Evidence based review of Legionella elimination in building water systems

Legionnaires disease can be acquired through exposure to Legionella pneumophila, a gram-negative bacteria ubiquitous in both natural and engineered water systems. Over the years, a number of disinfection techniques notably, chlorination, ozonation, thermal, UV and copper-silver ionization have been employed across different kinds of engineered water systems with diverse measures of success. Available evidence portends, most of the aforementioned techniques often have to be combined to achieve long-term efficacy. Remarkably, albeit the extensive research and reportage on Legionnaires outbreak in the developed world, very few studies have been carried out with regards to Africa. We reviewed existing literature on the application of the aforementioned techniques in buildings. Our study concurs with earlier studies; most of the disinfection techniques will have to be combined to achieve the desired efficacy. We found very scanty studies on Legionella or reportage of its outbreak within Africa. Our study also found very little in terms of any of the techniques been applied with the specific aim of reducing Legionella proliferation in engineered water systems within Africa. This is alarming, especially, on a continent where several communities have little or no access to quality water and healthcare. In light of the above, stronger measures such as sensitization, properly managed water distribution systems, as well as policies aimed at enforcing national and international guidelines on Legionella control is recommended.


INTRODUCTION
Legionnaire"s disease first commanded attention in 1976 during an American Legion convention in Philadelphia (Swanson and Hammer, 2000).Whilst initial documentation on Legionnaires was related to cooling towers, a study by Tobin et al. (1981) was one of the earliest to demonstrate Legionella could be found in water distribution systems of hotels and hospitals.Their study demonstrated infected water systems could be linked with cases of Legionnaires in the absence of airconditioning equipment.Legionnaires is largely caused by Legionella pneumophila a pathogenic bacteria for free living, ubiquitous, freshwater, and soil amoebae of the genera Acanthamoeba and Naegleria (Rowbotham, 1980).The disease is characterized by pneumonia, often afflicting the elderly or immunosuppressed individuals (Parry et al., 1985;Kümpers et al., 2008;Hilbi et al., 2010).Although Legionella is a genetically diverse species, L. pneumophila, one of the many Legionella species, is common in natural and engineered water systems and its single serogroup, L. pneumophila Sg1 accounts for almost 84% of Legionnaire's cases reported world wide (Cold Spring Harbor Laboratory, 2008).Though several disinfection modalities have been tested and approved over the years, the dilemma in choosing an efficacious technology still lingers.The application and draw backs of most of the contemporary techniques employed continue to be evaluated whilst new techniques or devices aimed at Legionella control in water systems are still being experimented.Perhaps, this is because each of the techniques presents distinct characteristics (Marchesi et al., 2011).Over the last two decades, most of the notably documented techniques (e.g.superheating, copper-silver ionization, ozonation, Ultra violet light and hyper chlorination) have undergone evaluation and review in literature elsewhere (Kim et al., 2002a;Campos et al., 2003;Lin et al., 2002;Muraca et al., 1987;Muraca et al., 1990;Lin et al., 1998Lin et al., , 2011)).Despite the above, these reviews along with documented efficacies often do not include Africa.While Legionella sampling in water systems is almost a routine in developed nations.Legionnaires disease is barely a subject in most African countries.Most patients are sub optimally treated for other diseases such as Tuberculosis in the face of likely symptoms.Owing to the fact that Legionella is ubiquitous in both natural and manmade water systems (CDC, 2016), the aforesaid is particularly disturbing considering the fact that piped-in water is non-existent in the poorest 40% of households in rural sub Saharan Africa (UNDESA, 2014).In this study, we review literature on some of the most commonly used techniques in Legionella control in engineered water systems.Keen attention is paid to water quality as well as the application of the respective techniques on the continent.

METHODS
The documented efficacy, advantage, disadvantage and effect on w ater quality of a number of disinfection modalities aimed at Legionella control and elimination w as carried out.Thermal disinfection, hyper-chlorination, copper-silver ionization, ozonation and UV light w ere selected, taking into consideration similar review s of the afore listed along w ith other techniques (Kim et al., 2002a;Campos et al., 2003;Lin et al., 1998).With the exception of copper silver ionization, the techniques w ere also selected based on existing evidence of their application across Africa for regular w ater treatment.Copper silver ionization w as how ever added based evidence of its "positive" review s in other literature.Finally, the review also sought to put together, documented efficacy in hospitals, hotels and w ater distribution netw orks; uncharacteristic of majority of such review s.

CLASSIFICATION OF WATER DISINFECTION TECHNOLOGIES IN LEGIONELLA CONTROL
Water disinfection employed in Legionella control is varyingly classified.In certain literature, the techniques are classified as "localised methods" or point of use e.g.ozonation, UV lights and "systemic" e.g.thermal disinfection or copper silver ionization (Peiró Callizo et al., 2005).Other literature also groups the techniques as systemic or emergency disinfection (Lin et al., 2011).Emergency techniques such as thermal disinfection often employed during outbreaks have been reported to lack residual effects over longer periods (Stout et al., 1986;Chen et al. 2005;Mouchtouri et al., 2007), w hile disinfection techniques such as copper-silver ionization w hich have been linked w ith long term Legionella control are still under evaluation albeit being implemented (Cachafeiro et al., 2007).

Method
Water temperature greater than 60°C inhibits the survival 19 and grow th of L. pneumophila in vitro (Muraca et al., 1987(Muraca et al., , 1990;;Campos et al., 2003).Although one of the earliest in Legionella control, available literature suggests it is ineff icacious unless repeatedly applied alongside faucets chlorine disinfection (Stout et al., 1986;Mouchtouri et al., 2007).Nonetheless, the temperature range for proliferation (20ᴼC to 43ᴼC) as w ell as inactivation (>44ᴼC, < 20ᴼC) are w ell documented (Konishi et al., 2006;Schulze-Robbecke and Buchholtz, 1992).Though varying forms of implementing this technique exists, the basic principle involves elevating the temperature of w ater in a storage tank above 70ᴼC w hile ensuring temperatures at distal fixtures are not below 60ᴼC.Distal fixtures may then be run at respective time intervals for days and monitored in accordance w ith required standards or regulations.

Characterization of efficacy, advantages and disadvantages
In a review , Campos et al. (2003), described thermal disinfection as a temporal control strategy as bacteria colonization is often evident months after implementation.Perhaps, due to "repository biofilm" w hich provide protective mechanism for survival and recolonization, complete elimination of Legionella in w ater systems remains farfetched (Mouchtouri et al., 2007).The above is evident in a study by Steinert et al. (1998), w ho observed re-colonization of tw o Legionella strains, three months after implementing the technique (70°C) in a hospital w ater system (Figure 1).A study by Chen et al. (2005), also observed regrow th"s, tw o months after implementing the technique at a medical centre (Figure 1).Other studies have also reported insignificant reduction in Legionella contamination counts w ithin the first months of its implementation (Marchesi et al., 2011).In terms of its edge over other disinfection techniques, thermal disinfection does not require any special equipment"s aside the use of devices that register w ater temperature.It can be implemented expeditiously in cases of outbreaks or emergencies.On the other hand, the possibility of scalding and the amount of w ork involved in monitoring distal sites could be time consuming.Challenges w ith its application in larger buildings e.g.hotels or hospitals w here stable temperature may be difficult to attain along entire w ater netw orks is also notew orthy (Chen et al., 2005;Mouchtouri et al., 2007).In view of the aforestated, routine implementation could be challenging especially as such facilities (Hotels or hospitals) w ill have to be unoccupied at best.Finally, dead legs, operation at deliberate low temperatures per concerns of scalding, as w ell as its implementation in old w ater distribution netw orks are also notew orthy inhibitions (Groothuis et al., 1985;Plouffe et al., 1983).

Effect of therm al disinfection on w ater quality and its application in Africa
The overall effect of thermal disinfection on w ater quality remains conspicuously better than other w ater disinfection technologies especially due to lack of biocidal effect of chemicals as may be the case in other disinfection modalities.Minus "circulation" through the w ater distribution netw ork along w ith "special devices" that monitor w ater temperature at distal sites, the technique is almost synonymous to boiling w hich remains perhaps, one of the default household w ater treatment methods in the developing w orld (Rosa and Clasen, 2010).Evidence of easy contamination after cooling coupled w ith economic and environmental unsustainability in its application (Luby et al., 2000;Gilman and Skillicorn 1985;Parikka, 2004) perchance, remains the defining reason w hy the technique is not used as a routine w ater disinfection modality.Interestingly, boiling remains relatively rare in Africa (4.5%) even though other African countries such as Uganda (39.8%) and Zambia (15.2%) report significantly high rates of its application (Rosa and Clasen, 2010).Nonetheless, w hereas boiling w ill serve the same purpose of Legionella inactivation (> 70°C), evidence of literature on boiling or heat treatment specifically aimed at Legionella control is almost nonexistent.Boiling is how ever largely documented as a household w ater treatment method.

Method
Hyper chlorination involves the addition of chlorine to w ater w ith existing chlorine residue.So far, tw o w ays of implementing the technique has been documented: (1) Shock hyper chlorination, w hich involves the pulse injection of chlorine into a given w ater system (range of 20 to 50 mg/l), and (2) continuous chlorination w hich involves continuous injection of chlorine in a form of gas, liquid or solid.The treated w ater is later drained and fresh w ater introduced to reduce chlorine concentration to about 0.5 to 1 mg/l.Typically, the installation involves, mounting chlorine injectors in highly pressured w ater supply lines w ith venturi orifice w hich create a vacuum; consequently, draw ing chlorine gas into the w ater stream.Overall, adequate contact time and mixing is required to achieve desired efficacy.

Characterization of efficacy, advantages and disadvantages
Although the technique represents the decisive treatment for bacteria control and elimination in most w ater systems most especially in old w ater distribution netw orks (Schoenen, 2002;Orsi et al., 2014), Legionella is reported to be more tolerant to chlorine than other bacteria including Escherichia coli (World Health Organization (WHO), 2007).Evidence of the presence of biofilm is reported to be a key element in its ability to offer such resistance.As described by Cooper and Hanlon (2010), continuous grow th and survival of L. pneumophila biofilms (1-2 month old at 50 mg/L over an hour) w ere observed despite high levels of chlorine.An earlier study by Kuchta et al. (1983) also show ed different Legionella strains w ere more resistant to chlorine compared to other coliform bacteria; even more resistant at low chlorine dosages.In terms of its long term efficacy, an environmental surveillance study by García et al. (2008), show ed L. pneumophila sero-group 1 survived despite successive episodes of hyperchlorination for 10, 5 and 17 years in a hotel, fishing boat and hospital w ater system respectively.By far, available studies continue to suggest, hyper chlorination appears to be only effective in its initial stages w ith bacteria levels often exceeding pre-treatment levels after a short w hile (Marchesi et al., 2011) (Figure 2).There is also evidence of facilities (e.g.hospitals) sw itching to other disinfection techniques having previously implemented hyper chlorination (Lin et al., 2011).Thus, hyper chlorination is inadequate for long term disinfection and only suitable for emergencies or during commissioning (DOH, 2009).Coupled w ith the above, the after effects of its application e.g.corrosion of pipping material over longer periods and the associated health effects of chlorine by-products such as trihalomethanes linked to cancer is also notew orthy (Gopal et al., 2007;USEPA , 2001;Miller, 2008;DOH, 2009;Mcdonnell et al., 1999).Finally, it has been suggested varying residual concentration due to changes in w ater flow rates may result in inadequate residual quantity at distal fixture, this, may in the long run have no significant effect on bacteria present (DOH, 2009).

Effect of hyper chlorination on w ater quality and its application in Africa
Although the use of chlorine eliminates w ater odour (Wajon et al., 1988), w ater quality is reduced (organoleptic and chemical) often, as a result of chlorine levels (>0.5 < 1.0 mg/L) (Orsi et al., 2014).In terms of health, an increased risk of bladder cancer also appears to be associated w ith the consumption of chlorinated w ater (Cantor, 1997) along w ith other illnesses such as asthma and dermatitis w hich have also been reported (Gorchev and Ozolins, 2011;Faw ell, 2003).Within Africa, available literature show s hyper chlorination is used extensively.Albeit reports of its effect on human health in countries such as Algeria (Benhamimed and Moulessehoul, 2010), hyper chlorination has been documented as an emergency technique during a Legionnaires outbreak in South Africa (NICD, 2016).Its application in conflict areas or disasters zones e.g.Kenya during Cholera outbreaks, alongside the promotion of hygiene and safe food preparation has also been reported (UNICEF, 2011).More so there is documented evidence on its use in the control of Hepatitis E outbreak in northern Uganda.These, reports show , higher concentration of residual chlorine is often required to facilitate control of the virus (IFRC, 2008).Overall, the use of chlorine w ithin drinking w ater netw orks either as centralised or point of use is quite extensive in many African countries e.g.Angola, Cameroon, Ethiopia, Guinea, Kenya, Madagascar, Malaw i, Mozambique, Nigeria, Rw anda, Tanzania, Uganda, and Zambia (Peletz and Mahin, 2009;Walfer, 2013).

Method
Copper-silver ionization involves passing w ater through a device w hich applies low potential electricity to electrodes of silver and copper, consequently, dissolving and distributing smaller ions (Cachafeiro et al., 2007).The Copper ions (Cu 2+ ) present, form electrostatic bonds w ith the negatively charged cell w alls of microorganisms (Legionella inclusive) resulting in the disruption of cell w all permeability (Muraca et al., 1990;Lenntech, 2011).On the other hand, the silver ions bond w ith the DNA and RNA of the microorganism, leading to immobilization and cell death (Lenntech, 2011).Overall, the concentration of ions and the nature of the w ater system must be carefully evaluated, in order to achieve eradication of Legionella.

Characterization of efficacy, advantages and disadvantages
Despite the conventional argument that most disinfection techniques cannot be used in isolation, there is evidence to suggest, copper-silver ionization continues to gain popularity across the globe as an effective and safe technique so long as ions are monitored according to requisite standards (Cachafeiro et al., 2007).In terms of its effect, its application leads to lysis and cell death by "distortion of cellular permeability coupled w protein denaturation" (USEPA, 2001).Its ability to kill as opposed to suppression or control of bacteria is also w ell documented (Lin and Vidic, 1996;Lin et al., 1998;DOH 2009).With respect to other microorganism aside Legionella, States et al. (1998) argue, copper silver-ionization is ineffective against amoeba and other non-Legionellaceae bacteria.Nonetheless, there is evidence the technique is efficacious against other w aterborne pathogens such as Pseudomonas aeruginosa, Stenotrophomonas maltophilia and Acinetobacter baumannii (Huang et al., 2008).Its application in both hot and cold w ater point of entry is also reported (Chen et al., 2008).
A survey by Stout and Yu (2003) across 16 US hospitals over a 5 to 11 year period, observed copper-silver ionization w as most effective compared to other previously implemented disinfection techniques.They reported 0% positivity for Legionella for 50 and 43% of the hospitals as of 1995 and 2000 respectively.This w as against 7 out of 15 of the hospitals reporting 30% Legionella positivity at distal fixtures prior to application.Only a single incidence of Legionnaires w as reported after the technique had been implemented.A study by Rohr et al. (1999), also show ed copper silver ionization achieved a reduction in Legionella counts from 40,000 to 7 cfu/L during the first year of disinfection (average silver ion concentration <10 ug/L).Nevertheless, an increment of 10,000 cfu/L w as observed in the third year over a four-year period (Figure 3).Thus, the study concluded, the effect of the technique w as only suitable for short-term applications and influenced by factors such as w ater temperature and content of silver ions in w ater (>45°C, up to 66.6°C).
Follow ing the concentration of nontuberculous mycobacteria and legionellae for three and four years respectively on the influence of the technique, Kusnetsov et al. (2001), reported electronically released copper and silver ions w ere inefficacious at taps and show ers of a hospital w ater system despite an increase in ion concentration in the circulating w ater.Mòdol et al. (2007) in an evaluation of the impact of the technique after the implementation of other techniques, reported a significant decrease in Legionella counts as w ell as reduction in nosocomial legionellosis (2.45 to 0.18 cases per 1000 patient discharges) Overall, although advantages such as ease of installation and maintenance have been reported, difficulty in monitoring residual levels, formation of insoluble hydroxides and the need for periodic cleaning of electrodes are also notew orthy (Lin et al., 2011;Campos et al., 2003).A review by Kim et al. (2002a), on studies conducted by other authors suggests the use of metal ions, "either copper alone or both copper or silver" is a good option for Legionella control especially in hot w ater recirculation systems.Nonetheless, its long term effect is yet to be listed as a major advantage.There are also studies on its application in sw imming pools although peer review ed literature on its efficacy is rarely been reported (Abad et al., 1994;Beer et al., 1999).

Effect of Copper-Silver Ionization on w ater quality and its application in Africa
According to Mcdonnell et al. (1999), the oral consumption of ions through the use of copper-silver ionization is often limited as ions are only added to recirculating lines in hot w ater systems.Largely, copper-silver ionization poses very minimal potential safety hazards compared to other chemical disinfection techniques even though its continuous use is how ever barely encouraged (Zheng et al., 2012).In a study on the technique, Lin et al. (2002), found variations in w ater quality as a cause of its inefficacy as on-site pH measurement w ere high (8.9).There are also reports on the effect of w ater type (hard or soft) on the efficacy of the technique (TARN- PURE Ltd, 2016).Within Africa, there seem to very little or nonexistent literature on the application of the technique in Legionella control.There are how ever some studies related to copper and silver metals.In Sw aziland, Varkey (2010) reported on the antibacterial effect of copper and silver metals on Coliform and E.
Coli through immersion of the metals in w ater.Paper filters containing AgNP and CuNP w ere tested on w ater sourced from contaminated streams in Limpopo South Africa resulting in marked reduction in coliform bacteria and E. coli counts and in the incorporation of a copper mesh in the design of clay pot w ater filters (CPWFs), E. coli and coliforms concentrations in contaminated w ater w ere reported to have reduced markedly (Varkey and Dlamini, 2012;Dankovich and Gray, 2011).

OZONATION Method
Ozone is an unstable potent biocide and oxidizing agent w hich must be produced on site due to its short half -life especially as rising temperature can results in increase in decomposition (NRC, 1980).Ozone can be produced in a number of w ays notably phosphorus contact, silent discharge, photochemical reactions, and electrochemical reactions (Wei et al., 2016).Generated electrically by exciting oxygen to a tri-atomic state , ozone can be injected to an allow able quantity but should be maintained at a residual dosage.

Characterization of efficacy, advantages and disadvantages
Comparably, ozone is more effective in L. pneumophila control compared to other oxidizing biocides (Dominique et al., 1988).In a review , Kim et al. (2002b) reported there is an appreciable use of ozone across Europe although its better complimented w ith other disinfectants such as chlorine.In light of it being unstable, ozone does not stay in w ater sufficiently long to provide the residual effects often required in the control of pathogenic bacteria (Kim et al., 2002b;WHO, 2007;Blanc et al., 2005).Ozone is reported to somew hat more effective at a low er temperature, w ith a potential of faster destruction of microorganisms at higher pH (Botzenhart and Tarcson, 1993) In terms of efficacy, Edelstein et al. (1982) documented, ozone reduced L. pneumophilla counts in one w ing of an unoccupied hospital w ith average ozone concentration of 0.79 mg/L.Blanc et al. (2005) how ever reported no significant reduction in Legionella counts after the application of just ozonation.Appreciable reduction in Legionella counts w ere how ever observed after the complimentary implementation of thermal eradication (Table 1).
Overall, ozone generation consumes a significant amount of energy and thus require a lot of monitoring in order to ensure pow er is optimized w hile achieving disinfection targets (Casey et al., 1998).Additionally, its requirement of onsite generation and the use of special equipment as opposed to other disinfection modalities make it less favourable.Based on its quick decomposition characteristics and its lack of residual effect, a grow ing body of review s and independent studies show ozonation is best implemented as a secondary disinfection technique (Campos et al., 2003;Kim et al., 2002a;Blanc et al., 2005).Reports of of its instantaneous effect in bacterial inactivation, rapid decomposition and on site generation are how ever listed as major advantages (Casey et al., 1998;Campos et al., 2003).

Effect of ozonation on w ater quality and its application in Africa
A varying body of literature show s, ozonation significantly improves w ater quality either in building w ater systems or even w hen used in aqua culture (Davidson et al., 2011;Von Gunten, 2003).A study by Von Gunten, (2003) reported ozone as an excellent disinfectant in the inactivation of protozoa compared to other very conventional disinfectants.In terms of w ater quality, ozonation has been documented to result in a significant increase in ultraviolet transmittance as w ell as a reduction in carbonaceous BOD and colour (Davidson et al., 2011).By combining ozonation w ith UV treatment, available literature has show n the potential for minimizing bromate as w ell as the oxidation or micro pollutants in w ater (Meunier et al., 2006).Practically, studies have also show n the application of ozonation for disinfection has a positive impact on distribution w ater system, reducing both levels of disinfection by products and complaints about the taste and odour of w ater (Dyksen et al., 2016).Essentially, it is evident problems in w ater quality that are often addressed w ith peroxide, chlorine etc. can also be addressed through ozonation (Eagleton, 1999).Within Africa, there is extensive evidence on the application of ozonation although much of its use is not directly aimed at the elimination of L. pneumophillia.The post-ozone plant at Wiggins w aterw orks w hich w as commissioned in 1984 in South Africa is one such example.The purpose of the plant has been to properly disinfect w ater by eliminating viruses in raw w ater pumped from the polluted low er Umgeni river (Rencken, 1994).There is also evidence on the application of ozonation in industrial w aste w ater and drinking w ater treatment plants in parts of South Africa ( Van der Merw e et al., 2012).Its use at the Midvaal w ater company Durban Water Recycle Plant and Magalies Water plant are also documented (Rajagopaul et al., 2008;Morrison et al., 2012).

Method
UV disinfection can be applied in tw o w ays: (1) Positioning UV lights at specific points w ithin a given w ater distribution system that service a designated area or (2) installation of a UV -sterilizers for the disinfection of incoming w ater.Monochromatic low pressure UV lamps and polychromatic medium pressure lamps are some of the lamps used (Oguma et al., 2004).

Characterization of efficacy, advantages and disadvantages
In terms of efficacy, available literatures show s UV radiation kills bacteria by hampering DNA replication through the production of "thymine dimers" (Gavdy and Gavdy, 1980 as cited in Hambidge, 2001).Typically, there are different types of UV lamb systems although low pressure UV bulb systems are almost an industry standard supplying monochromatic irradiation specific to 254 nm w avelength (Summerfelt, 2003).Its efficacy against other microorganisms is documented elsew here (Harris et al., 1987;Hijnen et al., 2006;Chevrefils et al., 2006).On the other hand, the use of UV alone in a persistently colonised hospital w ater distribution system ("point of use application") show ed it w as ineffective in the elimination of Legionella (Liu et al., 1995).The concurrent application of other disinfection techniques e.g.Superheat/Flush and shock chlorination how ever achieved some reduction in Legionella counts.In a study by Franzin et al. (2002) a reduction in Legionella counts (L.pneumophila sero-group 3) w as achieved even at distal sites using UV light.The authors therefore concluded UV disinfection could be suitable in small areas of w ater system.In Hall et al. (2003), UV w as efficacious in preventing Legionella contamination over a 13 year period.Overall, despite reports of the extensive application of UV in both cold and hot w ater systems (Liu et al., 1995;Kim et al., 2002a), its limitations include maintenance against the formation of scales, likelihood of malfunctions, lack of residual beyond points of application and turbidity (Muraca et al., 1990;USEPA , 2001;Campos et al., 2003).The use of filters in minimizing the accumulation of scale on UV quatz sleeves has also been reported (Liu et al., 1995).In terms of other advantages, UV light is easy to install (Campos et al., 2003).Conspicuously, continuous monitoring as may be the case in other disinfection modalities is also limited.Finally, although the energy used in typical UV disinfection system in an average size home is comparable to the energy used by a 40 w atts bulb (UV Dynamics, 2016), the energy cost associated w ith technique cannot be ignored.

Effect of UV irradiation on w ater quality and its application in Africa
Overall, the application of Ultraviolet light (UV -Lamp) is an established and increasingly popular alternative to chemicals for the disinfection of w ater distribution systems.Interestingly, the quality of w ater is one of the many defining factors in its application (Wright and Cairns, 1998).Amid it being a cost effective disinfection technology, there is evidence, UV provides no residual effect in w ater tow ards the protection against post treatment contamination (Clancy et al., 2000;Said and Otaki, 2013).A study by Choi and Choi (2010) has show n the technology has an effect on dissolved organic matter structure (DOM) in distribution systems including an increase in biodegradability (Frimmel, 1998;Drikas et al., 2004).By far, the aforementioned remains the most documented effect on w ater quality across an existing body of research (Kruithof et al., 1992;Oppenheimer et al., 1997).Within Africa, a study on a device (UV light 254 nm) that utilizes the technique to disinfect community drinking w ater demonstrated efficacy for close to 4.5 months; delivering w ater that meets WHO and USEPA bacteria standards (USEPA, 2001;Gadgil et al., 1998).At the Durban Metro Water and hospice for infants, the application of a UV unit also achieved a reduction in E. coli and total coliforms concentration per local standards (Gadgil et al., 1998).Largely, there is no literature on the application of this technology specifically aimed at Legionella control.There is how ever extensive publication on its application as a regular w ater disinfection technology coupled w ith the proliferation of numerous devices or units that utilize the technology for disinfection (Brahmi and Hassen, 2014;Gadgil et al., 1997).Perhaps, Solar Water Disinfection (SODIS) w hich is appears to be highly patronised due to the conspicuously low cost implication associated w ith it compared to the conventional use of UV lamps, positions the use of UV-light as somew hat unpopular.Reports on the efficacy of SODIS and application in developing nations is been reported (Mosler et al., 2013;Murinda and Kraemer 2008;Altherr et al., 2008;Conroy et al., 2001).Nevertheless, its limitations such as the need for sufficient solar radiation, relatively clear w ater and difficulty in treating large volumes cannot be overlooked (Mintz et al., 2001).

DISCUSSION
This work provides further knowledge on an already explored area of comparing different water disinfection techniques in Legionella control and elimination.Additionally, the study also attempts to identify cases of Legionnaires reportage in Africa as well as the application of some selected disinfection techniques in Legionella control within the continent.The overall aim of the fore going, is to provide an update on some of the most commonly used water disinfection techniques employed in Legionella control so as to provide a pool of evidence based options (locally and internationally) in the event of Legionnaires outbreak within Africa.
Globally, reports on the outbreak of Legionnaires occur almost every year; often reported by the developed countries.Countries such as the United States through the Centre for Diseases Control (CDC) are known to document almost 5,000 cases of Legionnaires" annually (Dooling et al., 2015).This is no different in Europe.Wellregulated surveillance programmes across almost 35 countries within Europe, exist with the mandate to collect and provide information with regards to the disease (Heuner and Swanson, 2008).The European Surveillance Scheme for Travel Associated Legionnaires Disease now known as EWGLINET, is known to be instrumental in recommending standards for Legionnaires surveillance (Ricketts and Joseph, 2005).The case is however very different in the African context.With the exception of a handful African countries e.g.South Africa, there is very little in terms of institutions or organizations aimed solely at monitoring Legionnaires on the national scale or on the continent at large.The lack of statutory notification of Legionnaires, coupled with clearly defined health based standards is evident in the lack of academic literature or studies related to Legionnaires on the continent, as may be the case in Europe.For instance, there are very few studies or reports on particular water disinfection techniques aimed at Legionella control or evidence of them being applied during an outbreak as shown in this review.Perhaps, political importance, clinical impact or rareness of disease (Heuner and Swanson, 2008) that are motivating factors in paying attention to most infectious diseases is minimal.
Unfortunately, the argument cannot be made that reportage or studies related to Legionella are very little because proliferation, infection or outbreaks rarely occur.L. pneumophila is known to grow and survive over a wide range of temperature (20, 40 and 50°C) (Rogers et al., 1994;Konishi et al., 2006) and these temperatures are typical in a greater part of sub Saharan Africa.Moreover, there is evidence to suggest temperature increase over land regions across the continent is consistent with anthropogenic climate change (Aalst et al., 2014), thus, the implication of the foregoing are favourable temperature for bacteria proliferation in already precarious water distribution networks across the continent.More so, at the root of the many challenges with water supply are the poor maintenance and servicing culture of already deteriorated pipe networks (Marin, 2009).Many of the water distribution networks are faced with massive leakages, irregular supply and low water pressure; these factors create avenue for easy water contamination (Marin, 2009).Implicitly, much of the water prior to being consumed at distal sites are already contaminated with pathogenic bacteria; a challenge that must be avoided from the onset as a step in achieving Legionella control (Borella et al., 2005;Rangel et al., 1999;WHO, 2007).
On the other hand, evidence of a single most consistent and permanent method at Legionella control is barely been established globally.Our review noted most of the disinfection techniques often had to be combined to achieve long term efficacy; consistent with literature elsewhere.For instance, while techniques such as thermal disinfection record some measure of efficacy, the lack of residual effect days after application, often require the application of other techniques e.g.chlorination.Thus, the application of such methods have demonstrated efficacy only in the short term.Additionally, it would also seem somewhat challenging to flush entire water systems of heavily occupied hotels or hospitals especially as occupancy and monitoring at distal sites are significant in achieving success.In terms of Chlorination, evidence of re-colonization as shown in this review, suggests it can arguably be placed under the category of short term or emergency techniques notwithstanding its residual effect.Moreover, the role of biofilm and Legionella resistance to Chlorine as reported in other literature (Kuchta et al., 1983;Cooper and Hanlon, 2010) imply, larger doses of chlorine are often required to achieve complete elimination.In view of the fact that chlorination is already used extensively as a common water disinfection technique within Africa (Whitacre, 2010), it seems at best, one of the most accessible disinfection techniques to be employed in the event of an outbreak.Chlorine dosage will however have to be based on local regulations or known international standards.
Copper silver ionization appears the most efficacious compared to the other techniques in this review.In addition to its ease of installation, evidence of its application even in hot and cold water systems with keen consideration to ion quantities, provides the basis for arguing, the technique can be employed during outbreaks and over longer periods.Additionally, its ability to kill as opposed to suppression, proves copper-silver ionization will be the most suitable disinfection modality in heavily colonised water distribution networks.The afore highlights, coupled with findings in other literature (Marchesi et al., 2011;Lin et al., 1998) therefore positions copper silver ionization, conceivably as one of the best alternatives for Legionella control.
With Ozonation, albeit, its extensive use within Europe and Africa as shown in this report, the need for onsite generation and special equipment could be challenging in most domestic installation or small public buildings in rural areas; heavily colonised by bacteria.Additionally, its lack of residual effect, energy consumption and need for monitoring, implies the technique will be difficult to apply in colonised water systems in places with less manpower or electricity.Nevertheless, its rapid destruction of microorganism at higher pH (Botzenhart and Tarcson, 1993) along with its ability to compliment other disinfection techniques, positions the technique as a viable option in Legionella control within Africa.
Finally, on the grounds of challenges such as maintenance against the formation of scales, lack of residual effect, use of filters for scale prevention and the need to combine with other disinfection techniques as already evidenced in this report, UV cannot be considered a "singular disinfection method".Albeit its ease of installation, the lack of evidence of its efficacy in large-scale water systems, suggests UV should be considered in less colonised water systems or systems with little water volume, where greater water quantity can be exposed at point of contact.On the hand, the use of Solar Water Disinfection (SODIS) in Africa as suggested in the earlier part of this study will serve the greater advantage of bacteria control in rural areas especially in the absence of the commonly "advanced technologies" in water disinfection.A foreseeable challenge to this measure will be its inability to be implemented in large scale water systems where treating large volumes of water will pose a challenge (Mintz et al., 2001).
Overall, while all the aforementioned techniques have their respective disadvantages, their proven reliability across different types of buildings and water systems, suggest they can all be described as efficacious in one way or the other.Their implementation will therefore vary on a number of factors such as cost, ease of installation, maintenance, age and design of plumbing system, availability of electricity (for certain techniques), as well as the measure of colonization of the water system under consideration.An overall consideration should however be the prevention of contamination from the onset as reported elsewhere as well as effective monitoring of water system towards the identification of colonization at the early stages.The need for sensitizing the general population about Legionnaires, symptoms as well as factors that can lead to contamination in water distribution networks is also important within Africa.

Figure 1 .
Figure 1.Comparison of the heat resistance of L. pneumophila serogroup 1 (SG1) and serogroup 2 (SG2).The heat resistance w as determined by plotting the number of survivors (CFU/ml) versus period of exposure at 60°C.Data are expressed as the means of 5 independent experiments.Source: Steinert et al. (1998).

Figure 2 .
Figure 2. Short duration of superheat and flush failed to eliminate the site positivity for Legionella in patient w ard and intensive care units (ICUs).Source: Chen et al. (2005).

Figure 3 .
Figure 3. Legionella pneurnophila counts (cfu/L; shaded bars) and w ater temperature (°C; w hite bars) in hot w ater system of a university hospital, w ith use of silver-copper ionization in the year before ionization, 3 years w ith first ionization unit (Section A), and 1 year w ith a new ionization unit (Section B; fourth year).Box plot: Median, 25 to 75th percentile, minimum to maximum.Source: Rohr et al. (1999).

Table 1 .
Water and sw ab samples obtained before control measures, during ozonation, and during ozonation and increased temperature (65°C).