Hospital-acquired infections at the 11th ECCMID in Istanbul: It was the tea flask that spread the germs

Hospital-acquired infections, or "nosocomial" infections in medical jargon, are a continual worry for specialists. They are becoming more and more common, and represent an ever-increasing economic burden, costing the UK alone over a billion pounds sterling every year. Not to mention threatening the health of more and more patients and causing thousands of deaths (more than 50000 in the US annually). For the germs that cause them have a very marked tendency to resist antibiotics and breed in places that are often difficult to detect. Many original communications on this subject are being presented at ECCMID, which is even organising a number of special symposia on nosocomial infections.

One of the startling communications was presented at Istanbul by Regina Watschinger, from Austria. She explained to ECCMID that her infection control team in Linz had isolated strains of Acinetobacter baumannii resistant to several antibiotics on the surface of gastric tubes. The bacterium, very common and usually harmless, becomes dangerous in certain conditions, especially in subjects whose resistance is compromised, such as patients in intensive care.

Having fruitlessly scrutinised all the possible infection sources, Watschinger decided to examine the tea flasks. For the tea, mass-produced like in many hospitals and stored in thermos flasks to keep it warm, was also used in Linz to rinse the gastric tubes.

"We therefore examined two thermos bottles in which the tea was kept warm before it was used to rinse the gastric tubes. In one of them, we successfully isolated a multiresistant Acinobacter baumannii strain", explained Watschinger. "Then we looked at 56 other tea flasks from nine departments in three different hospitals. Nearly half (26 to be precise) were contaminated in the same way."

Because of the porous rubber O-ring

Elaborate identification methods were used to show that the strains isolated on the thermos flasks were identical to those found on the patients� gastric tubes. The bacterial contamination was actually in liquid found between the inner and outer wall of the thermos flask. Repeated sterilisation can cause the O-ring to become porous and so contaminated liquid can leak through the O-ring. Another communication presented at Istanbul by the same team shows that traditional thermos flasks are real breeding-grounds for bacteria, and at least hospital high-risk zones such as intensive care, oncology and neonatal units should only use thermos flasks capable of being properly sterilized.

Unfortunately, the Linz case is not unique, as illustrated by other communications to the Congress, and many intensive care units have fallen prey to multiresistant Acinetobacter strains. When patients are infected, they can certainly be treated with carbapenems, highly powerful antibiotics capable of overpowering many normally resistant strains. But the situation is changing rapidly, and certain strains now resist even carbapenems.

That was what had happened at the intensive care unit in Izmir hospital, Turkey, explained Dr. Zeynep G�lay in his poster on Tuesday. The fact that the 11th ECCMID will be hearing nearly sixty communications on the different strains of Acinetobacter gives us some idea of the problem�s scale.

Water again

Among the other frequent culprits for hospital-acquired infections, Pseudomonas aeruginosa is well known and feared by doctors. For it also demonstrates remarkable resistance to antibiotics. So all possible precautions are taken to avoid outbreaks of infection. They are largely successful, indeed, since the number of Pseudomonas infections recorded in immunocompromised patients, who are highly vulnerable to all sorts of infection, has fallen markedly in the past decade.

Yet doctors were caught on the hop recently at the University Hospital of Basel (Switzerland) in the reverse isolation unit specially designed for bone-marrow transplant patients. That unit had undergone considerable expensive work, including the installation of a centralized water treatment system using best available technology, with nanofiltration and continuous water circulation in order to prevent stagnant areas, a classic breeding-ground for nosocomial germ colonies.

"Despite all that", bewails Andreas Widmer in his oral communication, to be presented on Wednesday morning, "we isolated Pseudomonas even before the patients had arrived in the unit!". Repeated thermic shocks with water heated to 80� over a ten- minute period did not help. From 5% of positive cultures, the level rose to 40%, and as far as 60% at which point there was an outbreak of infections.

After a detailed "fingerprint" was produced for some of the germs isolated on patients, they were found to be the same as germs found in the water supply. "Pseudomonas was found in the tap water and shower nozzles of every one of the 14 bedrooms in the special unit!" explains Widmer. The filters of the water intake system have been changed since, removing the source of the infection. About time too, given that mortality rates from Pseudomonas among such patients can border on 50%.

It will be clear by now that, when tracking the source of nosocomial infections, the water supply should be a major suspect. But it is not always easy to establish the link, as reported in Istanbul on Tuesday by Dominique Blanc, from the University Hospital of Lausanne (Switzerland), which suffered an outbreak of Burkholderia cepacia. This germ, a relative of Pseudomonas, was detected by his team in the respiratory tracts of 14 patients over a period of 17 days. The bronchoscopes and accessories were immediately checked, but found to be clear. In the end, the culprit proved to be the local anaesthetic solution manufactured by the hospital pharmacy!

"From comparison of the various germs isolated, it appears very probable that the contamination was caused by the use of ordinary distilled water. We found that a demijohn of ordinary distilled water in the pharmacy contained a strain with genetic fingerprints identical to those of the strain isolated in the solution and in the patients."

Legionellosis

The contaminated water connection is found in many other mini-epidemics that today affect most hospitals and are linked to Legionnaire�s disease, also called legionellosis, which can have a mortality rate of more than 40% if not treated. Hardly surprising, then, that the 11th ECCMID has taken a strong stance on this type of hospital-acquired infection, given the increasing number of potential contamination sources in the environment and the repeat outbreaks of infection. The "Recent Progress on Legionellosis" symposium at Istanbul reiterated that any type of water supply installation can encourage legionella to breed, be it drinking water pipes, taps, shower nozzles, humidifiers in the air-conditioning or other equipment such as nebulizers.

Repeat outbreaks are occurring in all European countries, with 2136 reported cases in 1999 against only 1360 in 1997. That works out at an estimated 5.4 cases per million inhabitants annually.

Victor Yu, from Pittsburgh, one of the speakers at the Legionellosis symposium, is horrified by this figure. "Specialized diagnostic laboratory tests are necessary and few hospitals in Europe use these tests routinely for patients with pulmonary infiltrates and fever. As a result, hospital-acquired Legionnaire�s disease is usually underdiagnosed." What is more, there are effective disinfection methods, but they must be combined with continuous monitoring of potential legionella breeding sites; only in this way can patients be fully protected.

Acinetobacter, Pseudomonas, Legionella, the same struggle: early on Wednesday morning, Victor Yu and Andreas Widmer led one of the 11th ECCMID�s "Meet the Experts" sessions, on the general subject of monitoring water in the hospital environment. There were five basic questions on the agenda: Why? Where? What? When? and How? Are the specialists sure that they know all the answers?

Contacts :

Regina Watschinger
Department of Medical Microbiology and Hygiene
Elisabethinen Hospital
Linz, Austria
Tel. (+43) 73276763680
e-mail: [email protected]

Zeynep G�lay
Department of Microbiology and Clinical Microbiology
Dokuz Eylul University School of Medicine
Izmir, Turkey
Tel. (+90) 23202595959
e-mail: [email protected]

Andreas Widmer
Division of Hospital Epidemiology, University Hospitals
Basel, Switzerland
Tel. (+41) 612653851
e-mail: [email protected]

Dominique Blanc
Centre Universitaire Hospitalier Vaudois (CHUV)
Lausanne, Switzerland
Tel. (+41) 21 314 02 60
e-mail: [email protected]

Victor L Yu
Professor of Medicine, University of Pittsburgh
Pittsburgh, Pennsylvania, USA
Tel. (+1) 412 688 61 79
e-mail: [email protected]