Memoire Online - Evaluation des risques sanitaires et écotoxicologiques liés aux effluents hospitaliers

Evens Emmanuel^lr^2*, Yves Perrodin^l, Gérard Kecle, Jean-Marie Blanchard², Paul Vermande²

¹ Laboratoire des Sciences de l'Environnement, École Nationale des Travaux Publics de l'État, Rue Maurice Audin, 69518 Vaulx-en-Velin, France

² Laboratoire d'Analyse Environnementale des Procédés et Systèmes Industriels, Institut National des Sciences Appliquées de Lyon, 20 avenue Albert Einstein, 69621 Villeurbanne Cedex, France

³ Unité d'Ecotoxicologie, Ecole Nationale Vétérinaire de Lyon, BP 83, 69280 Marcy l'Etoile, France

Keywords : Hospital wastewater, ecotoxicological risk assessment, pharmaceuticals, disinfectants, toxicity, Vibrio fischery, Pseudokirchnerie//a subcapitata, Daphnia magna

* Corresponding author. Tel : +(33) 4 72 04 72 89; fax:+(33) 4 72 04 77 43 E-mail address : evemml eyahoo.fr

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

In hospital a variety of substances are in use for medical purposes as diagnostics and research. After application, diagnostic agents, disinfectants and excreted non-metabolized pharmaceuticals by patients, reach the wastewater. This form of elimination may generate risks for aquatic organisms. The aim of this study was to present (i) the steps of an ecological risk assessment and management framework related to hospital effluents evacuating into wastewater treatment plant (WWTP) without preliminary treatment; and (ii) the results of its application on wastewater from an infectious and tropical diseases department of a hospital of a big city of the southeast of France. The characterization of effects has been made under two assumptions, which were related to : (a) the effects of hospital wastewater on biological treatment process of WWTP, particularly on the community of organisms in charge of the biological decomposition of the organic malter; (b) the effects on aquatic organisms. COD and BOD₅ have been measured for studying global organic charge. Assessment of organo halogenated compounds was made using AOX (halogenated organic compounds absorbable on activated carbon) concentrations. (3) Heavy metals (arsenic, cadmium, chrome, copper, mercury, nickel, lead and zinc) were measured. Low MPP (most probable number) for fecal bacteria has been considered as an indirect detection of antibiotics and disinfectants presence. For toxicity assessment, bioluminescence assay using Vibrio fischeri photobacteria, 72-h EC₅₀ algae growth Pseudokirchnerie//a subcapitata and 24-h EC₅₀ on Daphnia magna were used. The scenario allows to a semi-quantitative risk characterization. It needs to be improved on some aspects, particularly those linked: to long term toxicity assessment on target organisms (bioaccumulation of pollutants, genotoxicity, etc.); to ecotoxicological interactions between pharmaceuticals, disinfectants used both in diagnostics and in cleaning of surfaces, and detergents used in cleaning of surfaces ; to the interactions into the sewage network, between the hospital effluents and the aquatic ecosystem.

Hospitals use a variety of chemical substances such as pharmaceuticals, radionuclides, solvents, disinfectants for medical purposes as diagnostics, disinfections and research [1-3]. After application, some of these substances and excreted non-metabolized drugs by the patients enter into the hospital effluents [4, 5], which generally reach, as well as the urban wastewater (figure 1), the municipal sewer network without preliminary treatment [6, 7]. Unused medications also are sometimes disposed of hospital drains [5]. Pollutants from hospital were measured in the effluents of WWTP, and in surface water [8]. Due to laboratory and research activities or medicine excretion into wastewater, hospitals may represent an incontestable release source of many toxic substances in the aquatic environment [9].

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

Figure 1 : Problems of hospital effluents and their impacts
on WWTP and natural environments

The contact of hospital pollutants with aquatic ecosystems leads to a risk directly related to the existence of hazardous substances which could have potential negative effects on biological balance of natural environments. Risk is the probability of appearance of toxic effects after organism exposure to hazardous substances [10]. In the context of hospital wastewater discharge into the aquatic ecosystem, the exposure to hazardous substances, particularly disinfectants, non-metabolized pharmaceuticals and radionuclides, requires to consider possible risks for aquatic organisms. The fate of pharmaceuticals in the aquatic environment have been reported in different reviews of the literature [3, 4, 8, 11]. The ecological risk of glutaraldehyde, a dialdehyde usually recommended as the disinfectant of choice for reusable fiber-optic endoscopes, has been also treated in other study [9]. However, few studies treat with total risk resulting from the simultaneous exposure to various pollutants present in the hospital effluents.

French legislation fixes the conditions for the connection of hospital wastewater system into the urban sewer network [12]. In the Directive N° 793/93, on the human and ecosystem exposures to the classified toxic substances, the European Commission [13] requires to all member states to carry out a sanitary and ecological risk assessment for substances such as: drugs, disinfectants and radioactive substances. These regulations fall under the context of the risk management concerning human health, and also the management of those concerning the biological balance of the natural ecosystems. In a very general way, the risk management always passes - formally or not - by the preliminary phases of risk assessment [14]. The aim of this study was to present: (i) an implemented

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

framework for hospital wastewater management, which includes two steps : a "light" step based on the hazard assessment related to hospital effluents and, if proof of hazard existence occurred, the execution of a "heavy" step, based on an ecotoxicological risk assessment of hospital wastewater discharging into the urban sewer network, then into the natural environment (ii) detailed elaborated procedures for the steps of "hazard assessment" and "risk assessment" (iii) the results of their application on the effluents of an infectious and tropical diseases department (ITDD) of a hospital of a big city of the southeast of France.

Hospitals consume an important volume of water per day. The minimal domestic water consumption is 100 liters/person/day [15], whereas the value demand for the hospitals generally varies from 400 to 1200 liters/bed/day [16, 7]. In the United States of America, the hospital average water consumption is 968 liters/bed/day [17]. In France, the water average needs of university hospital centers is estimated at 750 liters/bed/day [7]. In the developing countries, this consumption seems to be around 500 liters/bed/day [18]. This important consumption in water of hospitals gives significant volumes of wastewater. Results of toxicity studies using the bacteria bioluminescence and Daphnia magna have revealed the important toxic activities of hospital wastewater on aquatic organisms [19].

The most frequent contaminants in hospital wastewater are : viruses and pathogenic bacteria (some of them are antibacterial resistant characters) [20], molecules from unused and excreted nonmetabolized pharmaceuticals [4], organohalogen compounds, such as the AOX (halogenated organic compounds adsorbable on activated carbon) [5], radioisotopes [21, 1].

Results on the microbiological characterization of hospital wastewater [20] reported these effluents have bacteria concentrations lower than 10⁸/100mL generally present in the municipal sewage system [21]. The low most probable number (MPN) detected for fecal bacteria in hospital is probably due to the presence of disinfectants and antibiotics [6, 20]. Markers of viral pollution of water, such as enterovirus, and other viruses have been identified in the hospital effluents [23]. Studies on the bacteria flora of hospital wastewater into WWTP showed that bacteria acquired resistant character [24]. Antibacterial resistant is a threat to the efficacy of antibacterial substances. The development of resistance to antimicrobial agents by many bacterial pathogens has compromised traditional therapeutic regimens, making treatment of infections more difficult [4]. Three factors have contributed to the development and spread of resistance: mutation in common genes that extend their spectrum of resistance, transfer of resistance genes among diverse microorganisms, and increase in selective pressures that enhance the development of resistant organisms [24, 25, 26, 4, 27, 28]

Hospital effluents reveal the presence of organochlorine compounds in high concentrations [6]. AOX
up to 10 mg/L were proved in the effluents of the hospitalization services of a university hospital

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

center [29]. The major mass carriers for the AOX in hospital effluents are most likely iodized X-ray contrast media, solvents, disinfectants, cleaners and drugs containing chlorine. Brominated organic compounds are negligible for the AOX in the hospital effluents [5]. In general, the maximum contribution of drugs to the AOX is not above 11% [30]. Beyond that it is also known that the AOX concentration in the urine of patients not treated with drugs is very low. It is normally between 0.001 to 0.2 mg/L [31]. Due to the dilution effect, no substantial contribution from this source is consequently expected [5]. The assessment of AOX shows that those non conventional pollutants have a bad biodegradability and a bad behavior of adsorption [8].

The conceptual framework for hazard assessment of hospital wastewater (figure 2), is based on a characterization of the hospital effluents in function: (i) of their chemical composition (measurement of global parameters and the minerai and organic pollutants); (ii) of their microbiological characterization; (iii) and of their intrinsic ecotoxicity.

Hospital effluents from (main or a department) collector

1^, vl^,

Chemical characterisation CC

- global parameters

- heavy metals

Microbiological characterisation MC - fecal coliforms

CCno regulations

Bioassais (UT = 100/ EC50)

- luminescence bacterial

- growth of alguae

- mobility of Daphnia

MC
10 NPP

Non hazardous effluents for

yes

UT no

egulation

the ecosystems

autorisation of discharge into the urban sewer network

yes

Hazardous effluents for the

Hypotheses:

ecosystems

- presence of toxic substances

- risks for ecosystems

- presence of disinfectants,

detergents and/or surfactants - presence of antibiotics

Ecotoxicological risk assessment

Figure 2 : Conceptual framework for ecotoxicological hazard
assessment of hospital wastewater

The selected parameters (stressors and assessment endpoints) for these characterizations were: (1) COD and the BOD₅ for the measurement of the total organic load; (2) the AOX (organohalogen compounds adsorbables on activated carbon) for the evaluation of the contained organohalogen compounds; (3) heavy metals (arsenic, cadmium, chromium, copper, mercury, nickel, lead and zinc) for the minerai pollution characterization; (4) the most probable number of fecal bacteria for the microbiological characterization (this parameter was also considered in this study like an indirect

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

detection of the massive presence of disinfectants and/or antibiotics); (5) the measurement of EC₅₀ of hospital wastewater on bacterial luminescence (Vibrio fischen), on the algae growth (Pseudokirchnerie//a subcapitata) and on the mobility of Daphnia magna for the characterization of the intrinsic ecotoxicity of the effluents.

The obtained results for these parameters have been compared with threshold values which were established in the following way : 1) global parameters: French regulations on effluents discharge (table 1); (2) ecotoxicological parameters: adopted threshold values at 2 Toxic Units (UT) [32, 19] for each of selected bioassays; (3) microbiological parameter : value threshold fixed at 1x10⁸ fecal coliforms for 100 ml, value corresponding to the average content of these fecal bacteria in the conventional urban sewer network [22].

Table 1: Threshold values for the different endpoints
Parameters	Threshold values	Origin
Physicochemical BOD₅ COD	mg/L 30 125	[12] [12]
AOX	1	[12]
Arsenic
Cadmium
Chromium	0.5	[12]
Copper	0.5	[12]
Mercury
Nickel	0.5	[12]
Lead	0.5	[12]
Zinc	2	[12]
Microbiological
Fecal coliforms	1x10⁸	[22]
Bioassais
EC₅₀ 30 minutes Vibrio fischeri	2 UT	[32, 19]
EC₅₀ 72-h Pseudokirchnerie//a subcapita	2 UT	[32, 19]
EC₅₀ 48-h Daphnia magna	2 UT	[32, 19]

For any ratio P_c/V_t > 1 (ID,: pollutant concentration in the hospital effluents; V_t: threshold values) and for any number in fecal bacteria lower than 1x10⁸ NPP for 100mL, the framework recommends to pass at the following step : "the ecotoxicological risk assessment of hospital wastewater".

The ecotoxicological risk assessment is a subset of the ecological risk assessment and can thus, for this reason, being treated according to an approach of the same type. Ecological risk assessment is a process that evaluates the likelihood to one or more stressors [33]. This process is based on two major elements: characterization of effects and characterization of exposure; these provide the focus for conducting the three phases of risk assessment (figure 3): problem formulation, analysis phase and risk characterization phase [34].

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré


		Problem Formulation		4


ul Th >. c Q<	Characterization Of Of Exposure		Characterization Ecological Effects


V V - Risk characterization

The step is a process for generating and evaluating hypotheses about why ecological effects have occurred, or may occur, from human activities [34]. It provides the foundation for the entire ecological risk assessment. Problem formulation results in three products [34]: (1) assessment endpoints that adequately reflect management goals and the ecosystem they represent, (2) conceptual models that describe key relationships between a stressor and assessment endpoint or between several stressors and assessment endpoints, and (3) an analysis plan.

This description, whose aim was to apprehend as well as possible the ecosystem exposure to the hospital effluents, was carried out for a management scenario of hospital wastewater usually observed in industrialized countries. This scenario envisages the connection of the hospital sewer network to the urban sewer network, as well as the biological WWTP which discharge its own effluents into the natural environment. A synthetic description of this scenario is presented in Figure 4. The full features ( ) indicated transport and transfers of the pollutants which has been taken into account in the

study, whereas the features in dotted lines ( ) indicate those which were not taken into account.

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

Two types of exposed ecosystems to the hospital wastewater pollutants have been considered in the studied scenario (table 2): (1) artificial ecosystems represented by the WWTP and (2) natural ecosystems represented by air, soils, surface water and groundwater.

Artificial WWTP bacteria, algae and protozoa (in case where the biological treatment units have reactors of decomposition functioning under the "aerobic" mode).

the primary (invertebrate), in particular of the crustaceans ; and secondary consumers, of which fish and water birds Groundwater Protection of fresh water resources

Development of the conceptual mode/ and choice of the parameters of evaluation

Within the framework of this evaluation, the WWTP, the fresh surface water and the species at the two first levels of food chains have been considered as the targets (figure 5). The fact, that the other ecosystems and the other species do not have been considered, does not mean that those are less importance in the ecological level, but simply which they were not taken into account in this first stage of the methodology development.

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

For the characterization of the effects, two assumptions were elaborated. They have been related to the ecological values to be protected: (a) "the discharge of hospital pollutants into the WWTP will not affect the biological treatment process of WWTP, with possible adverse effects on the community of organisms in charge of the biological decomposition of the organic malter "; (b) "the WWTP effluents will not have toxicological effects on the living species of the natural aquatic environments".

Measures	B : bacteria (luminescence)
Effects
		A : algae (growth)

			C : crustacea (mobility of Daphnia)

The characterization of the ecological effects of hospital pollutants on the bacteria, the algae growth and the crustacean survival, was carried out using standardized bioassays. In this context, the bacteria were represented by "Vibrio fischerf , the species constituting the primary producers (phytoplankton) were represented by the algae "Pseudokirchnerie//a subcapitata", and the fresh water crustaceans "Daphnia magna Strauss" ensured the representation of the primary consumers.

Analysis is a process that examines the two primary components of risk, exposure and effects, and their relationships between each other and ecosystem characteristics [34].

Wastewater from a hospital of a big city of the southeast of France were used for the realization of
the experimental phase of this study. It is a hospital of 750 beds approximately. Water consumption is
estimated at 1m³/lit/day. The effluents from the various departments are discharged into the hospital

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

network sewer. This network consists of several collectors broken down by service or group of related services. The institution has a combined sewage system. The existence of such network could increase the concentration of the nitrogen substances during the first raining days and a dilution of all the pollutants during the other raining days [35]. This network could also increase the concentration of certain heavy metals, particularly zinc.

Two campaigns of sampling (2001 and 2002) were realized on the effluents from the infectious and tropical diseases department (ITDD), with a capacity of 144 beds, of the hospital. Wastewater was collected before entering into the entire hospital sewer network, which discharges the total volume of effluents from the various departments into the urban wastewater network without pre-treatment. This ITDD collector does not receive effluents containing iodized X-ray contrant media from radiography department, substances which mainly contribute to AOX formation in hospital wastewater [8]. Water samples were collected by means of a telescopic perch in a 1-L glass flask. Ail the water samples and the mixture were kept at 4°C until analysis.

pH was measured directly on site after sampling with a pH meter HANNA instrument HI 8417 (accuracy pH #177; 0.01pH, mV #177; 0.2 mV #177; 1, °C #177; 0.4°C) digit and standard electrode HI 1131 B (refillable glass combination pH electrode).

Heavy metals have been determined according to ISO 11 885 protocol on filtered sample (0.45 pm) and acidified using nitric acid (pH<2) and using ICP-AES (Inductively Coupled Plasma-Atom Emission Spectroscopy).

Total suspended solids (TSS) concentrations were determined in conformity with the European standard NF EN 872 after filtration through a 1.2 pm membrane and dewatering at 105 °C.

Chlorides were determined by following the European standard NF EN ISO 10304-1 on diluted and filtered samples at 0.45 pm by using DIONEX DX-100 ion chromatograph with suppressed conductivity detection from 0.0 to 1000 pS. Ionpac AS14 4x250 mm analytical column (P/N 046124) was used for chloride sample analysis. AOX were measured according to European standard EN 1485. COD concentrations in 2001 samples was measured by potassium dichromate method using HACH spectrophotometer 2010 and test procedure provided by the supplier. French standard NF T90-001 had been followed in the determination of COD concentrations in 2002 samples. BOD₅ concentrations in the 2001 and 2002 samples were carried out by following European and French standard NF EN 1899-1.

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

Total Organic Carbon (TOC) was carried out on filtered samples at 0.45 pm and pre-treated with orthophosphoric acid (H3PO₄). French standard T90-102 was followed by using a carbon analyzer SPECTRA France, LABTOC model, with potassium per sulfate reagent (K2S208) and UV oxidation.

Fecal bacteria have been studied using French standard NF T 90-433 micro plaque. The French standard NF T 90-432 micro plaque and NF T 90-145 have been respectively used for the determination of faecal streptococci and Clostridia (anaerobic spore forming bacterium).

For the study of assessment endpoints, three standardized bioassays were carried out. Results of EC₅₀ for all these bioassays, with their confidence interval, are expressed in percentage of sample dilution in toxic unit TU (1 TU = 100 / EC₅₀)
·

The bioassay on bacteria luminescence was carried out with a LUMIStox system (Dr Lange GmbH, Duesseldorf, Germany) following the standard procedure of the European standard NF EN ISO 113483 (AFNOR, 1999). Tests were performed using gram negative marine bioluminescent bacteria of the species Vibrio fischeri NRRL-B-11177 of the Vibrionaceae family. In order to prevent the interferences of TSS on the bacteria luminescence, samples were filtered using a 0.45pm pore size membrane. The samples were treated with NaCI solution of 20 g/L and brought to 50 mS/cm of conductivity before the analysis. Starting from the concentration of the sample, eight consecutive elutions were tested (dilution factor 1:2); the inhibition of bioluminescence was measured at a wavelength of 490 nm, with readings after 5, 15 and 30 minutes of incubation at 15 °C. The EC₅₀ values were calculated as reported by Bulich [36].

The 72-h EC₅₀ algae growth toxicity test was monitored using French standard NF T90-375. Assays was carried out with the green algae inoculums Pseudokirchnerie//a subcapitata (formerly Selenastrum capricornutum) resulting from laboratory culture in exponential growth phases (POLDEN of the National Institute of Applied Sciences of Lyon -- INSA de Lyon). The sensibility of the laboratory species was controlled by regular tests with potassium dichromate. Standard diluted medium was used with 0.1 mg of EDTA per liter of assay solution. In order to avoid the interferences of suspended solids and other microorganisms on algae growth during the realization of the assay, experimental solutions were filtered at 0.45 pm. Experimental solutions were maintained at 4°C #177; 3°C. A set of five concentrations of experimental solution samples in the reference medium and a control were examined in each test. Assays were carried out in glass cups containing 25 mL of samples, with 3 replicates by concentration. The assay is static, under magnetic agitator and under constant luminosity, at 23°C #177; 2°C . Algae concentration were measured all the 24 hours using Malassez cell and optic microscope.

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

The determination of the inhibition of Daphnia magna mobility is a acute toxicity assay. Its objective is to identify the initial concentration of a pollutant in solution or an aqueous mixture which may immobilize in 24 or 48 hours 50% of exposed daphnia into polluted solutions. According to the European standard NF EN ISO 6341, the different assays were carried out on Daphnia sp. maintained in parthenogenetic culture in the laboratory (POLDEN of the National Institute of Applied Sciences of Lyon -- INSA de Lyon). The sensibility of the laboratory species was controlled by regular tests with potassium dichromate. Young female Daphnia, aged less 24 h were only used. The normal medium, without EDTA, was also used. The essays were realized at 20 #177; 2°C under darkness condition. All the assays were carried out in a limit of time from 6 to 48 h after sampling. Because hospital wastewater is considered as toxic for aquatic environment, a volume of 250 mL unfiltered samples was taken for each assay. In order to understand the effects of color, turbidity and TSS present in hospital effluent samples on Daphnia sp., the toxicity of 250 mL filtered sample (0.45 pm pore size membrane) was studied by comparing the results with the unfiltered volume of the same sample. The three required conditions for the validity of assays were observed: (i) the concentration of dissolved oxygen (DO), in the control group, was 2 mg/L at the end of each assay; (ii) the observed percentage of immobilization in the control group vessels was 10%; (iii) EC₅₀ in 24h of potassium dichromate was between 0.6 to 1.7 mg/L.

This operation is the final phase of ecological risk assessment and is the culmination of the planning, problem formulation, and analysis of predicted or observed adverse ecological effects related to the assessment endpoints [34]. There is a range of possible methods, of variable complexity [14]. The choice will depend on the operational constraints and the available data. Rivière [10] note "the ecological risk can be expressed of various manners: qualitative (absence or not of risk), semiquantitative (weak, average, high risk), in probabilistic terms (the risk is x%)".

The method known as "the quotient" is the most widespread method for the semi-quantitative characterization of risks. This method consists in calculating the ratio (or quotient) which is expressed as a "probable exposure concentration (PEC)" divided by a "probable non concentration effect (PNEC)" [34]. This "probable concentration without effect" can be estimated starting from the available data in the literature for the pure substances, and using experimental measurements (bioassays) for the mixture such as the hospital effluents. Although the toxicity of a chemical mixture may be greater or less than predicted from toxicities of individual constituents of the mixture, a quotient addition approach assumes that toxicities are additive or approximately additive [34]. This assumption may be most applicable when the modes of actions of chemicals in a mixture are similar, but there is evidence that even with chemicals having dissimilar modes of action, additive or near-additive interactions are common [37, 38, 39, 34].

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

When the quotient value "Q" is greater than 1, the risk is considered as significant, and all the more extremely as the quotient is large. Conversely, more the quotient is lower than 1, more the risk is regarded as weak. The "probable concentration without effect" on the organism is, in practice, generally represented by a EC₁₀, or a EC20, or a NOEC, divided by a safety factor (10 for example). In the absence of a EC₁₀ or of a NOEC, the EC₅₀ is sometimes used with a rated-up safety factor [14].

The highest concentrations obtained for the physicochemical characterization of the hospital wastewater from ITDD are summarized in tables 3. In all studied samples of the two campaigns (2001 and 2002), pH was always in an alkaline range (7.7 -- 8.8) with a variation lower than 1 pH unit.

	2002
8.8	8.2
359	127.1
1.24	1.61
298	236
1559	1530
2516	2664
350	3095
376	3183
<0,0005		ND
ND		0,011
ND		<0,007
ND		<0,004
ND		0,112
<0,0005		NM
ND		<0,0007
ND		<0,0035
ND		0,536

Low concentrations of bacteria flora were deducted for the hospital effluents. The results of the bacteriological characterization are summarized in table 4.

Parameters	Units	Maximal concentrations
		2001	2002
Fecal coliforms	MPN/100 mL	2 x 10³	1 x 10⁶
Fecal streptococcus	MPN/100 mL	1 x 10³	9 x 10⁵
Clostridia	UFC/20 mL	ND	104

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

The obtained results for the bioassays are synthesized in table 5. All obtained CE₅₀ from algae and Daphnia magna bioassays were greater than 2 TU. The results of toxicity test on Vibrio fischeri obtained for the year 2001, lead to EC₅₀ (5 minutes) greater than 50% of effluent for all the samples, i.e. with an ecotoxicity, expressed in UT, always lower than 2 UT. These results showed that 5 min assay can be considered as no toxic. However, significant differences were observed between EC₅₀ (5 minutes) and obtained results for EC₅₀ (15 and 30 minutes). In addition, there exists very little differences between the obtained results for 15 and 30 minutes assays. This report can be correlated with the contact time of 20 minutes contact required by chlorinated disinfectants to inactivate bacteria [40]. The results of 15 and 30 min greater than 2 TU. The maximal concentrations ranged from 4.2 to 4.6 showed that the hospital wastewater toxicity on Vibrio fischeri are similar to domestic wastewater toxicity. However, all the obtained results were lower than the means of 6.75 TU reported by the literature for the toxicity of hospital wastewater on V. fischeri completed after 30 min of exposure [19].

According to the proposed framework, the hazard assessment of hospital effluents to the aquatic ecosystems consists to compare the obtained results for physicochemical, microbiological and ecotoxicological characterizations with the threshold values presented in table 1 for the different parameters. Table 6 showed the results of this comparison.

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

Table 6 : Comparison of the highest concentrations with the threshold values

Parameters	Units	Highest measured concentrations	Threshold values	Ratio Pc/Vt
Physicochemical
BOD₅	mg/L	1559	30	>1
COD	mg/L	2664	125	>1
AOX	mg/L	1,61	1	>1
Chromium	mg/L	<0,004	0,5	<1
Copper	mg/L	0,112	0,5	<1
Nickel	mg/L	<0,0007	0,5	<1
Lead	mg/L	<0,0035	0,5	<1
Zinc	mg/L	0,536	2	<1
Microbiological
Fecal bacteria	NPP/100 mL	1 x 10⁶	1x10⁸	<1
Ecotoxicological
EC₅₀ 30 min. Vibrio fischery	UT	4,6	2	>1
EC₅₀ 72 h	UT	56	2	>1
Pseudokirchnerie//a subcapitata
EC₅₀ 24 h Daphnia	UT	117	2	>1
EC₅₀ 48 h Daphnia	UT	71	2	>1

With the exception of the heavy metals, all the ratios P_c/V_t carried out for the other physicochemical parameters were greater than 1. The same observation was made for the bioassays ratios. In addition, the results of genotoxicity tests on hospital wastewater using AMES and HAMSTER, reported in the literature, indicated that the effluents from clinicat services and hospital laboratories have presented a genotoxicity character [29].

The ratio, by dividing the MPN/100 mL of fecal bacteria from hospital wastewater with the average of those usually found in the urban effluents, was largely lower than 1, that could, at least partially, being related to the presence of disinfectants and/or antibiotics in the effluents.

Ail the results confirm the existence of hazardous substances in the studied hospital effluents, and thus the need for continuing the approach by the setting of the ecotoxicological risk assessment of hospital wastewater for the concerned aquatic ecosystems (WWTP and natural environment).

In the absence within the hospital of pollution control practices for wastewater, or of its own WWTP, all the contained pollutants into the ITDD effluents as those of the whole of the hospital are evacuated towards the municipal WWTP. In the proposed scenario, an &Oficial ecosystem "the WWTP" as well as the natural aquatic ecosystem were retained as targets, by restricting the study to the species of the two first levels of aquatic food chains.

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

Assumption: " the discharge of hospital pollutants into the WWTP will not affect the biological treatment process of WWTP, with possible adverse effects on the community of organisms in charge of the biological decomposition of the organic malter ".

The preservation of the biological efficiency of a WWTP can, in a first approach, being evaluated by means of the biodegradability studies of inflow pollutants. The biodegradability of organic substances is a measure of the speed and completeness of its biodegradability by microorganisms [41], and therefore the BOD₅/COD ratios could be used to analyze the difficulty or not for organic substances to be degraded [42]. A BOD₅/COD of ratio 0.5 or greater could be considered as threshold value to study the biodegradability of organic substances into the ITDD hospital wastewater. The variations of BOD₅, COD, and the BOD₅/COD ratio in the samples of ITDD hospital wastewater, for the 2002 campaign, are showed in table 7. BOD₅/DCO ratio oscillated between 0.38 and 0.57, which indicate that the pollutants would be sometimes difficult to degrade, which describes a potential impact on the WWTP efficiency.

Parameters	Units	Means	Minima	Maxima	SD	n
BOD₅	mg/L	892	200	1530	558	5
COD	mg/L	1745	520	2664	894	5
BOD₅/DCO		0,48	0,38	0,57	0,08	5

To evaluate in a semi-quantitative way the risks of a term-source on the ecosystems in a specific context, it is possible in a first approach, to consider the dilution coefficients generated by the global system. Within the framework of this study, three assumptions of dilution were considered for the risk characterization of hospital wastewater on the WWTP: (i) the daily flow of water supply by bed per day is equal to the volume of wastewater generated by bed per day; (ii) the ITDD generates a volume of wastewater of 144 m³/day. In absence of specific considerations on the interactions between the various pollutants inside the hospital sewer network, the contained organic pollutants in the effluents of the service will be diluted at least of 4 times in total volume, i.e. 750 m³/day of wastewater on average are generated by the different services of the hospital, before entering the urban sewer network; (iii) the ITDD effluents are treated into the WWTP of the considered city, this plant receives on average a hydraulic daily load of 87000 m³, which ensures a dilution of the measured pollutant concentrations in the hospital effluents at least of 600 times.

In this context, the ITDD effluents will not have a significant effect on total efficiency of the WWTP.
Indeed, if taking into account the fact that the evaluation of WWTP efficiency is expressed as a
percentage (70 to 90 %) of degradation of the organic matters, the WWTP mechanisms will be always

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

able to reach this efficiency level fixed by the regulation, since this performance at the end of process remains a function of the input concentrations. However, this method of evaluation will not allow to prevent the discharge into WWTP effluents of low biodegradable and toxic pollutants (like pharmaceutical residues and AOX) for the ecosystems.

Assumption: "the WWTP effluents will not have toxicological effects on the living species of the natural aquatic environments".

The ecotoxicity tests carried out for the hazard assessment can be reused in this step, but by interpreting their results in the light of the specific conditions of the studied scenario, in particular by taking in account the dilution of hospital wastewater in the urban network then in the target natural aquatic environment. It was seen previously that the dilution of hospital effluents in WWTP was equal to 600. For this, it is necessary to add, in the studied case, a dilution by 1000 of the WWTP effluents in the river water bodies. That led in fineto a dilution of 6x10⁵ of the hospital effluents to their arrivai in the receiving receptor. On this basis, the results of the various bioassays carried out on the effluents (table 8) show that dilutions in the natural environment are largely sufficient to protect itself from the studied ecotoxicity effects.

This very simplified and very operational first approach implies however assumptions which for some are rather pessimistic and, for others, relatively "imperfect" and being able, so to lead to an incomplete assessment of long-term impacts of the hospital effluents on the natural environments.

Concerning the "pessimistic" aspects, the reasoning is led as if the pollutants in the hospital effluents were not degraded, and any volatilization process has been occurred during their transport in the urban sewer network, and during their passage in the WWTP. However, this interpretation is not completely aberrant in comparison with the characteristics of some pollutants such as the AOX, which are considered to be non biodegradable with 90% by certain authors like Sprehe et al. [43]. If these assumptions had led in fine to a positive evaluation of the ecotoxicological risks, it would have been necessary to conduct a thorough study of the concerned phenomena. In the contrary case which we are concerned, savings of time and means (and thus "effectiveness") were carried out on these points.

Concerning the aspects "incomplete assessment", the approach based on standardized ecotoxicity test and the dilution of the effluents in the natural environment implies imperfections on several levels: (1) the battery of the selected bioassays is limited. Thus organisms such as fish, for example, were not taken into account, (2) the long-term effects of the pollutants in question on the ecosystems are complex and difficult to evaluate on the basis of mono-specific simple test of ecotoxicity. Thus

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

phenomena such as the genotoxicity of the pollutants or the their bio-accumulation in the food chains or the sediments of the river (with delay effect) were not treated. Field work and/or on reconstituted ecosystems in laboratory, such as tests on microcosms [44, 45] would make it possible to better apprehend these complex phenomena, (3) the reasoning on the basis of dilution cannot be sufficient in term of decision for the environmental protection. Indeed, of many other effluents are rejected into the same "target" medium. It will be thus more judicious in the future, and for an enlightened decision-making of the managers, to reason rather in term of contribution of the hospital effluents to the total risk generated by the discharge of all the industrial and urban effluents in the concerned river.

This study has demonstrated that it is possible to carry out the ecotoxicological risk assessment of hospital effluents by the use of standardized bioassays, global physicochemical parameters and the analysis of some targeted pollutants. The proposed scenario allows to a semi-quantitative risk characterization for the WWTP and the fresh surface water. The evaluation will have now to be improved on certain aspects, and will require in particular a better knowledge on the fates of pollutants in the urban sewer network and in the WWTP. This improvement of knowledge will relate in particular to the study of chemical and ecotoxicological interactions between pharmaceuticals, disinfectants, and surfactants. It seems necessary to characterize the ecotoxicological risk of the hospital effluents by experimental and fundamental studies on the fates of disinfectants, pharmaceuticals and surfactants present in the hospital effluents, while having care to include, on the ecotoxicological plan, the transfers towards the food chains.

Chapitre V Application de la méthodologie élaborée pour l'évaluation des risques écotoxicologiques liés aux effluents hospitaliers d'une ville d'un pays développé tempéré

Chapitre VI Etude spécifique sur le devenir de deux désinfectants largement utilisés dans les hôpitaux : l'hypochlorite de sodium et le glutaraldehyde

Evaluation des risques sanitaires et écotoxicologiques liés aux effluents hospitaliers

II. Ecotoxicological risk assessment of hospital wastewater : a proposed framework for raw effluents discharging Int° urban sewer network