A total of 6196 records was recovered and, after duplicate removal, 4413 records were screened. 4242 citations were eliminated in the first-level screening based on their title and abstract. 171 reports were sought for retrieval, 13 reports were not found, thus 158 reports were assessed for eligibility by full text during the second-level screening. 119 studies matching with inclusion criteria were included, 5 studies were included by citation searching, resulting in 124 studies finally included in this systematic review. The screening process is presented in the flow chart showed in Fig. 1.
The results of this review are presented in tables 1, 2, 3, and 4 according to their geographical area.
We identified a total of 124 studies searching for pharmaceutical residues in drinking water, 91 reporting positivity for one or more compounds, in concentrations range from a few to a few tens of nanograms. In the remaining 33 studies, target analytes were below the limit of detection. We tried to understand how widespread the phenomenon of contamination by medicines for human and veterinary use is, so we reported in which countries pharmaceutical residues have been traced or not in drinking water, and how many studies have been carried out in each country to understand the extent of research on the issue (Fig. 2).
Of the 124 studies included, 64 publications were from Europe (44 reported positivity for pharmaceutical residues), 33 from Asia (25 with positivity for pharmaceutical residues), 22 from Americas (19 with positivity for pharmaceutical residues), finally 3 from Africa and 2 from Oceania (2 and 1 reports that detected pharmaceutical residues, respectively). Spain was the country with more papers (20), China was the second country with 15 papers followed by Germany and Brazil with 12 and 8 papers, respectively. The most detected pharmaceutical categories overall and in Europe, Asia, and rest of the world are shown in Fig. 3. We also reported the most detected pharmaceutical molecules all over the world, Europe, Asia, and rest of the world in Fig. 4.
The most represented category considering all included papers was NSAIDs and analgesics, detected in 39 studies, followed by anticonvulsants and antibiotics (respectively detected in 34 and 33 studies). These categories are also the most detected dividing the results by geographical area, but in Asia the most detected pharmaceuticals are antibiotics, detected in 15 works from Asian studies of 33 total studies positive for antibiotics, in contrast with Europe where the first two detected categories are anticonvulsants and NSAIDs or analgesics, with 20 and 19 detections respectively, followed by antibiotics with 11 detections. Beta blockers were detected in 17 studies, all from Asia and Europe, no beta-blockers were detected in selected studies from Americas, Africa, and Oceania.
The most represented active principle considering total studies was Carbamazepine, detected in 24 studies. The other relevant active principles were Diclofenac (15 studies), Ibuprofen, Paracetamol and Sulfamethoxazole (13 studies).
The most used sample preparation technique to concentrate the sample and remove unwanted interferences is solid phase extraction (SPE). The most common analytical techniques used to find pharmaceutical residues in drinking water in the selected studies are liquid chromatography, high-performance liquid chromatography, ultra-high-performance liquid chromatography, tandem mass spectrometry (LC-MS/MS, HPLC-MS/MS, UHPLC-MS/MS), and gas chromatography mass spectrometry (GC-MS). When determining trace amounts of such pollutants, it is crucial to use an analytical method that may provide both selectivity and sensitivity.
The results of this review show agreement between different geographic areas and principal pharmaceuticals detected in drinking water. The possibility to detect a pharmaceutical molecule in drinking water depends not only on the amount of its clinical use and sold, but its presence in water is determined also by specific chemical properties that determine environmental persistence and from the efficacy of the removal from wastewater avoiding contamination of environmental waters, source for production of drinking water.
Active ingredients and their metabolites are frequently found in wastewater and environmental waters (Baron et al. 2014; Białk-Bielińska, 2016; Fekadu et al. 2019; Herrero-Villar et al. 2021; MacKeown et al. 2022). Studies have been conducted to determine the effective concentration of drugs in wastewater, both before and after the treatment carried out by STPs and in surface waters. NSAIDs, due to their widespread use, represent one of the most investigated categories of drugs. Several studies on acetylsalicylic acid and paracetamol have demonstrated the effectiveness of purification systems (Ternes 2001). Another drug belonging to the class of NSAIDs which persists in water even after purification is ibuprofen, traces of which have been found in water purification plants in Austria, Brazil, Germany, and Switzerland (Buser et al. 1999; Heberer 2002; Ternes 1998; Ternes et al. 2002), and it was calculated that more than 60 kg of diclofenac was discharged into the Baltic Sea in 2018, while in the first half of 2021 it was equal to 20 Kg from the two WWTPs tested. The presence of 4OH-diclofenac in effluents often in higher concentration compared to diclofenac mean that this still biologically active compound needs to be considered in future risk assessment (Kołecka et al. 2022). Traces of naproxen have been found in STP effluents in Canada (Metcalfe et al. 2003) and Spain (Quintana et al. 2005). Among the neuroactive drugs, carbamazepine is the most studied active principle as it is removed only in very small part (less than 10%) from the STP (Ternes 2001; Heberer 2002); the result is frequently detection of positive samples for this active principle. Another class of drugs that has been particularly studied is lipid-lowering drugs; in particular, clofibric acid has historical importance as the first pharmacologically active molecule to be found in water samples from municipal water purification plants.
The effects of long-term exposure to pharmaceutical residues through drinking water are difficult to assess. No long-term toxicity studies on humans are currently available, relating to doses much lower than therapeutic ones; however, the alterations observed in wildlife suggest potential effects on human health (Shore et al. 2014). Many of these molecules can act as endocrine disruptors: their structure could mimic that of other substances involved in the fine balance of the neuroendocrine system, hormones can act at very low doses and alter the normal physiology of the organism (Chen et al. 2019).
Furthermore, since the effects that active ingredients and metabolites in water can have on the environment and what danger they can represent for human health, greater use of drugs that is currently implemented in the veterinary field, both for curative and productive purposes especially in intensive farming, should not be underestimated. This sector is undoubtedly responsible for a large amount of environmental and water contamination that is difficult to control, and it can aggravate this situation.
Wastewater from intensive farms, generally undergoes less effective purification treatments than civil plants. To these residues must be added those of all the drugs widely used, illegally, on farm animals for auxinic, masking purposes, etc.
The possibility that pharmaceutical residues can cause direct damage to the environment and indirect damage to humans led US Food and Drug Administration and the European Agency for the Evaluation of Medicinal Products - EMEA), to include environmental risk assessment as an essential requirement in each dossier presented to put new drugs on the market, as provided in the guidelines issued by both agencies, FDA and EMEA, in 1997 and 2001 respectively, updated in 1998 and 2006 (FDA 1998; EMEA 2006b). Both documents initially require a careful assessment of the predicted environmental distribution of the active ingredient and its main metabolites. In practice, through mathematical formulas that consider the quantity of product sold, the maximum therapeutic dose foreseen per patient/day and daily produced wastewater per inhabitant, it is possible to estimate predicted concentrations.
For veterinary medicinal products also, the EMEA has issued a series of guidelines (EMEA, 2000; 2005; 2006a) which make it necessary to evaluate contamination and environmental persistence for the purposes of any marketing authorisation of veterinary medicines.