Cyanobacteria (Blue-Green Algae) in Inland and Inshore Waters: Assessment and Minimisation of Risks to Public Health

Guidance to Directors of Public

Health, to Heads of Environmental Health in Local Authorities (LAs), and to others in

Scotland, on possible risks to public health of cyanobacteria (blue-green algae) in inland

and inshore waters. It updates previous guidance under the same title that was published

by the Scottish Government Health Directorate previously in 2002, and revised in 2007.


3. PUBLIC HEALTH CONCERNS

3.1 Surveys in different parts of the world have found that between about 45% and 90% of blooms of cyanobacteria produce toxins (cyanotoxins). These toxins are largely retained within the cyanobacterial cells during their development and growth phases and are released, in the main, on cell death (Codd GA & Bell SG (1996)).

3.2 The range of cyanotoxins is diverse and includes neuro- and hepatotoxins, skin irritants, and inflammatory agents. A cyanobacterial bloom may contain more than one species, each producing the same or different toxins, either singly or in combination. In addition, the toxicity of one species might change over time dependent on the nutritional state of the algae, and for different places on a particular water body. Further information on cyanotoxins and their health effects is given in Annex B.

3.3 Evidence on toxicity is drawn from reports of the effects of exposure of people and of animals to cyanobacterial blooms and from laboratory investigations of cyanotoxins. Overall there have been relatively few published reports worldwide of significant human health impacts associated directly with exposure to cyanotoxins linked to cyanobacterial incidents. However, surveillance of illness associated with such exposures is likely to be incomplete.

3.4 In 1989, a group of soldiers took part in kayak training, including rolling and swimming exercises, at Rudyard Lake in Staffordshire. Two became severely ill with atypical pneumonia; others reported abdominal pains, vomiting, diarrhoea, blistering of the mouth and sore throats. Further incidents involving human health impacts have occurred after recreational contact with cyanobacterial scums and blooms in UK inland waters in recent years. The effects were probably associated with exposure to cyanobacteria and ingestion of the toxin-containing cyanobacterial scum.

3.5 Gastroenteritis, neurological effects, acute hepatocellular damage and skin irritation have been reported from other countries. Illnesses and deaths of haemodialysis patients, attributed to exposure to cyanotoxins in inadequately-treated water, occurred in Brazil in 1996. Further exposures of haemodialysis patients to cyanotoxins, followed by illness, occurred in Brazil in 2001.

3.6 Ingestion of hepatotoxic and neurotoxic scums of cyanobacteria is reported to have caused the deaths of cattle, horses, sheep, dogs, and a wide range of wild terrestrial animals and domestic and wild birds. There is also evidence that cyanotoxins have been major contributors to fish kills and the deaths of other aquatic animals.

3.7 Another potential source of intoxication for both animals and humans is via bioaccumulation of cyanotoxins in the food chain. The principal concern here would be accumulation of the toxins in shellfish including freshwater and brackish-water mussels and in fish. However, no cases of intoxication from such sources have been reported to date in Scotland.

3.8 Episodes of cyanobacterial contamination of drinking water supplies occur periodically:

(i) In September 1997, a massive cyanobacterial bloom affected the main water supply loch on Westray, Orkney Isles, and resulted in a ban on the use of water for drinking, cooking, and washing. Large quantities of water treatment chemicals were needed to reduce cyanobacterial concentrations to a level where even a reduced throughput could be maintained and aluminium levels in the final water eventually rose to a level considered unfit for consumption. The water had also become unacceptable due to taste and odour. The water authority arranged for potable water to be transported as bottled water and in tankers to serve the human population. Fortunately, the very large cattle herd on the island at the time was able to continue to drink the loch water without apparent ill effect. No cyanotoxins were actually detected in the drinking water.

(ii) In July 2005, consumers of treated water supplied from the Loch of Boardhouse in Orkney complained of an earthy taste and musty odour. A visual check of the loch identified green growth around the loch consistent with an algal or cyanobacterial bloom. Analysis confirmed the presence of mixed cyanobacterial species, predominantly of Anabaena, resulting in high concentrations of the taste and odour compounds 2- methylisobornereol and geosmin. Cyanotoxin (microcystin) concentrations were below 1 microgram per litre. A temporary powdered activated carbon dosing plant was installed which improved the taste and odour of the final treated water. The cyanobacteria had virtually disappeared by mid-August. Approximately 4,000 consumers were inconvenienced by disruption to their water supply and were supplied with bottled water. A similar problem with cyanobacteria and geosmin tainting affected the taste and odour of water from the Glenfarg reservoir in 2006, and was managed using activated carbon dosing.

(iii) Cases of cattle, dog, bird and fish deaths have occurred in Scotland in which cyanotoxins have been implicated as a cause or a contributory factor, based on toxicological and veterinary investigations. Further occasional reports of animal deaths attributed by their owners to contact with cyanobacterial scums have occurred. However, objective evidence is not always available to confirm an association with toxin exposure. In summer of 2003 there was good evidence to suggest that the deaths of two dogs in Fife were associated with ingestion of cyanobacterial sludge (detached shoreline mats) at Town Loch near Dunfermline. Restrictions were imposed and were supported by ongoing monitoring. Shoreline deposits were safely removed and disposed of. Another dog death on Shetland reported to SEPA in 2006 was investigated and cyanotoxin analysis suggested that toxin exposure was a strong candidate as the cause of death. However, problems with cyanobacterial recognition, incomplete knowledge of exposure circumstances, and post-event sampling and investigation, continue to constrain clear identification of the role of cyanobacteria and their toxins in some cases of suspected cyanobacterial poisoning.

(iv) The Scottish Environmental Incident Surveillance System (SEISS), operated by HPS includes a report from SEPA of an incident in August 2010 involving a canoeist who reported symptoms (not specified) following contact with water at the James Hamilton Heritage Park Loch in East Kilbride. Cyanobacteria (Microcystis sp) were detected in excess of 100,000 cells per ml., but no record of measured toxin levels was reported.

3.9 Although SEPA is able to provide an analytical service and advice, it does not operate a routine cyanobacterial national monitoring system, hence objective evidence is difficult to obtain to confirm specific associations between ill health and cyanotoxin exposure. Moreover, as the record of samples analysed for cyanobacteria tends to be patchy, it is difficult to relate trends in incidents involving cyanobacterial blooms with evidence of climate change. In a recent period 2008 - 2010, between 130-181 samples were received by SEPA for analysis each year, of which between 35%-54% were reported to contain cyanobacteria at concentrations exceeding 20,000 cells/ml. A number of lochs continue to be perennial 'hot spots' containing cyanobacteria at high concentrations throughout summer and into autumn. Cyanobacteria (Aphanizomenon flos-aquae) were detected in excess of 100,000 cells per ml. and resulted in the closure of the Great Scottish Swim event in August 2010 in Strathclyde Loch, North Lanarkshire.

3.10 Surveillance by HPS using the Scottish Environmental Incident Surveillance System (SEISS), from 2002 to 2010 identified a total of 423 water related cyanobacterial incidents reported across Scotland, ranging from 125 in 2002 (when surveillance started) to a minimum of 14 in 2009, increasing to 48 in 2010. Reporting to SEISS is however, likely to be selective and incomplete.

3.11 Cyanobacterial blooms, mats, and scums are inherently complex (Paragraph 3.2) and assessment of the associated risks to public health is not straightforward. Such assessments should therefore take account of specialist advice (Annex C). Where advice is not immediately available, action of the kind described later may still be appropriate.

Contact

Email: Janet Sneddon

Back to top