Outcome FROM PROPERTY

Reection and Reaction
academia. Such discussion should lead to creation

of mechanisms that allows a more streamlined and
uniform global approach, and to agreement on
protocols, analysis plans, and criteria for approval that
provide the pharmaceutical industry with sucient
basis to continue the development of agents for
conditions such as sepsis.
GST is an employee of Viropharma Inc, which has no products licensed or in

development in the area of sepsis.
1
Glenn S Tillotson
Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis

in the United States: analysis of incidence, outcome, and associated costs
of care. Crit Care Med 2001; 29: 130310.
Dellinger RP, Levy MM, Carlet JM, et al. Surviving sepsis campaign:
international guidelines for management of severe sepsis and septic shock:
2008. Crit Care Med 2008; 36: 296327.
Poole D, Bertolini G, Garattini S. Errors in the approval process and postmarketing evaluation of drotrecogin alfa (activated) for the treatment of
severe sepsis. Lancet Infect Dis 2009; 9: 6772.
http://www.fda.gov/OHRMS/DOCKETS/98fr/E8-24354.htm (accessed
Dec 2, 2008).
Viropharma Inc, Exton, PA, USA

Glenn.tillotson@viropharma.com
Household contact investigation of tuberculosis in lowincome and middle-income countries: public-health impact
Janina Morrison and co-workers1 showed in a systematic
review that household contact investigation merited
serious consideration as a means to improve tuberculosis
control in low-income and middle-income countries
with a high incidence of the disease. The pooled yield
among all household contacts was 45% (4348%,
I2=955%) for all active tuberculosis, and 23% (2125%,
I2=966%) for conrmed active tuberculosis.1 Substantial
heterogeneity exists across studies, possibly related
to variations in source, environmental, or contact
characteristics, screening methods, and diagnostic
thresholds. These factors might reduce the ability of
the pooled yield to predict the exact outcome in any
particular country.
Although the pooled yield of a case-nding
method might give an indication of the potential
cost-eectiveness, the proportion of tuberculosis
cases detected by the particular method might be
of greater relevance in the assessment of its publichealth impact. Since active case nding is often
limited to household contact investigations in lowincome and middle-income countries, it might be
reasonable to assume that all or practically all index
cases that prompted household contact investigation
were identied by passive case nding. Using this
assumption, the pooled estimates of the proportion
of active tuberculosis (with or without bacteriological
conrmation) based on the same studies included in
the Review,1 after correcting for over-dispersion,2 would
be 015 (011019, I2=956%) for household contact
www.thelancet.com/infection Vol 9 January 2009
Proportion of all active

tuberculosis detected
by household contact
investigation
Afonja et al
Alouch et al
Alouch et al
Andrews et al
Aziz et al
Bayona et al
Becerra et al
Claessens et al
Devadetta et al
Egsmose et al
Espinal et al
Gilpin et al
Guwatudde et al
Klausner et al
Kritski et al
Kumar et al
Lemos et al
Narain et al
Nsanzumuhire et al
Nunn et al
Saunders et al
Suggaravetsiri et al
Teixeira et al
Wares et al
WHO
Zachariah et al
054 (041065)
002 (0007)
008 (003016)
020 (015026)
034 (025043)
027 (022033)
005 (002009)
007 (005009)
020 (016025)
042 (035049)
021 (016027)
006 (002014)
020 (016025)
024 (019030)
021 (013031)
025 (016037)
013 (006022)
009 (006012)
008 (003017)
020 (013029)
017 (015020)
010 (008013)
018 (011027)
002 (001003)
035 (026045)
005 (002008)
Pooled proportion
015 (011019)
2=56510; df=25 (p=00001)
I2=956%
0
02
04
06
08
10
Figure: Forest plot of the proportion of all active tuberculosis (with or without bacteriological conrmation)
detected by household contact investigation based on studies included in the systematic review by Morrison
and co-workers1
Reection and Reaction
investigation (gure 1) and 085 (081089, I2=956%)

for passive case nding. Restricting pooled analysis to
conrmed active tuberculosis, the pooled estimates
would be 008 (005012, I2=93.5%) for household
contact investigation and 092 (088095, I2=935%)
for passive case nding. Such ndings may highlight
the key role of passive case nding in the control of
tuberculosis.3
The public-health impact of household contact
investigation is expected to be sustantially lower than
that of passive case nding. The incubation period of
tuberculosis varies from a few weeks to a few decades
and household contact investigation focuses on
examination at only one point of time. Additionally,
most infected hosts do not develop disease. Thus,
it may be more cost eective for low-income and
middle-income countries to spend limited public-health
resources on improving accessibility of a patient-friendly
health-care infrastructure4 and on increasing public
awareness of tuberculosis, upon which passive case
nding heavily relies. The feasibility of achieving the case
detection target of 70% by passive case nding has been
substantiated by early studies in India, which showed
that 70% of people with smear-positive tuberculosis had
symptoms and sought health care.5
In conclusion, although household contact
investigation may be considered in low tuberculosis
incidence, high-income countries,6 the available

evidence from meta-analysis does not favour household contact investigation in low-income and middleincome countries. The importance of improving case
detection among symptomatic patients self-reporting
to health services cannot be over-emphasised.3
*Kwok Chiu Chang, Chi Chiu Leung, Cheuk Ming Tam
Tuberculosis and Chest Service, Centre for Health Protection,
Department of Health, Hong Kong Special Administrative Region,
China
kc_chang@dh.gov.hk
We declare that we have no conicts of interest.
1
2
3
Morrison J, Pai M, Hopewell PC. Tuberculosis and latent tuberculosis

infection in close contacts of people with pulmonary tuberculosis in lowincome and middle-income countries: a systematic review and
meta-analysis. Lancet Infect Dis 2008; 8: 35968.
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials
1986; 7: 17788.
World Health Organization, International Union Against Tuberculosis and
Lung Disease, Royal Netherlands Tuberculosis Association. Revised
international denitions in tuberculosis control. Int J Tuberc Lung Dis 2001;
5: 21315.
Raviglione M, Frieden T. What are examples of eective tuberculosis
control programmes? In: Frieden T, ed. Tomans tuberculosis case
detection, treatment and monitoring: questions and answers.
WHO/HTM/TB/2004.334. Geneva: World Health Organization,
2004: 31821.
Banerji D, Andersen S. A sociological study of awareness of symptoms
among persons with pulmonary tuberculosis. Bull World Health Organ 1963;
29: 66583.
WHO. What is DOTS? A guide to understanding the WHO-recommended
TB control strategy known as DOTS. WHO/CDS/CPC/ TB/99.270. Geneva:
World Health Organization, 1999.
Ecacy and safety of cefepime

A Review by Dafna Yahav and colleagues1 found that
cefepime compared with other beta-lactam antibiotics
was associated with increased all-cause mortality, a
dierence driven by the febrile neutropenia subset of
patients (risk ratio [RR] 126, 95% CI 108149).
To better understand these dierences and to
determine if infectious or non-infectious causes
impacted the mortality results, we reviewed the
19 studies comprising the neutropenic fever subset
of these data. Whenever possible, the actual articles
were obtained from the FDA website or through
medical library holdings. Where abstracts were the
only information available or data within the published
literature were not adequate to answer all questions,
every attempt was made to contact the original
authors. Studies were specically reviewed for data
4
including number of deaths in each arm and causes of

death.
For these 19 studies, complete cause of death
information was obtained for 11 and partial cause of
death information for two. These 13 studies included
64% of the all-cause neutropenic deaths in Yahav and
colleagues paper. Review of causes of death among
these patients found no marked dierences between
cefepime and beta-lactam comparator for any infectious
cause (table 1). A higher proportion of patients died
secondary to progression of their underlying disease in
the cefepime arm compared with the other beta-lactam
arm. Furthermore, no patients were determined to have
died directly as a result of receiving therapy with any
agent, including cefepime (references 214, and personal
communication with the lead author of reference 4).
www.thelancet.com/infection Vol 9 January 2009
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Outcome FROM PROPERTY

Uploaded by

Copyright:

Available Formats

Outcome FROM PROPERTY

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Outcome FROM PROPERTY

Uploaded by

Copyright:

Available Formats

Reection and Reaction

academia. Such discussion should lead to creation

GST is an employee of Viropharma Inc, which has no products licensed or in

Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis

Viropharma Inc, Exton, PA, USA

Proportion of all active

Reection and Reaction

investigation (gure 1) and 085 (081089, I2=956%)

incidence, high-income countries,6 the available

Morrison J, Pai M, Hopewell PC. Tuberculosis and latent tuberculosis

Ecacy and safety of cefepime

including number of deaths in each arm and causes of

You might also like