Introduction
and "historical" outline of the disease's outbreak
Legionella pneumophila was first isolated as an infectious bacterial
agent during an epidemic of acute pneumonia, which broke out among
ex-servicemen participating in a meeting of the "American
Legion" held in Philadelphia in July 1976. Legionellosis
is also called "legionary's disease" due to this unexpected
breakout of the epidemic, which numbered 149 cases of serious
respiratory disorders and recorded 29 deaths in less than fifteen
days.
The source of the contagion was the air conditioning plant in
the hotel where participants had stayed. They were contaminated
by inhaling the aerosol formed by water polluted by the bacteria.
(Fig. 1) The Legionella remained anonymous till
1976 due to its special and unique ecological requirements in
the natural environment where it lives in symbiosis with Protozoa
in the biofilm. It is believed that Legionellae have always been
present in the natural environment and that man has coexisted
with them with no special drawbacks. Only recently, due to the
technological progress, which has provided the means for their
"proliferation" and transmission to man through aerosol,
have these water germs been able to express their pathogenicity
through the extended bacteria-human body interface created by
the new uses of water. To quote Patrick Grimont from the Pasteur
Institute in Paris "We had to wait for the age of airconditioning
to realise the pathogenic power of the Legionella".
 |
Fig.1
The first Legionella epidemic occurred during the American
Legion's celebrations as in the picture. During a meeting
of ex-servicemen, the American Legion's Congress held in
a hotel in Philadelphia in July 1976 recorded 149 cases
with 29 deaths. The name legionellosis or legionary's disease
refers to this event. |
DEFINITION
Legionellosis has been defined as "mid-seasonal pneumonia".
It has two distinct expressions from a clinical and epidemiological
perspective: the legionary's disease and Pontiac fever.
Both are characterised at the onset by loss or reduction
in appetite, malaise, head ache and muscular pain. High
fever around 39-40 °C and shivers set in within 24-48
hours.There is often a non productive cough and many patients
complain of abdominal pain and diarrhoea.Initially the clinical
picture simulates pneumonia caused by mycoplasma, but progress
is later towards the most typical bacterial alveolar pneumonia. |
TREATING
LEGIONELLOSIS
- Since Legionella is an intracellular microorganism, only
antibiotics endowed with good intracellular penetration
properties are effective.·
- In vitro studies (cell culture) have proved that macrolides,
quinolones, rifampicin, tetracycline and the combination
of trimethoprim-sulfamethoxazole are active. Erythromycin
has long been considered the choice treatment, but it has
shown lesser antibacterial power than new substances that
are currently used. Besides, erythromycin is often at the
root of intolerance problems and drug interactions.
- Mocrolides, azithromycin, clarithromycin and quinolones
(ciprofloxacin or levofloxacin) have been currently considered
as choice treatment in daily medical practice. Treatment
has generally produced an improvement after 3 - 5 days.
The scheduled treatment's duration ranges between 10 and
21 days (21 days in case of immunodepression).
- In case of serious infections some authors advice combining
two drugs (either macrolide + rifampicin or macrolide +
quinolone).
- We must probably repeat that, even if in vitro tests can
show a certain degree of sensitivity, penicillin, cephalosporins,
imipenem and aminoglycosides are not adequate to treat legionellosis
as they lack intracellular action. |
According to the World Health Organization (WHO), legionellosis
ranks among the 30 new infectious diseases that have emerged in
the past twenty years in an evolutionary context that has been
defined the "powerful return" of infectious diseases.
The reported picture taken from Dr. Filice (1994) and modified
has a definition of pneumonia caused by Legionella pneumophila,
while Fig. 2 summarises routes of infection that
lead to intracellular replication and multiplication in pulmonary
microphages, hence to legionellosis. (Steinert et al. 2001) We
also deem it useful to summarise in a box some considerations
on treatment inspired by the recent paper "Legionella et
légionellose" published by the Federal Office for
Public Health in Bern (2005).
Fig.2
Legionella pneumophila routes of infection
|
|
Routes
of human contamination starting from water: a) Legionella
is present in biofilm in the environment and it multiplies
in Protozoa; b) Legionella uses a "technical vector"
to colonise human airways; c) After penetrating into macrophages
Legionella multiplies in their vacuoles causing the death
of the host cell through its necrosis and apoptosis. (Quoted
from Steinert and modified, 2001) |
Risk situations and their repeatability
The use of water for various current human activities and especially
for "relaxing" situations (decorative fountains, "bubbling
water" tanks, hydromassages), treatment (respiratory therapies,
incorrectly handled nebulizers) and air conditioning (cooling
towers, condensation water, humidifiers) enables to spread Legionellae
through aerosol, thus encouraging their transmission to man. The
main reservoir is water and no interhuman transmission (from person
to person) is known. Tables 1 and 2
summarise the most interesting cases of outbreaks of legionellosis
epidemics specifying the place where the epidemic broke out, the
date, the number of cases recorded and the cause.
Epidemiologically we notice risk situations, which have a common
denominator in their repeatability, such as for example the legionellosis
outbreak recorded in Holland early in 1999: an aerosol produced
at a flower exhibition contaminated 226 people causing 21 deaths.
A similar situation was already produced in 1990 in Louisiana
when a nebulizer, which sprayed water as aerosol on exhibited
products in a food store (to keep fruit and vegetables cool) caused
32 cases of legionellosis among customers with 2 confirmed deaths.
Legionellosis epidemics are hard to identify and their source
of contagion must be speedily located to be decontaminated.
Hence the need for special instructions for the decontamination
of polluted installations. Another particularly significant epidemic
broke out among patients in the Pompidou Hospital in Paris in
July 2001 (12 cases with 6 deaths). This ultramodern health facility's
plant was contaminated by Legionellae because two years had elapsed
from the installation of the new plant to the opening of the hospital.
It was the time required to form biofilm in the water supply network.
The study of the circumstances of the outbreak of epidemics can
be packed with lessons to avoid repeating the mistake in the management
of water in houses and other facilities visited by man. For example
according to the table's data esemon recent epidemics we can observe
that cooling towers have repeatedly been the source of Legionella
pollution. In a preventive perspective it has hence been deemed
that special attention must be given to these infrastructures.
Tab.
1 and 2 Legionellosis
epidemics from 1976 to 2001 and from 2001 to 2004 |
|
|
Bacteriological profile Legionella's
"identikit"
To understand the spreading of legionellae and their epidemiological
implications it is necessary to trace a microbiological profile
and to make an overall study of the Legionellaceae family's habitat.
Legionellae are small gram negative (length: 2-5 mm; width: 0,5
mm), asporogenic, aerobic bacilli (Fig. 3 Fig 4
and Fig. 5). Of the 50 species described many
have flagella, which ensure their mobility. The Legionellaceae
family comprises only one Legionella genus. The species currently
known are 50 and 71 serum groups have been defined.
Depending on the regions, between 70% and 90% of legionellosis
cases are caused by Legionella pneumophila, serum group 1. Twenty-one
other species are pathogens for man, but they rarely appear. The
best known are (in alphabetical order): L. anisa, L. bozemanii,
L. cincinnatiensis, L. dumoffii, L. feeleii, L. gormanii, L. jordanis,
L. longbeachae, L. micdadei (Pittsburgh Pneumonia Agent), L. oakridgensis,
L. parisiensis and L. tucsonensis. Difficulties found by bacteriological
analysis in isolating Legionellae through culturebased studies
clash with their ubiquitous presence in water environments that
have poor nourishing properties. In the laboratory this water
bacterium with few nutritional requirements has a metabolism that
is inappropriate for culture media, which are very rich in nutritional
substances.
Subsequently most Legionella species cannot be cultivated in media
normally used by bacteriology for other infectious germs. Some
species, which have been highlighted only through their association
with amoebae in co-cultures, have been called LLAP (Legionella-like
amoebal pathogens). The study of the biofilm through epifluorescence,
gene amplification and direct immunofluorescence are the most
effective methods to highlight these germs and they subsequently
provide greater analytical reliability than Legionellae's ubiquity
and the degree of contamination of water.
| Fig.3
Legionella pneumophila: colonies
|
|
Legionella
pneumophila colonies. Bacterial culture in a Petri dish
in BCYE medium (yeast extract, cystein). |
Fig.4
Immunofluorescence microscopy applied to Legionella pneumophila. |
|
Fig.5
Optical microscopy applied to Legionella pneumophila. Gram
stained. |
|
Legionella
pneumophila under the optical microscope. Gram staining:
small Gram negative bacillus (length: 2-3 µm, width:
0.5 µm). |
Legionellae's ecological requirements
The ecology of Legionellae is complex and their development in
the natural environment depends on various types of factors, which
are briefly analysed below with their respective implications.
Physical Factors
The optimal temperature for the growth of Legionellae is 25°-45
°C (hence warm water encourages germ proliferation). Their
survival involves an extensive thermal spectrum and they can be
isolated from water with temperatures ranging from 57° to
63 °C. Especially their resistance over 60 °C causes many
drawbacks due to proliferation in hot water plants and man's subsequent
contamination.
Note that other types of bacteria generally die at these water
temperatures; hence the lack of competition between germs too.
Legionellae have a good survival rate in acid and alkaline environments:
they can stand pH fluctuations between 5.5 and 8.1. We must stress
that water stagnation and sedimentation encourage the germ's growth.
Ultraviolet radiations (UV rays) inhibit their development. This
sensitivity enables the application of UV radiations to decontaminate
environments that are highly polluted by Legionellae.
Fattori chimici
Gli ioni d'Argento (Ag++) e Rame (Cu++) risultano inibitori, quindi,
secondo alcuni autori, le tubature in rame inibiscono la colonizzazione.
Per contro, i siliconi, il teflon, favoriscono l'adesione e il
caucciù (giunti e ranelle) favoriscono uno sviluppo intenso.
In genere, tutti i materiali che rilasciano in un ambiente liquido
delle particelle organiche utilizzabili microbiologicamente favoriscono
la colonizzazione da parte della Legionella (si pensi ai diversi
tipi di plastiche).
Biological Factors
In biofilm legionellae live along with other microorganisms that
colonise surfaces. Environmental biofilm is formed by a complex
microbial community ("consortium"), which comprises
both aerobic and anaerobic bacteria, Protozoa, Nematoda and fungi
(mycetes) in a niche that contains the metabolites (nutritional
contribution: i.e. essential amino acids). It also attenuates
the influence of physical and chemical variations (temperature,
oxygen, biocides). For example synergic proliferation with bacteria
belonging to the Pseudomonas genus has been observed, while there
is antagonism in growth with the Aeromonas genus. Legionellae
have the important characteristic of surviving and multiplying
with Protozoa in the natural environment (free and ciliated amoebae),
which can be a reservoir, a vehicle for human infection and even
a protection (through cohighly resistant cysts). The degree of
symbiosis varies depending on the species and the Legionella strain
(virulent and non virulent) and depending on the Protozoa species.
(Harf C. Monteil H. 1988; Swanson M.S., Hammer B.K., 2000).
Microbiological and epidemiological
studies conducted on Legionella
When a case of legionellosis is identified environmental samples
must be taken to locate the source in order to possibly limit
the epidemic. It is hence important to establish which infection
reservoir considered is the one really involved in the contagion.
The survey conducted to highlight the germ comprises two compartments:
clinical material taken from the patient and the water environment.
In fact, legionellae have been isolated from water environments
such as streams, lakes, spa water, reservoirs, wells, aesthetic
fountains, taps and shower nozzles.
Cases
Today molecular biology techniques called fingerprinting enable
us to view on special filters some significant DNA fragments of
organisms isolated from water samples. The profiles obtained are
real fingerprints. These specific features of each microorganism
enable a direct comparison between bacterial strains (Grimont
et al., 1995). The method used enables to compare some DNA parts
of bacteria studied. (Gaia V. e Peduzzi R., 2002) The case we
have chosen as an example of epidemiological approach through
genetic methodologies concerns a patient suffering from legionellosis
who had spent a few weeks in a spa. At the end of the treatment
he moved to his holiday home in Ticino. The following week he
developed pneumonia and was admitted to hospital. Legionellae
were isolated from the patient's bronchial aspirate culture. Then
samples of water were taken from his flat in Ticino and from the
hospital, besides water from the spa and from the hotel where
he had stayed. Water samples from the patient's flat and from
the hospital proved negative, while the presence of various strains
of Legionella was detected in water taken from both the spa and
the hotel. It was thus possible to establish that the patient
had been contaminated during his stay in the spa. In fact the
analysis performed by means of molecular typing of the isolates
enabled to highlight identical genetic profiles in the strains
found in the spa water, in the hotel water and the one isolated
from the patient. Methods based on genetic analysis can thus be
very useful for epidemiological studies on the spreading routes
of pathogenic germs from water to man and on the diversion of
pathogens.
Case histories and their evolution
If we consider the details of clinical analysis requests sent
to the NRC for Legionella, we notice that the antigenbased analysis
has increased from 30% in 1998 to 86% in 2004, thus becoming a
necessary choice in a few years. The diagnosis based on the urinary
antigen sets some limits to the epidemiological study as it does
not have the Legionella's clinical strain anymore to genetically
compare it with the environmental strain. (Graph 1)
Regarding diagnosis on man we must observe that the urinary antigen
facilitates diagnostic activities, but the epidemiological survey
required to define the germ's origin cannot be conducted if the
documentation of the case of legionellosis is not followed by
the analysis of the germ isolated from the patient. Hence the
analysis of the urinary antigen is an immediate advantage for
the patient, but not for the environmental survey, which envisages
a comparison between the clinical strain and the environmental
one.
The incidence of the legionary's disease in Europe per million
inhabitants: Denmark's 20 cases per million are considered the
"Gold Standard". According to official declarations
we can observe that with 23 cases per million inhabitants Switzerland
was slightly above the standard in 2002-2003. France and Holland
have a slightly lower incidence, while Spain declares 33.5 cases
per million inhabitants. Always concerning 2002, Italy declares
10.45 cases per million inhabitants.
The European average is around 8 cases/million inhabitants. (Graph
2) The evolution of the number of cases of legionellosis
in Switzerland from 1995 to 2004 reveals the speedy progression
at the close of the last century followed by a stable course with
a tendency to diminish in 2004 (Graph 3). Taken
as a landmark the European plot has a similar course. This reduction
is present both in cases caused by travels and in declared cases.
The two plots tend to diminish from 2003. A comparative study
of national numbers with the rest of Europe yields the following
prospect: the average rate of infection in the 24 European countries
is 3.9 cases per million inhabitants. In 1997 the mortality rate
was 10% with 1,360 cases of legionellosis. The Swiss average was
3.71 cases per million inhabitants (there were 10 cases / million
inhabitants only in 1998). The sources of contamination in Switzerland
are reported in Graph 4. It provides details
of cases caused by travels and by a stay in hospital and delcases
of disease caught in a community environment. People who are most
frequently affected are those with reduced immune defences, besides
men over 50 years have a statistically higher risk than the rest
of the population.
We must however observe that appropriate antibiotic treatment
comprising erythromycine or other macrolides or compounds belonging
to the group of quinolones or tetracyclines (see box on "Legionellosis
therapy") is effective in most cases.
| Graf.
1 Analyses performed
(in %) to diagnose legionellosis caused by Legionella ICM
and CNR between 1999 and 2004 |
|
Type
of analyses performed to diagnose legionellosis caused by
Legionella ICM and CNR between 1999 and 2004. Percentages
referred to: the urinary antigen, bacterial culture and
serology. |
Graf.
2 Incidence of legionellosis
in Europe in 2002 |
|
Incidence
of legionellosis in Europe in 2002. Data referred to 15
countries. |
Graf.
3 Number of cases
of legionellosis declared in Switzerland |
|
Number
of cases of legionellosis declared in Switzerland from 1995
to 2004. |
Graf.
4 Number of cases
of legionellosis declared in Switzerland |
|
Number
of cases of legionellosis in Switzerland (2002-2003). Percentages
based on sources of contamination. |
Risk and Reference Range
Depending on the use of water, the load of legionellae can be
either tolerated or incompatible. For example, as per new Swiss
directives, we can deduce that the risk is high in intensive care
units, neonatology and transplant wards, which require corrective
measures already with 100 UFC/l (units forming colony per litre).
Hence a correlation between the use of water, the tolerated presence
of legionellae, the environment and the host's sensitivity is
essential especially concerning immunocompetence. A simple model
created by placing the risk level and the concentration of Legionellae
in UFC in the Cartesian coordinate system enables to locate the
various infrastructures where water is used (Graph 5).
Graf.
5 Correlation between
the use of water and the tolerated presence of legionellae |
|
Correlation
between the use of water and the tolerated presence of Legionellae
(units forming colony (UFC) and risk level). |
It clearly appears at the two ends that with low legionellae concentrations
the risk is high for the mentioned hospital wards (intensive care,
transplants, etc.), while the risk is low for cooling towers and
administrative and commercial offices despite the recorded high
concentrations of legionellae. We can illustrate this problem
by resorting to a practical case recorded by us: a patient discharged
from the intensive care unit recorded a relapse of legionellosis
on returning home where the level of legionellae in water was
high (Figure 6).
He first blamed the hospital, but in this case it was the drinking
water, which represented a high risk due to the patient's compromised
immune condition. In fact by molecular typing of the
strains, we proved that the patient had been contaminated at home.
Fig.6
Legionellosis relapse
in patient with kidney transplant |
|
Legionellosis
relapse in a patient who has undergone kidney transplantation.
The source of contagion, which triggered the relapse, was
water supplied to the house. |
Conclusions
Concluding we can highlight the following points:
- Problems related to water management have never found decisive
solutions even in industrialised nations with advanced hygiene
measures. The contamination of water from germs, whose existence
was even ignored, as occurred with Legionellae (which were unknown
till 1976) and new devices (air conditioning) enhance water's
role in spreading emerging pathogens.
- The knowledge of Legionellae's ecological needs and requirements
enables us to better fight their proliferation. The diagnosis
of legionellosis must be combined with the search for Legionellae
in the water environment (main source of infections) to be able
to remove the disease's contaminating reservoir.
- New analytical approaches, which are especially based on molecular
biology methods, enable to effectively fight the breakout of an
epidemic of legionellosis.
- The consequence of the diagnostic superiority of the urinary
antigen is a reduction in culture-based analyses and hence in
the isolation of the responsible bacterial strains, thus preventing
the typing method from tracing the source of the infection. This
drawback can be bypassed by perfecting new molecular typing methods
designed to enable the characterisation of the strain directly
on the clinical sample and hence without requiring the culture
phase.
- However the polluting source must be sought, checked and decontaminated.
It is also possible to trigger the legionary's disease as a professional
disease in certain professions, which produce aerosol in the place
of work. In fact, after the diagnosis, the first question patients
are asked with a questionnaire concerns the work environment.
Raffaele Peduzzi
Professor of Microbiology,
University of Geneva Switzerland.
Director of the Cantonal Bacteriological Institute
Bellinzona - Canton Ticino - Switzerland.
Valeria Gaia
Centro Nazionale di Referenza per Legionella, Bellinzona
Loris Landi
Dirigente medico A.S.L. Napoli 5, Ercolano
