Source to sink tracking of selected human pharmaceuticals from two Oslo city hospitals and a wastewater treatment works

Abstract

The occurrence of twenty pharmaceutical compounds was quantitatively determined in effluents from two major Oslo city hospitals, Rikshospitalet and Ullevål, along with influent, sludge and final effluent from the city's VEAS wastewater treatment works (WTW). Composite hospital effluents were collected over a twelve week period and were showed to contain paracetamol, metoprolol, diclofenac, ibuprofen, 17beta-Estradiol, estriol, estrone, oxytetracycline, tetracycline, doxycycline, chlorotetracycline, demeclocycline, trimethoprim, ciprofloxacin, sulfamethoxazole, cyclophosphamide and ifosfamide. Three pharmaceuticals were not detected above the limit of detection; cefuroxime, 17alpha-ethinylestradiol and meclocycline. Composite influent, sludge and effluent samples were collected from VEAS WTW over a seven week period. The influent into VEAS WTW contained all of the same selected substances detected in the hospital effluents, except for oxytetracycline, chlorotetracycline, demeclocycline, cyclophosphamide and ifosfamide. The percentage of pharmaceuticals entering the works from the hospitals was <10% for all of the selected compounds. VEAS sludge samples contained a different profile of substances reflecting their physico-chemical properties. Hydrophobic antibiotics, such as oxytetracycline, tetracycline and ciprofloxacin, were detected in all of the collected sludge samples. Their absence in the collected influent samples suggests that they enter the works bound to effluent particles, with the dissolved fraction observed in the hospital effluents partitioning onto particulate matter within the sewerage network. The final effluent from VEAS WTW contained reduced concentrations of many pharmaceuticals, including paracetamol, ibuprofen and sulfamethoxazole. For other compounds, such as metoprolol, diclofenac and trimethoprim, there were often higher concentrations in the effluent than the influent. These effluent concentrations represent median inputs varying from low g day(-1) (e.g. paracetamol and ibuprofen) to nearly 200 g day(-1) (e.g. metoprolol and trimethoprim) into Oslofjord. A simple risk assessment showed that the antibiotic ciprofloxacin may at times pose an acute risk to the Oslofjord aquatic environment.

Contribution of hospital effluents to the load of pharmaceuticals in urban wastewaters: Identification of ecologically relevant pharmaceuticals

“…Several studies have reported the presence of antibiotics in hospital effluents, being the fluoroquinolones among the most detected. For instance, the measured concentrations of ciprofloxacin are in agreement with previous findings reported in literature to hospital effluents in Norway (up to 23,336 ng L − 1 ) (Thomas et al, 2007), Switzerland (31,980 ng L − 1 ) (Kovalova et al, 2012) and Italy (1400-26,000 ng L − 1 ) (Verlicchi et al, 2012a), while Duong et al (2008) found lower concentrations of ciprofloxacin (1100-10,900 ng L − 1 ) in Taiwan. On the other hand, higher levels were detected in a university hospital in Germany (up to 51,000 ng L − 1 ) (Ohlsen et al, 2003) and in Sweden (3600-101,000 ng L − 1 ) (Lindberg et al, 2004).…”

“…Comparatively to previous findings, Thomas et al (2007) reported higher levels of acetaminophen (329,852 ng L − 1 ) in the effluents of hospitals from Oslo (Norway), while ibuprofen did not exceed 8957 ng L − 1 .On the other hand, lower concentrations of these pharmaceuticals were detected in the effluents of two Italian hospitals, where acetaminophen levels went from 1400 to 5900 ng L − 1 and ibuprofen from 380 to 3200 ng L − 1 (Verlicchi et al, 2012a). Opposite to this trend, Sim et al (2011) never detected acetaminophen or ibuprofen in effluents from four general hospitals in Korea.…”

“…However, for ofloxacin higher levels were detected in Italy (3300-37,000 ng L − 1 ) (Verlicchi et al, 2012a), USA (up to 35,500 ng L − 1 ) (Brown et al, 2006) and Germany (up to 31,000 ng L − 1 ) (Ohlsen et al, 2003), but not in Sweden (200-7600 ng L − 1 ) (Lindberg et al, 2004). Relatively to sulfamethoxazole, the concentrations found reaching up to 8714 ng L − 1 , which were, in general, higher than data reported in literature (Brown et al, 2006;Kovalova et al, 2012;Ohlsen et al, 2003;Thomas et al, 2007;Verlicchi et al, 2012a). Nevertheless, levels up to 12,800 ng L − 1 were detected in hospital effluents in Sweden (Lindberg et al, 2004).…”

“…In fact, more pronounced contributions were described in literature for antibiotics (Beier et al, 2011;Ort et al, 2010;Thomas et al, 2007;Verlicchi et al, 2012a), reaching, in some cases, contributions as high as 272% (ciprofloxacin) (Thomas et al, 2007), 94% (clarithromycin) (Beier et al, 2011) or 67% (azithromycin) (Verlicchi et al, 2012a). On the other hand, for some of the most consumed analgesics/NSAIDs (for instance, ibuprofen, diclofenac or acetaminophen) hospital contribution reported in literature did not exceed 15% (Beier et al, 2011;Langford and Thomas, 2009;Thomas et al, 2007;Verlicchi et al, 2012a), which is in agreement with our results in what concern to ibuprofen and diclofenac (contribution up to 4.2 and 9.5%, respectively) (data not shown), however for acetaminophen, the contribution of university and general hospitals went to 483 and 115%, respectively (data not shown). Nevertheless, the X-ray contrast agent iopromide, which had a mean total mass load of approximately 303 g d − 1 coming from hospital effluents, only contributed with approximately 13% of its total mass load found in WWTP influent, though Ort et al (2010) reported a minor contribution (less than 5%).…”

“…Nevertheless, few data is available on the contribution of hospital effluents towards the load of pharmaceuticals in WWTPs (Beier et al, 2011;Langford and Thomas, 2009;Ort et al, 2010;Thomas et al, 2007;Verlicchi et al, 2012a). At the same time, available data regarding the environmental risk posed by hospital effluents to aquatic organisms is still sparse and often limited to predicted (Escher et al, 2011;Souza et al, 2009) rather than measured concentrations (Verlicchi et al, 2012a).…”