ARTICLE IN PRESS
WAT E R R E S E A R C H
42 (2008) 395 – 403
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Removal of antibiotics from wastewater by sewage
treatment facilities in Hong Kong and Shenzhen, China
A. Gulkowskaa, H.W. Leunga, M.K. Soa, S. Taniyasub, N. Yamashitab, Leo W.Y. Yeunga,
Bruce J. Richardsona, A.P. Leic, J.P. Giesya,d,e, Paul K.S. Lama,
a
Centre for Coastal Pollution and Conservation, Department of Biology and Chemistry, City University of Hong Kong,
Tat Chee Avenue, Kowloon, Hong Kong SAR, People’s Republic of China
b
National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba, Ibaraki 305-859, Japan
c
College of Life Sciences, Shenzhen University, Shenzhen 518060, People’s Republic of China
d
Department of Veterinary Biomedical Sciences and Toxicology Centre, University of Saskatchewan, Canada S7K 3J8
e
National Food Safety and Toxicology Center, Zoology Department and Center for Integrative Toxicology, Michigan State University,
East Lansing, MI 48824, USA
art i cle info
ab st rac t
Article history:
Concentrations of nine antibiotics [erythromycin-H2O (ERY-H2O); trimethoprim (TMP);
Received 13 May 2007
tetracycline (TET); norfloxacin (NOR); penicillin G (PEN G); penicillin V (PEN V); cefalexin
Received in revised form
(CLX); cefotaxim (CTX); and cefazolin (CFZ)] were measured in influent and effluent
17 July 2007
samples from four sewage treatment plants (STPs) in Hong Kong as well as in influent
Accepted 19 July 2007
samples from one STP in Shenzhen. Levels of PEN V and CFZ were below method detection
Available online 27 July 2007
limits in all of the samples analyzed. CLX concentrations were the highest in most of the
Keywords:
Antibiotics
STPs
Influent
Effluent
Removal efficiency
Hong Kong
China
Hong Kong samples, ranging from 670 to 2900 ng/L and 240 to 1800 ng/L in influent and
effluent samples, respectively, but CLX was not detected in the samples from Shenzhen.
Comparatively lower concentrations were observed for ERY-H2O (470–810 ng/L) and TET
(96–1300 ng/L) in the influent samples from all STPs in Hong Kong. CTX was found to be the
dominant antibiotic in the Shenzhen STP influents with a mean concentration of 1100 ng/L,
but occurred at lower concentrations in Hong Kong sewage. These results likely reflect
regional variations in the prescription and use patterns of antibiotics between Hong Kong
and Shenzhen. Antibiotic removal efficiencies depended on their chemical properties and
the wastewater treatment processes used. In general, relatively higher removal efficiencies
were observed for NOR (5–78%) and TET (7–73%), which are readily adsorbed to particulate
matter, while lower removal efficiencies were observed for ERY-H2O (9–19%), which is
relatively persistent in the environment. Antibiotics were removed more efficiently at Hong
Kong STPs employing secondary treatment processes compared with those using primary
treatment only. Concentrations of NOR measured in effluents from STPs in Hong Kong were
lower than the predicted no-effect concentration of 8000 ng/L determined in a previous
study. Therefore, concentrations of antibiotics measured in this preliminary study would
be unlikely to cause adverse effects on microorganisms used in wastewater treatment
processes at the sampled STPs.
& 2007 Elsevier Ltd. All rights reserved.
Corresponding author. Tel.: +852 2788 7681; fax: +852 2788 7406.
E-mail address: bhpksl@cityu.edu.hk (P.K.S. Lam).
0043-1354/$ - see front matter & 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2007.07.031
ARTICLE IN PRESS
396
1.
WA T E R R E S E A R C H
Introduction
Pharmaceuticals and related products have become chemicals of emerging environmental concern in recent years
(Richardson et al., 2005). Pharmaceuticals have been used as
human medicine to treat or prevent microbial infections, and
as veterinary drugs and husbandry growth promoters in
aquaculture and livestock operations (Halling-Sørensen et al.,
1998). These contaminants, including their precursor compounds and transformation products, are discharged into the
environment intentionally and unintentionally during manufacturing processes and through consumption or disposal of
used and unwanted drugs (Daughton and Ternes, 1999).
Pharmaceuticals have been detected in groundwater (Lindsey
et al., 2001), surface water (Xu et al., 2007; Gulkowska et al.,
2007), streams (Kolpin et al., 2002), as well as in sludge, soil
and sediment samples (Dı́az-Cruz and Barceló, 2005; Göbel
et al., 2005; Kim and Carlson, 2007). Pharmaceuticals have
also been frequently detected in sewage treatment plant (STP)
effluents (Daughton and Ternes, 1999; Ashton et al., 2004),
since STPs were not designed to remove them (Xu et al., 2007).
For example, trimethoprim (TMP) was found to be only
partially removed by STPs (Hernando et al., 2006). Furthermore, unusual wet weather runoff that exceeded STP treatment capacity can result in larger volume discharge or direct
introduction of pharmaceuticals into the environment
(Daughton and Ternes, 1999).
Antibiotics are regarded as ‘‘pseudopersistent’’ contaminants due to their continual introduction into the ecosystem
(Richardson et al., 2005; Hernando et al., 2006). The occurrence of antibiotics in the environment has therefore received
considerable attention. It has been demonstrated that antibiotics are, in general, poorly absorbed by the human body,
and thus are excreted either unchanged or transformed, via
urine and feces (McArdell et al., 2003). Several studies have
shown a relationship between local sales of human pharmaceuticals and their concentrations in STP influents (HallingSørensen et al., 1998; Göbel et al., 2005), meaning that
wastewater from STPs could potentially be significant point
sources of human antibiotics in the aquatic environment.
Another concern about antibiotic residues in the environment is their potential adverse effects to various organisms.
For example, bacteria isolated from sewage bioreactors have
been shown to exhibit resistance to some antibiotics including TMP, erythromycin (ERY), tetracycline (TET), ciprofloxacin
and ampicillin (Costanzo et al., 2005). Veterinary antibiotics
have been shown to cause oxidative damage in liver cells of
rainbow trout (Oncorhynchus mykiss) (Gagńe et al., 2006), and
were lethal to brine shrimp (Artemia) (Migliore et al., 1997).
Since antibiotics are specifically designed to be biologically
active, unintended chronic exposure to antibiotics could
cause adverse effects at lower concentrations than other
chemicals. In addition, different antibiotics possessing a
common mechanism of action could have significant additive
or synergistic adverse effects, despite relatively low environmental concentrations of the individual chemicals (Daughton
and Ternes, 1999).
In a previous study, antibiotics, including TET, TMP,
erythromycin-H2O (ERY-H2O), norfloxacin (NOR) and cefalexin
42 (2008) 395– 403
(CLX), were detected in Hong Kong coastal waters (Gulkowska
et al., 2007). The higher concentrations of antibiotics near STP
outfalls suggested sewage discharges as potential sources. To
date, there is no information on antibiotic concentrations in
either STP influents or effluents or on removal efficiencies of
antibiotics from STPs in Hong Kong or southern China. The
objectives of the present study were therefore to determine
concentrations of selected antibiotics [ERY-H2O, TMP, TET,
NOR, penicillin G (PEN G), penicillin V (PEN V), CLX, cefotaxim
(CTX) and cefazolin (CFZ)] in both influents and effluents of
four Hong Kong STPs, and in influents of one STP in Shenzhen
(a city in Guangdong Province, China, adjacent to Hong Kong).
These antibiotics were selected based on anecdotal evidence
of the most frequently used pharmaceutical compounds in
China (Richardson et al., 2005). In addition, the removal
efficiencies of individual antibiotics in the Hong Kong STPs
were determined.
2.
Materials and methods
2.1.
Chemicals and standards
ERY and TMP were purchased from Sigma-Aldrich (St. Louis,
MO, USA). TET hydrochloride, NOR, CLX, and potassium salts
of PEN G and PEN V were obtained from Riedel-de Haën
(Seelze, Germany). CTX and CFZ sodium salts were purchased
from Fluka (Buchs, Switzerland). Oasiss hydrophilic–lipophilic balanced (HLB; 6 cc, 200 mg) solid-phase extraction (SPE)
cartridges were purchased from Waters (Milford, MA). Milli-Q
water was used throughout the study. HPLC-grade methanol
and acetonitrile, formic acid (99%) and disodium ethylenediamine tetraacetate (Na2EDTA) were purchased from Wako
Pure Chemical Industries Ltd. (Japan). Ammonium hydroxide
(NH4OH) was obtained from Riedel-de Haën (Seelze, Germany).
Stock solutions of individual antibiotics (100 mg/L) were
prepared following the procedures reported previously (Gulkowska et al., 2007). Briefly, two groups of standards were
prepared. Group I chemicals (ERY, TET, NOR and TMP) were
dissolved in methanol at pH 3, whereas Group II chemicals
(PEN V, PEN G, CLX, CTX and CFZ) were dissolved in Milli-Q
water. Mixed working solutions (1, 5, 10, 20, 50 and 100 mg/L)
were freshly prepared prior to extraction.
2.2.
Sample collection and preparation
Samples of STP influents and effluents were collected
simultaneously from four Hong Kong STPs in Wan Chai,
Shatin, Tai Po and Stonecutters Island in December 2006,
while influent samples were collected from Nan Shan STP in
Shenzhen in January 2007. The types of treatment employed
and the treatment capacities for each STP are given in Table 1.
Two-liter grab samples were collected from each location by
use of a stainless-steel bucket, which was pre-cleaned by
rinsing with Milli-Q water, methanol and then water from the
specific location. Samples were stored in 1-L polypropylene
bottles, kept in coolers and transported to the laboratory. The
samples were vacuum-filtered through glass fiber filters
(4.5 mm, Advantec, Toyo Roshi Kaisha Ltd., Japan) immediately
Table 1 – Sewage treatment system and treatment capacity of individual STPs (information from the Drainage Services Department, Hong Kong)
STP
Organic
matter
removal
Nutrient
removal
Sources of the
catchment area
Treatment process
About
50%
removal
in BOD5
90%
removal
in BOD5
Not
applicable
Main areas of Kowloon
and northeastern Hong
Kong Island
50%
removal
in total
nitrogen
Shatin new town, Ma On
Shan new town and
surrounding villages
(Eastern New Territories)
Chemically enhanced primary treatment process in
which chemicals, including ferric iron and polymers, are
used to treat the sewage. Treated sewage is discharged via
an outfall in the western part of Victoria Harbor
Secondary treatment processes including (i) screening of
coarse material; (ii) settlement of grit particles; (iii)
primary sedimentation of suspended matter; and (iv)
biological treatment of sewage
Population
Dry
weather
flow
(m3/day)
3.5 million
by the year
2021
1725
million by
the year
2021
150,000 at
present and
830,000 by
the year
2010
94,300 at
present
1–2
n.a.
21
20
16
20
90%
removal
in BOD5
50%
removal
in total
nitrogen
Tai Po new town and the
Tai Po industrial estate
(Eastern New Territories)
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
Shatin
369,512 at
present
Tai Po
376,077 at
present
Wan Chai
East
Shenzhen
Nan Shan
136,000 in
2004–2005
n.a.
Secondary treatment processes including (i) removal of
coarse solids; (ii) removal of grit; (iii) primary
sedimentation; (iv) biological treatment for primary
sewage; and (v) anaerobic digestion for surplus activated
sludge
Primary treatment screening plant that screens
suspended matter with diameter 46 mm
Primary treatment screening plant
n.a. represents not available.
ARTICLE IN PRESS
Solid
retention
time
(day)
42 (2008) 395 – 403
Hydraulic
retention
time (h)
WAT E R R E S E A R C H
Stonecutters
Island
Treatment capacity
397
ARTICLE IN PRESS
398
42 (2008) 395– 403
WA T E R R E S E A R C H
2.1 mm i.d. 10 mm length). The gradient program of mobile
phases for HPLC and the MS/MS parameters for the instrument were described in detail in our previous study (Gulkowska et al., 2007).
after collection to remove particulate matter prior to extraction. The filtered samples were then stored in the dark at 4 1C,
and were extracted within 48 h of collection.
Samples were pre-treated following the methods described
by Gulkowska et al. (2007). Na2EDTA was added to the samples
as a chelating agent. The two groups of antibiotics were
extracted at different pH values (pH 3 and pH 7.5–8.0 for
Group I and II antibiotics, respectively).
2.3.
2.5.
Each ion of interest in the chromatogram was selected and
integrated. External calibration curves of six concentrations
(1, 5, 10, 20, 50 and 100 mg/L) were constructed for the
quantification of target analytes in the samples. Standard
calibration curves exhibited excellent linearity (correlation
coefficient 40.999). Final extracts with concentrations that
fell outside the ranges of the calibration curves were diluted
with methanol to an appropriate factor and re-injected.
Sample extraction
Influent and effluent samples were extracted using Oasis HLB
extraction cartridges. Each cartridge was pre-conditioned
prior to loading the water sample following the procedures
described in our previous study (Gulkowska et al., 2007). An
aliquant of sample water (200 mL) was then loaded onto the
cartridge, eluted at a rate of 1 drop/s, and the eluate was
discarded. The cartridge was then washed with 4 mL water to
remove excess Na2EDTA. Finally, the target fraction was
eluted with 4 mL acetonitrile or methanol for Group I and II
chemicals, respectively. The eluate was reduced to a volume
of 0.5 mL under a gentle stream of nitrogen and the final
volume was adjusted to 2 mL with water, thoroughly mixed
and then transferred into an amber autosampler vial for
instrumental analysis.
2.4.
Data quantification
3.
Results and discussion
3.1.
Quality assurance and quality control
Procedures for the determination of the accuracy of the
sampling and extraction methods, as well as the method
detection limits (MDLs) applied in this study were the same as
in our previous study except that the sample volume was
reduced to 200 mL (Gulkowska et al., 2007). The procedural
recoveries for influent and effluent samples were 76–100%
and 89–100%, respectively (Table 2), indicating the applicability of the current extraction method to both influent and
effluent samples. Concentrations reported in the present
study were not corrected for recoveries. The MDLs for effluent
samples ranged from 4.0 to 37 ng/L. Higher MDLs were
determined for influent samples (4.0–93 ng/L), possibly due
to the existence of matrix effects in the raw sewage.
Concentrations of all target analytes in field and procedural
blanks were lower than the MDL for each analyte. Positive
controls were also analyzed in the present study to check for
possible degradation, loss or contamination of target analytes
during the sampling and transportation. Two hundred
Instrumental analysis
Concentrations of antibiotics in STP influents and effluents
were analyzed using high-performance liquid chromatography interfaced with tandem mass spectrometry (HPLC–MS/
MS). Separation of analytes was performed by an Agilent
HP1100 liquid chromatograph (Agilent, Palo Alto, CA) interfaced with an Applied Biosystems API 2000 triple quadrupole
tandem mass spectrometer equipped with a Turbo IonSpray
source operated in both negative and positive modes. A 10 mL
aliquot of extract was injected onto a XBridgeTM C18 column
(Waters Corp., 2.1 mm i.d. 50 mm length, 5 mm) equipped
with a guard column (Waters Corp., XBridgeTM C18, 5 mm,
Table 2 – Quality control and quality assurance parameters for individual antibiotics
Concentration (ng/L)
n
ERYH2O
TMP
TET
NOR
PEN G
PEN V
CLX
CTX
CFZ
Procedural blank
Field blank
MDL for influent samples
MDL for effluent samples
MDL for Milli-Q water
10
5
–
–
–
oMDL
oMDL
47
12
3.0
oMDL
oMDL
24
6.0
2.0
oMDL
oMDL
70
14
5.0
oMDL
oMDL
90
16
18
oMDL
oMDL
4.0
4.0
8.0
oMDL
oMDL
5.0
5.0
22
oMDL
oMDL
93
37
12
oMDL
oMDL
15
12
22
oMDL
oMDL
11
14
5.0
Procedural recovery
Matrix-spike recovery (influent)
Matrix-spike recovery (effluent)
Positive control
12
5
4
5
100711
100710
10073
110722
9775
10074
10078
96710
9578
9974
9975
9476
Recovery (mean7SD) (%)
94711
9977
10076
100710
80714
76719
98715
8973
9074
7976
9777
9877
80722
10076
9879
9478
8977
9973
9973
89714
71718
9478
9774
9578
n represents number of samples.
ARTICLE IN PRESS
WAT E R R E S E A R C H
milliliters of Milli-Q water was pre-spiked with 200 mL of
100 mg/L standard mixtures, brought to the field site, exposed
to the same conditions as the real samples and then
transferred back to the laboratory for analysis. Recoveries of
all the target analytes in the positive controls ranged from
79% to 110% (Table 2).
3.2.
Occurrence of individual antibiotics in STP influents
Concentrations of selected antibiotics in STP influents varied
among locations (Table 3). Concentrations of PEN V and CFZ
were lower than their respective MDLs in all samples
analyzed. The absence of PEN V and PEN G in the Hong Kong
samples may be attributed to their chemically reactive blactam rings that are easily hydrolyzed by bacterial blactamases (Cha et al., 2006). CLX occurred most often at
the highest concentrations (670–2900 ng/L), followed by ERYH2O (470–810 ng/L) and TET (96–1300 ng/L). These results
demonstrated that tetracyclines and macrolides (ERY-H2O)
are widely used in Hong Kong. CTX was detected at the lowest
concentrations (oMDL—24 ng/L) in sewage influents from
399
42 (2008) 395 – 403
Hong Kong. In contrast, CTX was the most abundant
antibiotic detected in influents from the Nan Shan STP in
Shenzhen with a mean concentration of 1100 ng/L.
The highest concentrations of TET (1300 ng/L), NOR (460 ng/L),
TMP (320 ng/L) and CLX (2900 ng/L) in influents were measured at the Shatin STP, whereas the greatest ERY-H2O
concentration (810 ng/L) was detected at the Wan Chai STP,
which possessed the least treatment capacity of the STPs
included in the present study (Table 1). The influent samples
from Stonecutters Island contained lower concentrations of
NOR (280 ng/L), TMP (210 ng/L) and CLX (1900 ng/L) than those
from the Shatin STP. The lowest concentrations of all
antibiotics in influent samples were detected at the Tai Po
STP: 470 ng/L for ERY-H2O, 96 ng/L for TET, 110 ng/L for NOR,
120 ng/L for TMP and 670 ng/L for CLX. As such, it is apparent
that the concentrations of antibiotics in water were independent of the sewage treatment capacity of particular STPs in
Hong Kong.
The relative concentration profiles for the seven antibiotics
detected in influents varied among locations (Fig. 1). Influents
from the STPs at Stonecutters Island, Shatin and Wan Chai
Table 3 – Antibiotic concentrations (mean7SD) in influent and effluent samples from STPs in Hong Kong and Shenzhen
STP location
Sample
Concentration (ng/L)
ERY-H2O
TET
NOR
TMP
PEN G
CLX
CTX
Wan Chai
Influent
Effluent
810711
85072.1
66072.8
620712
11072.5
11077.8
12070.7
17072.5
oMDL
oMDL
1200718
98076.0
oMDL
oMDL
Tai Po
Influent
Effluent
47072.5
52078.5
9671.6
18076.7
11070.4
8570.1
12072.1
14075.7
oMDL
oMDL
670713
24075.7
2470.1
oMDL
Shatin
Influent
Effluent
740714
600711
130070
370732
460748
10078.7
32071.4
12074.2
oMDL
oMDL
290073.5
330713
oMDL
3473.2
Stonecutters Island
Influent
Effluent
55078.0
51079.0
550710
510713
28079.0
320710
21073.0
23077.0
oMDL
oMDL
190078.0
180073.0
oMDL
oMDL
Shenzhen Nan Shan
Influent
590715
15070.7
370721
20071.8
2970
oMDL
1100716
Sewage Treatment Plant
Shenzhen
ERY-H2O
TET
NOR
TMP
PENG
CLX
CTX
Stonecutters
Island
Shatin
Tai Po
Wan Chai
0
20
40
60
Antibiotics composition (%)
80
100
Fig. 1 – Composition profiles of antibiotics in influent samples from STPs in Hong Kong and Shenzhen, China (erythromycinH2O (ERY-H2O); tetracycline (TET); norfloxacin (NOR); trimethoprim (TMP); penicillin G (PEN G); cefalexin (CLX); and cefotaxim
(CTX)).
ARTICLE IN PRESS
400
WA T E R R E S E A R C H
42 (2008) 395– 403
exhibited similar relative concentration profiles of the antibiotics, in which CLX was the most abundant, accounting for
55%, 50% and 42% of the total, respectively, and was followed
by ERY-H2O (13–27%) and TET (16–23%). A similar composition
pattern was also observed in influent samples from the Tai Po
STP except that TET contributed a lesser amount (6%) as
compared with other STPs (Fig. 1).
In contrast, a completely different antibiotic profile was
observed in the influent of the Shenzhen Nan Shan STP.
Approximately 44% of the total antibiotic concentration was
contributed by CTX, and CLX was not detected. This spatial
variation in antibiotic profiles possibly reflects regional
differences in the prescription and use patterns of antibiotics
between Hong Kong and Shenzhen.
Antibiotic concentrations reported in different countries
vary considerably, indicating different antibiotic consumption
rates. The concentrations of ERY-H2O detected in the present
study were in the ranges detected in STPs from Wisconsin,
USA (180–1200 ng/L) (Karthikeyan and Meyer, 2006), but were
comparatively higher than those in Howdon STP (71–141 ng/L),
United Kingdom (Roberts and Thomas, 2006). Similarly,
concentrations of NOR in influents from the Shatin STP
(460 ng/L) and Stonecutters Island STP (280 ng/L) were higher
than those reported in Sweden, New Mexico, USA and Finland
(Lindberg et al., 2005; Brown et al., 2006; Vieno et al., 2006). A
relatively lower concentration of TET (96 ng/L) was detected in
the Tai Po STP in the present study as compared with those
from Wisconsin, USA (240–1200 ng/L). The highest levels of
TMP detected in Sweden (1300 ng/L) and New Mexico, USA
(1400 ng/L) were approximately 4–12 times higher than the
levels found in STPs in the present study.
The highest concentrations of NOR (320 ng/L), TMP (230 ng/L)
and CLX (1800 ng/L) were detected in effluent samples from
Stonecutters Island STP, although the highest concentrations
of these antibiotics in influents were detected at the Shatin
STP. These results suggest different antibiotic removal
efficiencies at various STPs. The lowest concentrations of
TET (180 ng/L), NOR (85 ng/L) and CLX (240 ng/L) in effluent
samples were again observed at the Tai Po STP, whereas those
for ERY-H2O (510 ng/L) and TMP (120 ng/L) were measured in
samples from the Stonecutters Island and Shatin STPs,
respectively.
The concentrations of ERY-H2O (510–850 ng/L) detected in
the present study were relatively higher than those recorded
in the USA (90–300 ng/L) (Karthikeyan and Meyer, 2006),
United Kingdom (150–260 ng/L) (Roberts and Thomas, 2006)
and Italy (median concentration of 47 ng/L) (Zuccato et al.,
2005), but were lower than levels (maximum concentration of
6000 ng/L) detected in Germany (Hirsch et al., 1999). Higher
levels of NOR (85–320 ng/L) were also observed as compared
with those reported from Sweden (7–37 ng/L), France
(50–80 ng/L), Greece (70 ng/L), Italy (60–70 ng/L) and Sweden
(30 ng/L) (Andreozzi et al., 2003). Levels of TMP (120–230 ng/L)
were relatively lower than the maximum concentrations
measured in effluents from Sweden (1300 ng/L) (Lindberg et
al., 2005), California, USA (1200 ng/L) (Renew and Huang,
2004), and United Kingdom (410 ng/L) (Roberts and Thomas,
2006). The levels of TET (180–620 ng/L) were comparable with
those recorded in the USA (50–850 ng/L) (Karthikeyan and
Meyer, 2006).
3.3.
The composition profiles of individual antibiotics in the
effluent samples from Stonecutters Island and Wan Chai
STPs were similar to their corresponding influent samples
(Fig. 2). However, the relative concentration patterns of
antibiotics in effluents from the Shatin and Tai Po STPs
differed from the patterns in their corresponding influent
samples. The most abundant antibiotic measured at these
sites was ERY-H2O (45% for the Tai Po and 39% for the Shatin)
instead of CLX (Fig. 2 and Table 3). This variation can probably
be attributed to at least two factors: antibiotic prescription
Occurrence of individual antibiotics in STP effluents
Sewage Treatment Plant
The concentrations of antibiotics in effluents varied among
STPs (Table 3). Similar to influent samples, PEN V, PEN G and
CFZ were not detected in all samples. Comparatively higher
levels of CLX and ERY-H2O were measured, with concentrations ranging from 240 to 1800 ng/L and 510 to 850 ng/L,
respectively. Lower levels were detected for TET (180–620 ng/L),
NOR (85–320 ng/L) and TMP (120–230 ng/L). CTX occurred at
the lowest levels, ranging from oMDL to 34 ng/L.
3.4.
Comparisons between influent and effluent samples
Stonecutters
Island
ERY-H2O
TET
NOR
TMP
PENG
CLX
CTX
Shatin
Tai Po
Wan Chai
0
20
40
60
Antibiotics composition (%)
80
100
Fig. 2 – Composition profiles of antibiotics in effluent samples from STPs in Hong Kong (erythromycin-H2O (ERY-H2O);
tetracycline (TET); norfloxacin (NOR); trimethoprim (TMP); penicillin G (PEN G); cefalexin (CLX); and cefotaxim (CTX)).
ARTICLE IN PRESS
WAT E R R E S E A R C H
401
42 (2008) 395 – 403
Table 4 – Average antibiotic removal efficiencies (%) in wastewater from Hong Kong STPs
STP location
Wan Chai
Tai Po
Shatin
Stonecutters Island
Removal efficiency (%)
ERY-H2O
TET
NOR
TMP
PEN V
PEN G
CLX
CTX
CFZ
5
12
19
9
7
88
73
8
5
20
78
16
42
17
62
11
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
20
65
89
9
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a. represents not available: antibiotic concentrations were oMDL for influent and/or effluent samples.
patterns in Hong Kong, which determine or affect influent
concentrations, and antibiotic removal efficiencies at individual STPs, which affect effluent concentrations.
Antibiotic removal efficiencies, calculated as the difference
between mean concentrations of antibiotics in the influent
and effluent samples taken simultaneously, varied among
compounds and STPs. Removal efficiency is affected by
several factors including (i) specific treatment processes
employed by individual STPs; (ii) the sewage residence time
at different STPs; and (iii) chemical structures and properties
of the antibiotics. The removal efficiencies of the five
antibiotics commonly detected from individual STPs in Hong
Kong are summarized in Table 4.
3.4.1.
Treatment processes employed in STPs
The present results demonstrated that removal efficiencies
varied between different antibiotics and different STPs. In
general, the proportion of antibiotics removed was lower at
the Wan Chai STP (5.0–20%) and Stonecutters Island STP
(8.0–9.0%), which employed primary treatment. In comparison, secondary treatment processes used at the Tai Po and
Shatin STPs may account for the higher removal efficiencies
of antibiotics (20–65% for the Tai Po STP and 19–89% for the
Shatin STP). However, the reason(s) for the difference in
removal efficiencies between these two STPs remains largely
unknown, but could be due to the differences in the daily
loading of antibiotics into and the treatment capacity of the
individual STPs.
3.4.2.
Sewage residence time at STPs
Hydraulic retention time (HRT) is one of the major factors in
determining the antibiotic removal efficiency. In the present
study, the HRT was longer than 15 h for secondary treatment
plants (Shatin and Tai Po), but only 1–2 h for the primary
treatment plant (Stonecutters Island STP) (Table 1). Longer
exposure time for wastewater treatment generally results in
the removal of greater proportions of antibiotics (Batt et al.,
2006).
3.4.3.
Chemical structures and properties of antibiotics
In the present study, the greatest antibiotic removal capacities occurred at the Shatin STP (Table 4). Relatively lesser
proportions of NOR (20%) and CLX (65%) were removed by the
Tai Po STP. A previous study suggested that antibiotics with
lower adsorption coefficients would be expected to remain to
a greater extent in the aqueous phase, which is finally
discharged through waterways into the aquatic environment
(Ohlenbusch et al., 2000). The persistence of antibiotics also
determines their removal capabilities during treatment
processes. Macrolide antibiotics have been shown to be more
persistent than some of the other antibiotics (Huang et al.,
2001), which may account for the relatively small amount of
ERY-H2O (19%) removed at the Shatin STP. On the other hand,
the relatively higher removal rates of NOR (78%) and TET
(73%) may be attributed to their tendencies to adsorb quickly
to soils, sediments and/or sewage sludge (Huang et al., 2001;
Golet et al., 2002). It has been suggested that sorption to
sewage sludge is the primary removal mechanism for
fluoroquinolones such as NOR (Golet et al., 2003). Similar
results have been documented for STPs in Sweden (87%
removal for NOR) (Lindberg et al., 2005), Finland (100%
removal for NOR) (Vieno et al., 2006) and USA (68–100% for
TET) (Karthikeyan and Meyer, 2006). However, it should be
noted that removal efficiencies vary among STPs, and only
20% of NOR was removed in wastewater at the Tai Po STP.
CLX, a b-lactam antibiotic, is expected to possess similar
chemical properties as penicillins, and thus is easily transformed through hydrolysis. This may account for the
relatively high removal rate of CLX (89%) at the Shatin STP.
The removal efficiency for TMP at the Shatin STP (62%) was
comparable to those reported for STPs in New Mexico, USA
(50–100%) (Brown et al., 2006).
An increase in antibiotic concentrations in the final effluent
was observed in the present study (Table 3). Concentrations of
ERY-H2O and TMP were higher in effluents than influents
from the Wan Chai STP; the same pattern was found for ERYH2O, TET and TMP from the Tai Po STP, and NOR and TMP
from the Stonecutters Island STP. Higher effluent concentrations of TMP and doxycycline compared with influents were
reported in STPs in Sweden (Lindberg et al., 2005). Dextropropoxyphene, mefenamic acid, propranolol, tamoxifen, ERY and
TMP concentrations were also found to be greater in effluents
than in influents of STPs in the Lower Tyne catchment, United
Kingdom (Roberts and Thomas, 2006). Possible explanations
for these observations include (i) deconjugation of conjugated
metabolites during the treatment process (Miao et al., 2002);
(ii) an underestimation of the actual amount due to particulate matter with adsorbed antibiotics being filtered out during
sample preparation; and (iii) a change in the adsorption
behavior of the analytes to particles during treatment
processes, influencing the ratio between influent/effluent
water (Lindberg et al., 2005).
ARTICLE IN PRESS
402
WA T E R R E S E A R C H
3.5.
Potential toxic effects of fluoroquinolone antibiotics to
bacteria in STPs
Fluoroquinolones are an important group of antibiotics that
are used to treat a number of human and animal infections
(Renew and Huang, 2004). However, they have been shown to
elicit genotoxic effects in a genetically modified bacterial
strain of Salmonella typhimurium at a concentration of 5000 ng/L
(Hartmann et al., 1998). In view of this, a predicted no-effect
concentration (PNEC) of total fluoroquinolones, estimated to
be 8000 ng/L, was derived for organisms in STPs based on
acute ecotoxicity data (Golet et al., 2002). In the present study,
the maximum concentration of NOR (455 ng/L) detected in the
influent samples from the Shatin STP was lower than the
corresponding PNEC, indicating that it is unlikely that adverse
effects to organisms involved in the degradation processes
during wastewater treatment would occur due to this
antibiotic.
4.
Conclusion
Seven out of nine antibiotics, ERY-H2O, TET, NOR, TMP, PEN G,
CLX and CTX, were detected in STP influents and effluents in
Hong Kong and Shenzhen, China. Different antibiotic composition profiles were observed in influents to STPs in Hong
Kong relative to the Shenzhen STP, probably indicating
regional variations in the prescription and use patterns of
antibiotics between these two cities in South China. Removal
efficiencies varied among individual antibiotics, depending
on their chemical structures/properties, as well as the
wastewater treatment processes utilized at various STPs.
Antibiotics that readily adsorb onto particulate matter, such
as NOR and TET, exhibited comparatively higher removal
efficiencies, whereas ERY-H2O was removed to a limited
extent due to its persistence in the environment. Higher
antibiotic removal efficiencies were observed at the Shatin
STP, which used secondary treatment processes, than at the
Wan Chai and Stonecutters Island STPs, where only primary
treatment was used. Influent concentrations of NOR reported
in this study were not likely to cause adverse effects on
microorganisms involved in wastewater treatment processes.
Acknowledgments
The work described in this paper was supported by a strategic
research grant at the City University of Hong Kong (Grant no.
7001936), and the Area of Excellence Scheme under the
University Grants Committee of the Hong Kong Special
Administration Region, China (Project no. AoE/P-04/2004).
We thank Dr. Margaret Murphy, City University of Hong Kong,
for her critical review of this manuscript.
R E F E R E N C E S
Andreozzi, R., Raffaele, M., Nicklas, P., 2003. Pharmaceuticals in
STP effluents and their solar photodegradation in aquatic
environment. Chemosphere 50 (10), 1319–1330.
42 (2008) 395– 403
Ashton, D., Hilton, M., Thomas, K.V., 2004. Investigating the
environmental transport of human pharmaceuticals to
streams in the United Kingdom. Sci. Total Environ. 333 (1–3),
167–184.
Batt, A.L., Bruce, I.B., Aga, D.S., 2006. Evaluating the vulnerability
of surface waters to antibiotic contamination from varying
wastewater treatment plant discharges. Environ. Pollut. 142
(2), 295–302.
Brown, K.D., Kulis, J., Thomson, B., Chapman, T.H., Mawhinney,
D.B., 2006. Occurrence of antibiotics in hospital, residential,
and dairy effluent, municipal wastewater, and the Rio Grande
in New Mexico. Sci. Total Environ. 366 (2–3), 772–783.
Cha, J.M., Yang, S., Carlson, K.H., 2006. Trace determination of
b-lactam antibiotics in surface water and urban wastewater
using liquid chromatography combined with electrospray
tandem mass spectrometry. J. Chromatogr. A 1115 (1–2),
46–57.
Costanzo, S.D., Murby, J., Bates, J., 2005. Ecosystem response to
antibiotics entering the aquatic environment. Mar. Pollut. Bull.
51 (1–4), 218–223.
Daughton, C.G., Ternes, T.A., 1999. Pharmaceuticals and personal
care products in the environment: agents of subtle change?
Environ. Health Perspect. 107 (6), 907–938.
Dı́az-Cruz, M.S., Barceló, D., 2005. LC/MS2 trace analysis of
antimicrobials in water, sediment and soil. Trends Anal.
Chem. 24 (7), 645–657.
Gagńe, F., Blaise, C., André, C., 2006. Occurrence of pharmaceutical products in a municipal effluent and toxicity to rainbow
trout (Oncorhynchus mykiss) hepatocytes. Ecotoxicol. Environ.
Saf. 64 (3), 329–336.
Göbel, A., Thomsen, A., McArdell, C.S., Joss, A., Giger, W., 2005.
Occurrence and sorption behavior of sulfonamides, macrolides, and trimethoprim in activated sludge treatment. Environ. Sci. Technol. 39 (11), 3981–3989.
Golet, E.M., Alder, A.C., Giger, W., 2002. Environmental exposure
and risk assessment of fluoroquinolone antibacterial agents in
wastewater and river water of the Glatt Valley watershed,
Switzerland. Environ. Sci. Technol. 36 (17), 3645–3651.
Golet, E.M., Xifra, I., Siegrist, H., Alder, A.C., Giger, W., 2003.
Environmental exposure assessment of fluoroquinolone antibacterial agents from sewage to soil. Environ. Sci. Technol. 37
(15), 3243–3249.
Gulkowska, A., He, Y., So, M.K., Yeung, L.W.Y., Leung, H.W., Giesy,
J.P., Lam, P.K.S., Martin, M., Richardson, B.J., 2007. The
occurrence of selected antibiotics in Hong Kong coastal
waters. Mar. Pollut. Bull. 54 (8), 1287–1293.
Halling-Sørensen, B., Nors Nielsen, S., Lanzky, P.F., Ingerslev, F.,
Holten Lützhøft, H.C., Jørgensen, S.E., 1998. Occurrence, fate
and effects of pharmaceutical substances in the environment—a review. Chemosphere 36 (2), 357–393.
Hartmann, A., Alder, A., Koller, T., Widmer, R., 1998. Identification
of fluoroquinolone antibiotics as the main source of umuC
genotoxicity in native hospital wastewater. Environ. Toxicol.
Chem. 17 (3), 377–382.
Hernando, M.D., Mezcua, M., Fernández-Alba, A.R., Barceló, D.,
2006. Environmental risk assessment of pharmaceutical
residues in wastewater effluents, surface waters and sediments. Talanta 69 (2), 334–342.
Hirsch, R., Ternes, T., Haberer, K., Kratz, K., 1999. Occurrence of
antibiotics in the aquatic environment. Sci. Total Environ. 225
(1–2), 109–118.
Huang, C.H., Renew, J.E., Smeby, K.L., Pinkerston, K., Sedlak, D.L.,
2001. Assessment of potential antibiotic contaminants in
water and preliminary occurrence analysis. Water Resour.
Update 120, 30–40.
Karthikeyan, K.G., Meyer, M.T., 2006. Occurrence of antibiotics in
wastewater treatment facilities in Wisconsin, USA. Sci. Total
Environ. 361 (1–3), 196–207.
ARTICLE IN PRESS
WAT E R R E S E A R C H
Kim, S.C., Carlson, K., 2007. Temporal and spatial trends in the
occurrence of human and veterinary antibiotics in aqueous
and river sediment matrices. Environ. Sci. Technol. 41 (1),
50–57.
Kolpin, D.W., Furlong, E.T., Meyer, M.T., 2002. Pharmaceuticals,
hormones, and other organic wastewater contaminants in US
streams, 1999–2000: a national reconnaissance. Environ. Sci.
Technol. 36 (6), 1202–1211.
Lindberg, R.H., Wennberg, P., Johansson, M.I., 2005. Screening of
human antibiotic substances and determination of weekly
mass flows in five sewage treatment plants in Sweden.
Environ. Sci. Technol. 39 (10), 3421–3429.
Lindsey, M.E., Meyer, M., Thurman, E.M., 2001. Analysis of trace
levels of sulfonamide and tetracycline antimicrobials in
groundwater and surface water using solid-phase extraction
and liquid chromatography/mass spectrometry. Anal. Chem.
73 (19), 4640–4646.
McArdell, C.S., Molnar, E., Suter, M.J.F., Giger, W., 2003. Occurrence
and fate of macrolide antibiotics in wastewater treatment
plants and in the Glatt Valley Watershed, Switzerland.
Environ. Sci. Technol. 37 (34), 5479–5486.
Miao, X.S., Koenig, B.G., Metcalfe, C.D., 2002. Analysis of acidic
drugs in the effluents of sewage treatment plants using liquid
chromatography–electrospray ionization tandem mass spectrometry. J. Chromatogr. A 952 (1–2), 139–147.
Migliore, L., Civitareale, C., Brambilla, G., Dojmi Di Delupis, G.,
1997. Toxicity of several important agricultural antibiotics to
Artemia. Water Res. 31 (7), 1801–1806.
42 (2008) 395 – 403
403
Ohlenbusch, G., Kumke, M.U., Frimmel, F.H., 2000. Sorption of
phenols to dissolved organic matter investigated by solid
phase microextraction. Sci. Total Environ. 253 (1–3), 63–74.
Renew, J.E., Huang, C.H., 2004. Simultaneous determination of
fluoroquinolone, sulfonamide, and trimethoprim antibiotics
in wastewater using tandem solid phase extraction and liquid
chromatography–electrospray mass spectrometry. J. Chromatogr. A 1042 (1–2), 113–121.
Richardson, B.J., Lam, P.K.S., Martin, M., 2005. Emerging chemicals
of concern: pharmaceuticals and personal care products
(PPCPs) in Asia, with particular reference to Southern China.
Mar. Pollut. Bull. 50 (9), 913–920.
Roberts, P.H., Thomas, K.V., 2006. The occurrence of selected
pharmaceuticals in wastewater effluent and surface waters of
the lower Tyne catchment. Sci. Total Environ. 356 (1–3), 143–153.
Vieno, N.M., Tuhkanen, T., Kronberg, L., 2006. Analysis of neutral
and basic pharmaceuticals in sewage treatment plants and in
recipient rivers using solid phase extraction and liquid
chromatography–tandem mass spectrometry detection.
J. Chromatogr. A 1134 (1–2), 101–111.
Xu, W., Zhang, G., Zou, S., Li, X., Liu, Y., 2007. Determination of
selected antibiotics in the Victoria Harbour and the Pearl River,
South China using high-performance liquid chromatography–electrospray ionization tandem mass spectrometry.
Environ. Pollut. 145 (3), 672–679.
Zuccato, E., Castiglioni, S., Fanelli, R., 2005. Identification of the
pharmaceuticals for human use contaminating the Italian
aquatic environment. J. Hazard. Mater. 122 (3), 205–209.