Mortality after Staphylococcus aureus bacteraemia in two hospitals in Oxfordshire, 1997-2003: cohort studyBMJ 2006; 333 doi: https://doi.org/10.1136/bmj.38834.421713.2F (Published 03 August 2006) Cite this as: BMJ 2006;333:281
All rapid responses
As family physicians in an academic setting in a 'Third World'
country (South Africa), we find the article 'Mortality after
Staphylococcus aureus bacteraemia in two acute hospitals in Oxfordshire'
by David H Wyllie and colleagues very interesting. We need to point out
from the very outset that none of us are microbiologists but that we more
than frequently care for patients with intractable bacterial infections at
one tertiary hospital and one district hospital.
We are aware that in Europe, The Netherlands appears to have found an
answer to hospital acquired infections - Methicillin Resistant
Staphylococcus Aureus, as indeed published literature shows.(1) We note
with interest that in the same issue of The BMJ, consultant microbiologist
John Paul while paying tribute to the Dutch approach, he cautions 'To do
this properly would require a huge investment in facilities, however, and
might take a decade or so to bear fruit.'(2) We agree with this assertion
and wish to point out that in impoverished Third World countries, such an
approach would for ever remain a pipe dream.
Countries of the Third World face numerous acute and chronic problems
with regard to caring for the sick. One of the major obstacles is that
these countries inherited a model of health care which is largely
unsuitable for the care of their populations. That model of health care is
what medical social scientists frequently refer to as biomedicine or the
engineering model of health care. This model, which continues to be
sustained by most health ministries cannot be an answer to poverty,
malnutrion, destitution, squalor and extremely poor or absent sanitation.
Wyllie and colleagues state 'A key issue is whether bacteraemia was the
underlying cause of death in these patients. It is difficult to attribute
the cause of death in cases of S aureus bacteraemia because of the many
coexisting risk factors for death, such as acute and chronic illness, and
a high degree of medical inervention.'(3). In our situation, a large
number of patients with bacteraemia are immunocompromised and may have TB
as well as numerous opportunistic infections.
As family physicians in a Third World country, we strongly advocate
the management of patients in their own homes. This, without any doubt
would reduce the 'pandemic' of hospital acquired infections. We suggest in
the spirit of Alma Ata (1978), that governments divert the enormous
resources currently allocated to hospitals to family medicine and primary
health care in the communities.
1. Bootsma MC, et al, Controlling methicillin-resistant Staphylococcus
aureus: quantifying the effects of interventions and rapid diagnostic
testing. Proc Natl Acad Sci USA 2006; 103:5620-5
2.John Paul: Surveillance and management of all types of Staphylococcus
aureus bacteraemia: BMJ Vol 333, 5 August 2006, 269-270
3.Wylie, David H et al: Mortality after Staphylococcus aureus bacteraemia
in two acute hospitals in Oxfordshire, 1997-2003: cohort study: BMJ Vol
333 5 August 2006
Competing interests: No competing interests
We read with interest the study by Wyllie et al analysing the
possible impact of MRSA bacteraemia on short-term mortality . Although
they have alluded to the fact that their study is limited by the lack of
data on coexisting risk factors for death, this is pertinent in the
context of MRSA bacteraemia and indeed very crucial when interpreting the
results of their evaluation.
As an illustrative example, we had assessed the relationship of
mortality in hospital patients with MRSA bacteraemia to the presence of
hyperglycaemia on admission during a similar time period (2000-2002)
amongst the 309 patients aged 16 years or above with MRSA bacteraemia in
our centre and found that mortality was significantly more in the highest
tertile of blood glucose distributions [2,3]. This has particular
inference, if we take into consideration other commonly associated risk
factors for vascular events such as hypertension .
Of note, increased screening and surveillance for MRSA was gaining
momentum during this period. Wyllie et al have demonstrated that a higher
proportion (n=97) of the patients at risk of death within 30 days of
diagnosing MRSA bacteraemia were amongst those discharged from Medicine
. In view of the global prevalence of chronic illnesses such as
diabetes mellitus and the projections for 2030 , the results of such
retrospective analyses should be interpreted cautiously.
1. Wyllie DH, Crook DW, Peto TEA. Mortality after Staphylococcus
aureus bacteraemia in two hospitals in Oxfordshire, 1997-2003: cohort
BMJ 2006; 333: 281-284.
2. Varughese GI, Miltsios K, Buch HN, Orendi JM, Scarpello JH.
Mortality rates in hospital patients with hyperglycaemia and MRSA
bacteraemia. Br J Diabetes Vasc Dis 2006; 6: 42-44.
3. Varughese GI, Miltsios K, Barton DM, Buch HN, Orendi JM, Scarpello
JH. An audit of laboratory blood glucose determinations and their relation
to mortality in hospital in-patients with MRSA bacteraemia. Diabetic
Medicine 2004; 21(Supp. 2): P192.
4. Varughese GI, Patel JV, Lip GY. Blood pressure control in the
setting of diabetes mellitus: new targets, new hope for improvement? J Hum
Hypertens 2006 Apr 13; [Epub ahead of print].
5. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of
diabetes: estimates for the year 2000 and projections for 2030. Diab Care
2004; 27(5): 1047–1053.
George I. Varughese (Specialist Registrar)
Abd A. Tahrani (Specialist Registrar)
John H.B. Scarpello (Consultant Physician)
DEPARTMENT OF DIABETES, ENDOCRINOLOGY & GENERAL (INTERNAL)
UNIVERSITY HOSPITAL OF NORTH STAFFORDSHIRE,
STOKE-ON-TRENT ST4 6QG (UK)
Competing interests: No competing interests
1. I believe that this study was not a cohort study as such, since it
was not a prospective observational study but rather a retrospective
2. The study analyzed the episodes of staphylococcus bacteraemia rather
than the first episode of staphylococcus bacteraemia in each patient only:
without a doubt, a few patients must have had repeatedly positive blood
cultures with the same pathogen and probably more so in the MRSA group. It
is not clear from the study what proportion of the 441 bacteraemia
episodes included in the study might have been from the same patients,
something that might have biased the outcomes in the MRSA or the MSSA
group or both.
3. The study covered a period of seven years: within these seven years
medical advances have certainly occurred, that might have affected both
the occurrence of bacteraemia (cohorting, intravenous line care etc) and
also the outcome of it (more antibiotics available, better culture systems
etc). Stratifying the data to six-month blocks eliminates some but not all
of this bias.
4. The importance of comorbidities actually being responsible for this
effect can not be underestimated: the studies quoted have not adequately
excluded comorbid illness as a possible associated factor.
• The study by McClelland et al . did not use a comorbidity index but
artificially separated patients to over 65 and under 60. Different models
that are not presented in the study were constructed, some of them using
the number of comorbid conditions instead of assigning a weight to each
one of them.
• The study by Blot et al. analysed the MRSA bacteraemias separately from
the MSSA bacteraemias and compared each group to their respective
controls, but not to each other. Moreover, the patients belonging to the
MRSA group had a mean of 29 days of hospitalization prior to the
bacteraemia (median 24) versus 10 days of hospitalization (median 6) for
the MSSA group (p <_0.001. similarly="similarly" the="the" mean="mean" apache="apache" ii="ii" scoring="scoring" for="for" two="two" groups="groups" was="was" statistically="statistically" different.="different." it="it" appears="appears" that="that" even="even" a="a" system="system" like="like" might="might" have="have" not="not" captured="captured" true="true" previous="previous" comorbidity="comorbidity" impact="impact" in="in" each="each" group.="group." p="p"/>5. Even if the difference in mortality is true, we have to be reminded
that it does not prove causality: once colonization of the anterior nose
with Staphylococcus aureus has happened, autoinfection happens more
commonly in MRSA-colonized patients than in MSSA colonized patients
because intercurrent antibiotic treatment eliminates one but not the other
(Archer G) . Of note that this notion has been recently challenged but the
demonstration of intracellular reservoirs Staphylococcus aureus in the
nasal mucosa (Clement S).
McClelland RS et al. Staphylococcus aureus bacteremia among elderly
vs younger adult patients:comparison of clinical features and mortality.
Arch Intern Med. 1999 Jun 14;159(11):1244-7.
Blot SI, Vandewoude KH, Hoste EA, Colardyn FA.
Outcome and attributable mortality in critically Ill patients with
bacteremia involving methicillin-susceptible and methicillin-resistant
Staphylococcus aureus. Arch Intern Med. 2002 Oct 28;162(19):2229-35.
Archer G, Climo M. Staphylococcus aureus bacteremia—consider the
source. N Engl J Med. 2001;344:55–56.
Clement S et al. Evidence of an intracellular reservoir in the nasal
mucosa of patients with
recurrent Staphylococcus aureus rhinosinusitis.
J Infect Dis. 2005 Sep 15;192(6):1023-8. Epub 2005 Aug 9.
Competing interests: No competing interests