. 2020) and Pakistan (Ashfaq et al. 2019), but in surface waters from Greece (Nannou et al. 2015), Vietnam (Ngo et al. 2020) and Sri Lanka (Guruge et al. 2019) high RQ was found.

Triclosan showed high RQ in this study, and in Greece (Nannou et al. 2015), Indian (Singh and Suthar 2021) and Uganda (Nantaba et al. 2021) surface waters; however, in China (Zheng et al. 2020) and Sri Lanka (Guruge et al. 2019), low or medium RQ was found.

This estimation of RQs was made for each compound separately, but it must be taken into consideration the fact that in the aquatic environment PPCPs never occur individually and the mixture of various pharmaceuticals may lead to different toxicity risks on aquatic organisms. However, the estimations made in this study are based on the RQ of a single pharmaceutical (Kosma et al. 2014; Nannou et al. 2015).

How could be observed, the risk assessment for PPCPs can vary around the world, and among regions from the same country, since it depends on the quantified concentrations, which vary according to the PPCP usage, sanitary conditions, removal efficiency in wastewater treatment plants and environmental conditions.

Prioritization of CECs in Brazilian aquatic environments

In Brazil, there is no national environmental legislation to regulate CECs in the environment or even in drinking water. The monitoring initiatives come from academics and environmental agencies, such as the São Paulo State Environmental Company (CETESB) (Aragão et al. 2020). Because of this, it is important to create a list of priority contaminants in the country, which in addition to taking into account data on the consumption of pharmaceutical and personal care products, occurrence data and the environmental risks involved must be considered. Norman Network and the European Watch-List are two important examples of prioritization of CECs in the environment. Norman Network seeks to promote the exchange of information on emerging environmental substances from different countries (Norman 2019), while the European Watch-List is used by European Union with the goal of obtaining monitoring data for pollutants for which available data to assess their risks are still insufficient to allow conclusions on their effects (Eur, E. C 2013).

Based on occurrence data and environmental risk, it was possible to highlight the most concerning CECs in Brazilian aquatic environment. CECs with moderate or high ecological risk were classified as environmentally hazardous compounds and therefore should receive more attention and further studies on occurrence, persistence and toxicity should be performed. About 25% of the reviewed CECs are antibiotics and other 21% are anti-inflammatories drugs, with emphasis on diclofenac, acetaminophen and ibuprofen that presented frequency of detection of 41%, 32% and 24%, respectively, presenting high risk to aquatic biota. These compounds may result in negative effects on wild bivalves after long-term exposures and even on organisms from higher trophic level due to food-chain transfer (Almeida et al. 2020). Therefore, considering the risk associated with the occurrence of these compounds in Brazilian aquatic ecosystems, they can be considered of concern and included in the list of priority contaminants.

On the antibiotics group, only sulfamethoxazole presented moderate risk for algae. This is one of the most consumed antibiotics by the Brazilian population and, similar to caffeine, has been considered a marker of anthropogenic contamination being ubiquitous in water bodies worldwide (Thiebault 2020).

Caffeine was the most detected chemical in the Brazilian surface water presenting low to high environmental risk. Moreover, it is an anthropogenic marker, helping to identify places where effluents are discarded without treatment and that probably are contaminated with other CECs present in wastewater. Therefore, this compound must be also considered a priority in environmental monitoring studies.

In addition these, 17α-ethynylestradiol, 17ß estradiol, bisphenol A, methylparaben and triclosan presented high frequencies of detection and concentration that caused high risk to algae, crustaceans and fish. These CECs are reported in several environmental matrices worldwide and have the potential for endocrine disruption.

Therefore, taking into account the occurrence data and the estimated environmental risks, Figure 3 shows the CECs of most concern in each Brazilian region, highlighting that risk quotient takes into account the maximum environmental concentration found, being an indicative of the environmental risk. Considering a general ranking, the most concerning CECs in Brazil are diclofenac, acetaminophen, caffeine, methylparaben, sulfamethoxazole, bisphenol A, ibuprofen, 17α-ethynylestradiol, 17ß-estradiol and triclosan. Monitoring studies for these compounds should be carried out in order to generate data to assist decision-making and the creation of environmental legislation in the future.

Fig. 3
figure 3

CECs of greatest concern in Brazilian surface waters, considering occurrence and ecological risk. In parentheses is presented the maximum risk quotient found for each chemical

Full size image

In the European Union, they have a surface water Watch List (WL) under the Water Framework Directive (WFD) which is a mechanism for obtaining data on potential water pollutants for the purpose of determining the risk they pose. This list is updated every 2 years. From the compounds ranked to be studied in Brazil as a priority, diclofenac was included in the first WL (2015), 17α-ethynylestradiol and 17ß-estradiol in the first (2015) and second (2018) WL and sulfamethoxazole was recently added in the third WL (2020). These substances are considered in these lists because they are considered substances that may pose a significant risk, at Union level, to or via the aquatic environment, but for which monitoring data are insufficient to come to a conclusion on the actual risk posed (Eur 2020).

In this study, it was possible to notice that the prioritizing pharmaceuticals in Brazilian urban rivers impacted by domestic effluents is directly related to medicine consumption. From twenty most commercialized pharmaceutical substances in Brazil, nine (45%) were among the most detected compounds in Brazilian surface waters. Therefore, the relationship between regional usage of pharmaceuticals and levels quantified in different regions presents a challenge to prioritization in a country with several socioeconomic characteristics, like Brazil.

Uncertainties and limitations

This study was carried out based on measured concentrations and ecotoxicity data from literature. For some chemicals, that ecotoxicity data are not available (benzoylecgonine, citalopram, cocaine, nimesulide, among others), it was not possible to carry out risk analysis. Therefore, results of priority ranking of CECs in Brazil may change or remain unchanged if there are other updated measured and ecotoxicity data.

Compounds detected with low frequency (<10%), as amitriptyline, azithromycin, avobenzone, benzylparaben, erythromycin, glibenclamide, losartan, valsartan, among others, were not taken into consideration in the risk assessment, but they also may present risk to the aquatic biota. In addition, few studies were found in the north (Thomas et al. 2014), northeast (Chaves et al. 2020) and midwest (Sposito et al. 2018), and some chemicals (e.g. 17ß-estradiol, 17α-ethynylestradiol, estrone, naproxen and sertraline) were reported in more than one sampling locations, but from the same region, being insufficient to represent the complete ranking of CECs in the country.

Conclusions

Monitoring data of CECs in Brazil is still limited; most studies had occurred mainly in the south and southeast regions. However, based on current knowledge, it was possible to build a priority ranking of CECs by frequency of occurrence and environmental risk levels. The reviewed CECs were mainly distributed in highly urbanized and industrialized regions, including the São Paulo State. The water bodies from the southeast region are the most studied, consequently have a greater amount of different chemicals reported and seemed to be the hot-spot region where several chemicals were ranked with high risk. Levels of CECs in Brazilian aquatic environment were usually higher than in other emerging countries (China, India, South Africa and Pakistan) and lower than South American countries.

Analyzing the environmental risks caused by CECs in Brazilian surface water, it was observed that pharmaceuticals presented higher risks to aquatic organisms compared to PCPs. Crustaceans (Daphnia) were the most sensitive organism for risk assessment of CECs. Ranked CECs with the highest risk were anti-inflammatories drugs (acetaminophen, diclofenac and ibuprofen), caffeine, steroid estrogens (17ß-estradiol and 17α-ethynylestradiol) and preservative (methylparaben). These results indicate a great threat to Brazilian aquatic ecosystem. Therefore, risk management actions must be carried out to promote safety to aquatic biota and protection of the environment.