Visit to Public Health Laboratory
Service Headquarters, Colindale, 22 July 1997
1. Six members of the Sub-Committee,
with staff, met at the headquarters of the Public Health Laboratory
Service (PHLS) in Colindale, for informal briefing on PHLS's work
on infections resistant to antibiotics.
2. Professor Brian Duerden,
Deputy Director, introduced PHLS. PHLS is funded by the Department
of Health and the Welsh Office to provide microbiology and epidemiology
services in England and Wales. It has two central facilities in
Colindale, the Central Public Health Laboratory (CPHLmicrobiology)
and the Communicable Diseases Surveillance Centre (CDSCepidemiology),
and 48 Public Health Laboratories at NHS hospitals around the
country. There are Reference Laboratories at CPHL, and some of
the PHLs have Reference Units for particular infections. PHLS
has close contact with the many NHS clinical microbiology laboratories
which are not PHLs, and with the corresponding services in Scotland,
Northern Ireland and the Irish Republic.
3. PHLS can track the micro-organisms
causing disease and monitor their resistance to antimicrobial
agents in three ways:
(i) general surveillance: all
laboratories regularly report all cases of specified infections
to the centre;
(ii) Reference Laboratories receive
micro-organisms and data for more detailed investigations; they
may in turn rely on a network of "sentinel laboratories"
for specific surveillance;
(iii) special "snapshot"
surveys of limited duration, organised from the centre.
4. These processes are not without
problems. First, they depend on voluntary reporting; PHLS would
like to see legislation for mandatory reporting. Secondly, they
do not provide "denominator" data: i.e. how reported
cases compare with the situation in the population at large. Thirdly,
they are skewed by the tendency for most Reference Laboratories
(other than the Laboratory for Enteric Pathogens: see below) to
see only "interesting" cases, including cases exhibiting
resistance. Fourthly, they depend on different laboratories working
to standards and definitions which are both reliable and compatible.
All PHLS and NHS laboratories are regularly tested with blind
samples sent round by NEQAS; the British Society for Antimicrobial
Chemotherapy (BSAC) currently has a group working on standard
methods for the UK (with PHLS input); and the EU and the USA have
a task force on co-ordinated surveillance; but the situation remains
far from perfect.
5. Dr Barry Cookson, Director,
Laboratory for Hospital Infection, CPHL, spoke about methicillin-resistant
Staphylococcus aureus (MRSA). SA is present in the nose
of one person in three (persistent in one in ten, otherwise transient),
and is not usually resistant; but resistance develops rapidly,
usually in the context of a hospital. SA colonises skin but usually
cannot invade it unless there is a wound; if it does invade, and
sepsis arises, it becomes serious. SA is spread by direct contact.
6. "We are running out of antibiotics."
In the 1970s MRSA in the UK was checked, by a combination of gentamicin
and better infection control; but in the 1980s MRSA developed
gentamicin-resistance. In the UK MRSA can still be treated with
vancomycin and a number of other antibiotics. But in some parts
of the world vancomycin may be the only effective antibiotic,
and low-level vancomycin-resistant MRSA has emerged in Japan.
7. MRSA prevalence varies widely
between countries, regions, hospitals, and even individual wards.
However two recent strains, EMRSA-15 and -16, have spread across
the UK and crossed to Holland. EMRSA-16 can be virulent; it is
the first strain with two super-antigens, and can cause chest
infections in particular.
8. Ways to control MRSA include
infection control, policies to control use of antibiotics, and
other factors such as staff to patient ratios; but no system is
wholly reliable. Nurses are crucial, since they have most direct
contact with patients. Any survey of MRSA in a hospital must take
account of how many carriers brought it in with them. Many serious
cases arise from auto-infection, e.g. in a patient already carrying
MRSA who undergoes surgery and whose lesions are then invaded
as they heal.
9. Dr David Livermore, Head,
Antibiotic Reference Unit, CPHL (from 1st September 1997), described
the work of the ARU: it tests "difficult" organisms
for susceptibility or resistance to antibiotics, and checks surprising
results from other laboratories; it monitors and researches resistant
strains, and provides advice; and it supplies strains for use
10. ARU is trying to move from snapshot
surveys, which are resource-intensive, to continuous local monitoring.
As noted above, this depends on quality assurance (QA); susceptibility-testing
in the UK is much less uniform than in the USA. BSAC and PHLS
are currently testing new methods, and PHLS hope to have sentinel
laboratories in place by the end of 1998.
11. Dr Livermore mentioned penicillin-resistance
in Streptococcus pneumoniae (= pneumococcus). In the
UK resistance has risen to about 4 per cent. Iceland has
suffered a dramatic rise, from one per cent in 1980 to 30 per
cent in 1990. This is believed to have originated via travellers
returning from Spain.
12. Dr Livermore also mentioned
resistance to vancomycin. Because of how this antibiotic works
(it binds to a substrate rather than an enzyme), resistance was
thought to be impossible. However in about 1987 there emerged
the vancomycin-resistant enterococcus (VRE); how this came about
is a mystery. As to why, it happened while avoparcin, a related
antibiotic, was being used in Europe as an animal growth-promoter.
This "can't have helped"; but the human and animal strains
of enterococcus concerned are not the same; and the USA has more
VRE than Europe, although avoparcin is not used there. Plasmid
transfer of vancomycin-resistance to MRSA is now a "substantial
risk", and has been demonstrated in the laboratory on the
skin of a mouse; but the low-level VRSA found in Japan is not
due to gene transfer and probably arose by mutation.
13. Dr Bernard Rowe, Director,
Laboratory for Enteric Pathogens, CPHL, regards resistance to
antibiotics as "unwelcome but inevitable". He spoke
about salmonella, of which there are about 2000 serotypes, causing
typhoid, paratyphoid and enterocolitis, most commonly due to food
14. Since 1989, Salmonella typhi
has acquired resistance to chloramphenicol; the incidence is now
around 30 per cent. Resistance emerged in India and Pakistan,
and most cases found in the UK can be traced to infection in SE Asia.
15. There are about 30,000 cases
a year of human salmonella food-poisoning in England and Wales,
mostly acquired by eating UK food. As the table shows, multi-drug
resistance is now most common in S. typhimurium, the second
most prevalent serotype in humans in the UK. A particularly worrying
strain is S. typhimurium DT104, which is multi-resistant.
This is the major cause of salmonellosis in cattle, sheep, pigs
and poultry; it also infects humans, and its spectrum of resistance
is widening. Dr Rowe attributes this to use of antibiotics in
food animals. For instance, late in 1993 enrofloxacin was licensed
for veterinary use; DT104 promptly acquired resistance to the
very similar human antibiotic ciprofloxacin. It should be noted
that, though highly resistant, DT104 is no more virulent for humans
than other salmonellae, and in most cases antibiotics are not
16. Dr Rowe believes that farmers
are using antibiotics as a substitute for good hygiene and husbandry;
in his view, the ban on use of human antibiotics as animal growth
promoters, introduced after the Swann Report of 1969, should be
extended to cover prophylaxis. (There is a grey area around "metaphylaxis":
treatment of animals believed to be incubating disease.) He detects
a recent change of heart on this issue in agribusiness.
Multi-drug resistance in Salmonella
in England and Wales, 1990-96
|Per cent of salmonellosis in humans 1996
||MDR per cent 1990
||MDR per cent 1996
|| 63|| 0.8
|| 19|| 18
|| 4|| 11
|| 2|| 2
MDR = Multi-drug resistant,
i.e. to four or more antibiotics
17. Dr Patrick Wall,
Consultant Epidemiologist, CDSC, described ENTERNET, a project
for international surveillance of food poisoning. ENTERNET, funded
by the EU under BIOMED, involves EU member states and also Switzerland
and Norway. It began as Salm-Net to monitor salmonellae, and has
been extended to take in the antibiotic resistance of salmonellae
and E. coli 0157. Participating laboratories commit themselves
to acquire a minimum data set, and training in standard methods
is provided by PHLS if required. Dr Wall gave striking examples
of successful "detective work", identifying international
outbreaks of food poisoning and tracing them to their source.
Recently the food industry has become more involved with the PHLS
in initiatives to improve the safety of food, and PHLS are now
working with the supermarkets to develop codes of practice for
food production and handling.
18. Dr Francis Drobniewski,
Mycobacterium Reference Unit, Dulwich PHL, spoke about multi-drug
resistant tuberculosis (MDR-TBdefined as resistant to isoniazid
and rifampicin). TB is responsible for more deaths worldwide than
any other infectious disease (3m per year). The emergence of resistance
is encouraged by failure of patients to complete antibiotic courses
(typically 6 months or longer), by poor prescribing, and by the
physiology of certain groups of patients, particularly people
with AIDS, among whom MDR-TB causes 80-90 per cent mortality.
There have been two recent outbreaks of MDR-TB among people with
AIDS in London hospitals (Chelsea and Westminster 1995, St Thomas'
1996); these have given rise to expensive litigation, and a major
programme to build negative-pressure isolation rooms. The WHO
has set up a Global Surveillance System, involving 22 reference
laboratories of which the MRU is one. The MRU has been appointed
the WHO European Region Co-ordinating Laboratory in recognition
of its work in the field of drug resistance. There is a lot of
MDR-TB in Eastern Europe; the eastward expansion of the EU is
therefore a cause for concern.
19. Dr J Watson, Consultant
Epidemiologist, CDSC, described MYCOBNET, the UK surveillance
system for TB set up in 1995. Data for the whole UK is collected
at PHLS CDSC. In the UK as a whole MDR-TB is rare but increasing;
it is concentrated in London, among people with AIDS, among recent
immigrants (50 per cent of TB patients in the UK were born
overseas; screening on or after arrival is impractical), among
homeless people (who are particularly prone to non-compliance)
and among people who have had TB before. MDR-TB is a global problem;
but, as a recent WHO review has shown, data from around the world
are patchy and inconsistent.
20. Dr Ros Stanwell-Smith,
Consultant Epidemiologist, CDSC, stressed the need not merely
to "count corpses", but to improve the situation by
changing behaviour. In her view, PHLS leads the world at the former
but could do more of the latter. By way of example, she described
how she provides surgeons with prompt and graphic feedback concerning
resistant infection of surgical wounds. Hospital policies on use
of antibiotics are effective only if staff can be persuaded to
follow them; one good way to find out what is actually happening
is to ask patients. Continuing education of medical staff is a