Escolar Documentos
Profissional Documentos
Cultura Documentos
Corbis
New at WFP The World Food Programme (WFP), the UNs rst response
organisation for natural disaster relief
and the ght against world hunger,
has appointed Ertharin Cousin as
its new Executive Director. Cousin,
currently US representative to the UN
agencies for food and agriculture, will
be taking over from current Executive
Director Josette Sheeran in April.
Paul Grundy
Editorial
287
Editorial
FPG/Getty Images
Corbis
288
Comment
for who should be treated and for how long, are based
on the consensus opinion of experts in helminth control5
and do not take account of analytical methods based on
current understanding of the transmission dynamics of
helminths. The key questions for control by chemotherapy
alone focus on how to achieve the greatest impact with
the minimum use of anthelmintic drugs (panel). Disease
transmission during drug treatment is a dynamic process,
with loss and acquisition of worms determined by many
factors, including coverage and ecacy. Models of these
processes provide predictive information about the eect
of intervention at the community level, whether by mass
or targeted chemotherapy.7 This knowledge, however, is
rarely applied in practice.
Helminths have much more predictable dynamics than
most viruses, bacteria, and protozoa. After treatment,
worm populations return to precontrol levels in a
monotonic way, because of density-dependent processes
that inuence parasite reproduction, infection, and mortality and are partly related to the build-up of a small
degree of acquired immunity; the long life expectancy of
established worms in the human host (years rather than
days) is also a factor.5 This understanding should be used
more widely to rene community-based guidelines and
to inform policy makers of what to expect from a given
intervention programme.
289
Comment
290
2
3
4
5
6
7
8
9
Comment
Published Online
January 7, 2012
DOI:10.1016/S01406736(11)61928-4
See Articles page 315
291
Comment
Toshirou Nishida
Department of Surgery, Osaka Police Hospital, Osaka 543-0035,
Japan, and Department of Surgery, Osaka University Graduate
School of Medicine, Osaka, Japan
toshin@mvb.biglobe.ne.jp
I declare that I have no conicts of interest.
1
292
Comment
7
8
Singer DE, Chang Y, Fang MC, et al. Should patient characteristics inuence
target anticoagulation intensity for stroke prevention in nonvalvular atrial
brillation? Circ Cardiovasc Qual Outcomes 2009; 2: 297304.
Connolly SJ, Ezekowitz MD, Yusuf S, et al, for the RE-LY Steering Committee
and Investigators. Dabigatran versus warfarin in patients with atrial
brillation. N Engl J Med 2009; 361: 113951.
Patel MR, Mahaey KW, Garg J, et al, for the ROCKET AF Steering
Committee and Investigators. Rivaroxaban versus warfarin in nonvalvular
atrial brillation. N Engl J Med 2011; 365: 88391.
Heneghan C, Ward, A, Perera R, and The Self-Monitoring Trialist
Collaboration. Self-monitoring of oral anticoagulation: systematic review
and meta-analysis of individual patient data. Lancet 2011; published online
Dec 1. DOI:10.1016/S0140-6736(11)61294-4.
Bloomeld HE, Krause A, Greer N, et al. Meta-analysis: eect of patient
self-testing and self-management of long-term anticoagulation on major
clinical outcomes. Ann Intern Med 2011; 154: 47282.
Schulman S, Kearon C, Kakkar AK, et al, for the RE-COVER Study Group.
Dabigatran versus warfarin in the treatment of acute venous
thromboembolism. N Engl J Med 2009; 361: 234252.
The EINSTEIN Investigators. Oral rivaroxaban for symptomatic venous
thromboembolism. N Engl J Med 2010; 363: 2499510.
Granger CB, Alexander JH, McMurray JJ, et al, for the ARISTOTLE
Committees and Investigators. Apixaban versus warfarin in patients with
atrial brillation. N Engl J Med 2011; 365: 98192.
Wallentin L, Yusuf S, Ezekowitz MD, et al, for the RE-LY investigators.
Efficacy and safety of dabigatran compared with warfarin at different
levels of international normalised ratio control for stroke prevention in
atrial fibrillation: an analysis of the RE-LY trial. Lancet 2010;
376: 97583.
293
Comment
294
Comment
6
7
8
9
10
11
12
13
Published Online
January 11, 2012
DOI:10.1016/S01406736(11)61842-4
See Articles page 342
295
Comment
3
4
5
Comment
10
11
12
13
14
To submit a paper go to
http://ees.elsevier.com/thelancet
Sabine Kleinert
The Lancet, London NW1 7BY, UK
Sabine.kleinert@lancet.com
297
Comment
Katherine Rolfe
World Report
John Donnelly
Dominic Chavez
Many of the children in Witto Payam village, South Sudan, have nodding syndrome
299
World Report
The long-running dispute over patents for the BRCA1 and BRCA2 genes granted to Myriad
Genetics may nally be laid to rest by the US Supreme Court. Sharmila Devi reports.
300
Sharmila Devi
www.thelancet.com Vol 379 January 28, 2012
World Report
Safety ags for Boehringer Ingelheims antiblood clotting agent dabigatran serve as a reminder
about the risks of rapidly adopting newly approved drugs. Asher Mullard reports.
Asher Mullard
301
World Report
AP
Greek patients queue outside a pharmacy during the pharmacists strike in January
302
Eva Karamanoli
www.thelancet.com Vol 379 January 28, 2012
Perspectives
Film
Beats and the brain
a lm about Parkinsons
disease seen through the prism
of music
Ray Chaudhuri at Londons Kings
College Hospital and Oxford neurosurgeon Tipu Aziz. Chaudhuri proposes
an apomorphine pump, Aziz discusses
deep brain stimulation and the future
potential of stem cells. Barbara opts for
the pump, but it takes a year of ghting
through the bureaucracy and hassles
over who will pay before she can return
to Kings to have the pump tted in
2010. As the apomorphine diuses
through her body, her body relaxes and,
for a time, she is able to go on tour with
Jon and his band, Colosseum.
But this isnt a lm about the clinical
course of Parkinsons disease, or the
303
Perspectives
Book
Joseph Rotblats conscientious life in science and politics
304
Andrew Robinson
andrew.robinson33@virgin.net
Perspectives
Prole
Ndola Prata: ghting for womens reproductive health
Ever since I was a little girl, I always wanted to x things,
says Ndola Prata, Scientic Director of the Bixby Center for
Population, Health and Sustainability at the University of
California, Berkeley. Prata battles daily for more pragmatic
approaches to reproductive and sexual health care. I nd
myself questioning why we dont focus on interventions
that can be scaled up, thus reaching most of the women
in need, she says. Her conviction that its worth ghting
for what you believe in comes from her parents, she says.
Growing up in conict-ridden Angola in the 1970s wasnt
easy. Once Portugal acceded independence in 1975, some
Angolan citizens left but Pratas parents resolutely stayed
put, believing that Angola would become a great country.
Prata was already well versed in the challenges of
reproductive health in her country by the time she went
to medical school. In my extended family you couldnt go
5 minutes without someone talking about a problem related
to reproductive health, she says. Although she initially
wanted to be a neurosurgeon, her trademark practicality
kicked in, and she realised that Angola would be better
served if she tried to address its high fertility rates and poor
reproductive care. By the late 1990s, Prata had practised
medicine in Angola for a decade and had also completed
an MSc in medical demography from the London School of
Hygiene and Tropical Medicine in the UK. In 1998, she moved
to the School of Public Health at the University of California,
Berkeley, where she has stayed ever since, combining her
work for the Bixby Center with her role as Associate Professor
of Maternal and Child Health. Prata was fairly certain her
move to the USA would be long termshe fell in love with
an American scientist, who she would later marry.
Much of Pratas time, she says, is spent ghting for
realistically aordable methods of improving reproductive
health to be used more widely. Her battle lines are often
etched on that tricky demarcation between care that is ideal
and that which is good enough. In this, she says, I often
clash with proponents of gold standards of care. Some
interventions have such high standards, that they are just not
achievable on a large scale in resource-poor settings within
an acceptable timeframe. This practical streak is partly a
legacy, Prata says, of coming from a country like Angola
you do what you can with what you have. Community
health workers, such as traditional birth attendants, for
instance, can be useful in improving access to maternal care,
she says, but their use has stirred controversy. They play
an important role in family planning and maternal health
because they are there with the women during pregnancy
and birth. Like it or not, scientists need to engage with
these workers to ensure their advice is medically sound. Not
acknowledging their role is more damaging, she argues.
www.thelancet.com Vol 379 January 28, 2012
Priya Shetty
305
Comment
Published Online
December 16, 2011
DOI:10.1016/S01406736(11)61884-9
e20
Comment
4
5
10
AG is Chair and JWW Vice Chair of the Presidential Commission for the Study of
Bioethical Issues.
11
1
2
e21
Comment
e22
Comment
2
3
e23
8
9
10
11
12
Perspectives
306
Private Collection/Photo Peter Nahum at The Leicester Galleries, London/The Bridgeman Art Library
Perspectives
Niall Boyce
The Lancet, London NW1 7BY, UK
307
Obituary
Robert Ader
Psychologist and co-founder of the eld of
psychoneuroimmunology. He was born on
Feb 20, 1932, in the Bronx, NY, USA, and
died on Dec 20, 2011, in Pittsford, NY, USA,
aged 79 years.
The change that Robert Ader helped initiate in medical science
began with a serendipitous discovery. It happened in the early
1970s, when he and Nicholas Cohen were studying taste
aversion. The researchers had been giving rats a saccharin
solution accompanied by an injection of cyclophosphamide,
an immunosuppressive drug that induces gastrointestinal
upset. When the injections stopped, they found as expected
that the rats had become conditioned to avoid consuming
the sweet solution. To complete the experimental protocol,
they forced the rats to take the saccharin solution using eye
droppers. This is where the surprise arrived.
Ader and Cohen found that some of the animals they had
force-fed with the saccharin later died. The magnitude of the
avoidance response and the mortality rate of the rats was
directly related to the volume of solution consumed. What
could be going on? A hypothesis that seemed reasonable
to me was that, in addition to conditioning the avoidance
response, we were conditioning the immunosuppressive
eects (of cyclophosphamide), Ader said in a 2010
interview. In other words, the taste of saccharin alone was
enough to stimulate neural signals that suppressed the
rats immune systems, just as if they had been overdosed
with the immunosuppressant. Ader and Cohen went on
to conrm this hypothesis in a controlled experiment,
308
Stephen Pincock
stephen.pincock@journalist.co.uk
Correspondence
Mohga M Kamal-Yanni
mkamalyanni@oxfam.org.uk
Published Online
January 13, 2012
DOI:10.1016/S01406736(12)60040-3
Action to preserve
WHOs core functions
cannot wait for
organisational reform
Submissions should be
made via our electronic
submission system at
http://ees.elsevier.com/
thelancet/
309
Correspondence
*Ingegerd stman-Smith,
Anne de-Wahl Granelli
ingegerd.ostman-smith@pediat.gu.se
Division of Paediatrics, Department of Clinical
Sciences, The Sahlgren Academy, Gothenburg
University, Section of Paediatric Cardiology, Queen
Silvia Childrens Hospital, Smrslottsgatan, stra
Sjukhuset, 416 85 Gothenburg, Sweden
1
310
Correspondence
*Ingegerd stman-Smith,
Anne de-Wahl Granelli
ingegerd.ostman-smith@pediat.gu.se
Division of Paediatrics, Department of Clinical
Sciences, The Sahlgren Academy, Gothenburg
University, Section of Paediatric Cardiology, Queen
Silvia Childrens Hospital, Smrslottsgatan, stra
Sjukhuset, 416 85 Gothenburg, Sweden
1
310
Correspondence
Krishnarajah Nirantharakumar
k.nirantharan@bham.ac.uk
University of Birmingham, Birmingham B15 2TT, UK
1
Authors reply
Dierences in results between
de-Wahl Granelli and colleagues
study1 and ours are likely to have arisen
because of antenatal screening and
timing of pulse oximetry. Detection
of critical congenital heart disease
by antenatal ultrasound is very
variable: we detected 50% of cases
of critical congenital heart disease
antenatally compared with only 3%
in de-Wahl Granelli and colleagues
study.1 Detection rates of between
15% and 50% have been reported
among UK health regions.2 We noted
a lower incremental value of pulse
oximetry than did de-Wahl Granelli and
colleagues, almost certainly because
many more cases were identied
by antenatal screening in our study
hospitals. Our detection rate for critical
congenital heart disease of 75% with
pulse oximetry for the full cohort
is higher than theirs. We agree that
routine visualisation of the outow
tracts should further improve antenatal detection in isolated cardiac
lesions; however, consistency of
detection has yet to be shown.
de-Wahl Granelli and colleagues
screened at a median of 38 h after
birth, compared with 12 h in our
study, which might explain our higher
false-positive rate. Discharge from
UK hospitals is frequently within
24 h, so delayed pulse oximetry is not
feasible and also increases risks of late
diagnosis. Importantly, half (28/57) of
eligible babies with critical congenital
heart disease in de-Wahl Granelli and
colleagues study required admission
www.thelancet.com Vol 379 January 28, 2012
*A K Ewer, L J Middleton,
S Thangaratinam, K S Khan, J J Deeks
a.k.ewer@bham.ac.uk
School of Clinical and Experimental Medicine,
University of Birmingham, Birmingham, UK (AKE);
*Birmingham Womens Healthcare NHS Foundation
Trust, Birmingham B152TG, UK (AKE); Birmingham
Clinical Trials Unit, University of Birmingham,
Birmingham, UK (LJM); Queen Mary University of
London, Barts and the London School of Medicine,
London, UK (ST, KSK); and Public Health,
311
Correspondence
Paul M Ridker
pridker@partners.org
Center for Cardiovascular Disease Prevention, Brigham
and Womens Hospital, Boston, MA 02215, USA
1
Authors reply
Paul Ridker questions the ability of
coronary artery calcium to identify
appropriate patients for statin
therapy. However, his implication that
increased coronary artery calcium is a
marker for poor response to statins is
not supported by the available data.
Statins clearly reduce recurrent
coronary events in secondary prevention
trials in which patients have established
coronary disease and presumably
elevated coronary artery calcium. The
St Francis Heart Study1 tested this
hypothesis more directly, randomising
1005 patients with elevated coronary
Correspondence
ADDITION-Europe and
the case for diabetes
screening
quite small. Were the benet signicant, it would have required screening
of 1829 people,2 and treatment of
78 screen-positive ones, for 53 years
to prevent one cardiovascular event.
Without comparison with an unscreened group, the study could
not explore the specic benets of
diabetic screening. Glycated haemoglobin (HbA1c) in screen-positive
people decreased by 0304%
(1920 mmol/mol) over 53 years,
whereas it would be expected to
increase. Meta-analyses have shown
that intensive glucose lowering,
although reducing rates of non-fatal
myocardial infarction and surrogate
measures of microangiopathy, does
not aect cardiovascular mortality,
stroke, or patient-important microvascular disease.3
But the downsides of screening are
evident. The diagnosis of diabetes is a
surrogate for challenges with employment and insurability; a need for clinic
visits, tests, self-monitoring, and
hypoglycaemic drug use (some 60%
in ADDITION-Europe); and inconvenience and cost. Moreover, although
ADDITION-Europe showed no psychological eect of screening itself, those
who screened positive had signicantly
worse general health, greater anxiety,
and were more depressed.4
Current evidence does not support
population screening for diabetes. This
chorus, coming from public health
advocates, professional societies, and
industry, is not without consequence:
Heath5 has eloquently described
how screening policies for noncommunicable diseases could result in
catastrophic spending by individuals
and institutions in low-income and
middle-income countries.
We declare that we have no conicts of interest.
Authors reply
ADDITION-Europe is not a trial of
screening for diabetes. We reviewed
the evidence for screening and identied the key uncertainties: the eect
of early treatment in the lead time
after detection by screening and the
harms associated with screening.1
ADDITION-Europe addressed the
eect of early treatment. Inclusion
of a no-screening control group in
one centre enabled us to add to the
growing evidence of minimal harms
associated with screening.2 A diagnosis
of diabetes certainly has adverse consequences, which are brought forward
by screening. However, our study
suggests that stepwise screening
might attenuate these eects by
facilitating psychological adjustment.3
We share John Yudkin and colleagues
healthy scepticism about the eects
of rapid, intensive glucose lowering
in patients with longstanding poor
glycaemic control, but we question
its relevance in this context. Patients
at high cardiovascular risk, such as
those with screen-detected diabetes,
should not be denied access to eective
treatments such as antihypertensives,
statins, and metformin, which were
prescribed to 45%, 16%, and 04% of
ADDITION-Europe participants, respectively, before diagnosis and to 80%,
313
Correspondence
76%, and 52% 5 years later. ADDITIONEurope might have been too small
and too brief, but the most likely
explanation for the non-signicant
17% reduction in risk of cardiovascular
events was the quality of routine care
and hence the limited dierences in
treatment between study groups. This
in turn explains the number needed
to treat (NNT) of 78 screen-detected
patients treated for 53 years to prevent
one cardiovascular event.
However, it would be wrong to infer
from this nding that screening is not
worthwhile. In assessing the potential
benets of screening, the NNT should
reect changes in treatment before
and after diagnosis (as shown above),
the duration of the lead time between
detection by screening and clinical
diagnosis, the likelihood of treatment
irrespective of a diagnosis of diabetes,
and the eectiveness of treatment.
Our estimates suggest that around
ten patients with screen-detected
diabetes would need to be treated for
10 years to prevent one event.4
We have reduced the uncertainty,
but whether screening represents an
ecient use of scarce resources remains
unclear. Tackling the growing burden of
diabetes requires a balanced investment
in primary prevention, better care for
those with the disorder, and perhaps
also in some settings earlier detection
via screening in high-risk groups.5
Corbis
314
Cited or read?
The need to establish quantitative
and qualitative indices of scientic
production by a researcher or
institution is hotly debated. Worldwide, the impact factor and H-index
are regarded as the best available.1,2
The advantages and disadvantages
have been widely discussed.3 However,
the question might arise: is an article
written to be cited or to be read?
In 2009, I published an article in the
American Journal of Medical Genetics
on non-immune hydrops fetalis.4
The journal has an impact factor
of 2505 and, as of June, 2011, the
article had been cited seven times
(two self-citations), according to ISI
Web of Science. However, between
publication and June, 2011, the article
was downloaded 3167 times from
the journals website (data from
the publisher)the highest number
of any article during that period. I
assume that many colleagues dealing
with hydrops fetalis read the article
Carlo Bellini
carlobellini@ospedale-gaslini.ge.it
Neonatal Intensive Care Unit, Gaslini Childrens
Hospital, University of Genoa, Department of
Pediatrics, 16147 Genova, Italy
1
Department of Error
Williams HC, Dellavalle RP, Garner S. Acne vulgaris.
Lancet 2012; 379: 36172In this Seminar
(Jan 28), an aliation was missing for Sarah
Garner. The aliation should be The
Commonwealth Fund, New York, NY, USA. This
correction has been made to the online version
as of Jan 27, 2012, and to the printed Seminar.
Correspondence
New antithrombotic
drugs for atrial
brillation: caution is
needed
Authors reply
We thank Sarah Bell and colleagues
for reinforcing the point that approval
processes should be optimised, in
terms of delay, without decreasing the
rigour of these processes.
We are aware of several recently
published cases emphasising that
prescription of dabigatran could be
deleterious, particularly in elderly
patients with poor renal function and
low bodyweight.1
Guideline developers should consider exclusion criteria as well as the
characteristics of patients enrolled in
randomised controlled trials. Bell and
colleagues legitimately underline that
patients with a calculated creatinine
clearance lower than 30 mL/min
were excluded from the RE-LY trial,
and that the study recruited very
few patients older than 80 years.2
We would add that this was the case
for all trials of new antithrombotic
drugs in atrial brillation,24 and that
additionally, in RE-LY, only 002% had
a bodyweight less than 50 kg,2 whereas
Submissions should be
made via our electronic
submission system at
http://ees.elsevier.com/
thelancet/
e24
Correspondence
Published Online
November 18, 2011
DOI:10.1016/S01406736(11)61756-X
e25
3
4
Correspondence
165
Price
Pre-intervention
trend
Volume
Pre-intervention
trend
155
140
120
145
100
135
80
125
115
105
60
40
095
20
085
075
0
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Year
Figure: Price and volume of artemether-lumefantrine in the public sector before and after the Global Fund and WHO actively facilitate
the switch to ACTs
ACTs=artemisinin-based combination therapies. In 2004, the Global Fund, working with WHO, provided many of its recipient countries with
the impetus and nancing necessary to rapidly change from suboptimal therapies (eg, chloroquine) to ACTs.
5
e26
Articles
Summary
Background D2 gastrectomy is recommended in US and European guidelines, and is preferred in east Asia, for
patients with resectable gastric cancer. Adjuvant chemotherapy improves patient outcomes after surgery, but the
benets after a D2 resection have not been extensively investigated in large-scale trials. We investigated the eect on
disease-free survival of adjuvant chemotherapy with capecitabine plus oxaliplatin after D2 gastrectomy compared
with D2 gastrectomy only in patients with stage IIIIIB gastric cancer.
Methods The capecitabine and oxaliplatin adjuvant study in stomach cancer (CLASSIC) study was an open-label,
parallel-group, phase 3, randomised controlled trial undertaken in 37 centres in South Korea, China, and Taiwan.
Patients with stage IIIIIB gastric cancer who had had curative D2 gastrectomy were randomly assigned to receive
adjuvant chemotherapy of eight 3-week cycles of oral capecitabine (1000 mg/m twice daily on days 1 to 14 of each
cycle) plus intravenous oxaliplatin (130 mg/m on day 1 of each cycle) for 6 months or surgery only. Block
randomisation was done by a central interactive computerised system, stratied by country and disease stage.
Patients, and investigators giving interventions, assessing outcomes, and analysing data were not masked. The
primary endpoint was 3 year disease-free survival, analysed by intention to treat. This study reports a prespecied
interim ecacy analysis, after which the trial was stopped after a recommendation by the data monitoring committee.
The trial is registered at ClinicalTrials.gov (NCT00411229).
Findings 1035 patients were randomised (520 to receive chemotherapy and surgery, 515 surgery only). Median follow-up
was 342 months (254417) in the chemotherapy and surgery group and 343 months (256419) in the surgery only
group. 3 year disease-free survival was 74% (95% CI 6979) in the chemotherapy and surgery group and 59% (5364) in
the surgery only group (hazard ratio 056, 95% CI 044072; p<00001). Grade 3 or 4 adverse events were reported in
279 of 496 patients (56%) in the chemotherapy and surgery group and in 30 of 478 patients (6%) in the surgery only
group. The most common adverse events in the intervention group were nausea (n=326), neutropenia (n=300), and
decreased appetite (n=294).
Interpretation Adjuvant capecitabine plus oxaliplatin treatment after curative D2 gastrectomy should be considered as a
treatment option for patients with operable gastric cancer.
Funding F Homann-La Roche and Sano-Aventis.
Introduction
Gastric cancer is the second most common cause of
cancer-related mortality worldwide, with 988 000 new
cases and 736 000 deaths per year.1 Surgery is the main
treatment for operable gastric cancer; however, recurrence
rates are high (about 4080% in advanced cases).2,3 In east
Asia, particularly in Japan and South Korea, D2 gastrectomy
is the standard surgical treatment for localised gastric
cancer.4,5 In Europe, two randomised controlled studies
done in the UK6 and the Netherlands7 showed little initial
dierence between D1 and D2 surgery. However, longterm follow-up of the Dutch trial showed a reduction in
gastric cancer-specic deaths with extended surgery,8 a
nding that has been supported by ndings from a smaller
Taiwanese study.9 As a result of these ndings, D2
gastrectomy is now recommended in European10 and US
treatment guidelines11 for resectable disease.
www.thelancet.com Vol 379 January 28, 2012
Published Online
January 7, 2012
DOI:10.1016/S01406736(11)61873-4
See Comment page 291
315
Articles
Methods
Study design and patients
1035 randomised
429 completed
observation phase
22 lost to
follow-up
84 died
515 analysed
346 completed
treatment phase
18 lost to
follow-up
61 died
520 analysed
316
Procedures
All patients had curative D2 gastrectomy within 6 weeks
before randomisation. At least 15 lymph nodes were
examined to ensure adequate disease classication. All
surgeons had experience doing this type of surgery
(>50 procedures per year). To further ensure the quality of
surgery, standard operating procedures were predened
and given to all surgeons before the start of the study, and
surgery was photographed.
Patients assigned to the chemotherapy group received
eight 3-week cycles of oral capecitabine (1000 mg/m
twice daily on days 114 of each cycle) plus intravenous
oxaliplatin (130 mg/m on day 1 of each cycle). Dose
reductions or interruptions were allowed to manage
potentially serious or life-threatening adverse events.
Subsequent dose escalations for capecitabine were
allowed after two more cycles in the absence of
grade 24 toxic eects, if oxaliplatin was permanently
discontinued. For toxic eects judged to be due to only
one drug, the dose of the other drug was not modied. In
cases of oxaliplatin-related neurological adverse events,
capecitabine could be continued as monotherapy.
Oxaliplatin monotherapy was not allowed if capecitabine
was discontinued.
www.thelancet.com Vol 379 January 28, 2012
Articles
Statistical analysis
3 year disease-free survival in the oxaliplatin and
capecitabine and surgery only groups was predicted to be
650% and 562%, respectively (hazard ratio [HR] 075,
with the assumption of a 3 year dropout rate of 10% with
recruitment for 12 months). The null hypothesis (no
dierence in 3 year disease-free survival between study
groups) was tested with Cox proportional hazards
regression stratied by country and disease stage, with
covariates of age, sex, and nodal metastatic status. A
sample size of 512 patients in each group was planned to
record 385 disease-free survival events with 80% power
at a 005 signicance level by a two-sided log-rank test.
An interim ecacy analysis was scheduled for
after 257 events (668% of the number of events at
nal analysis), with stopping boundaries decided by
application of the Lan-DeMets method24 and OBrienFleming spending function25 using the actual number of
events. Interim ecacy analyses for disease-free survival
and overall survival were done with Cox proportional
hazards regression by intention to treat. All ecacy
analyses were repeated per protocol to test the sensitivity of
the results (data not shown). Safety was assessed per
protocol.
Time to event endpoints were analysed with KaplanMeier survival methods. Estimates of treatment eect
were calculated as HRs with 95% CIs. Study treatment
groups were compared with a two-sided log-rank test.
We also did a prespecied analysis of disease-free
survival in subgroups.
www.thelancet.com Vol 379 January 28, 2012
558 (116)
358 (70%)
90% (90100)
90% (90100)
162 (015)
162 (015)
114 (017)
114 (017)
AJCC/UICC23 stage
IB
0 (0%)
II
261 (51%)
253 (49%)
IIIA
184 (36%)
193 (37%)
IIIB
69 (13%)
73 (14%)
IV
1 (<1%)
1 (<1%)
0 (0%)
Tumour stage
T1
3 (1%)
8 (2%)
T2
282 (55%)
282 (54%)
T3
229 (44%)
227 (44%)
T4
1 (<1%)
3 (1%)
Tumour location*
Antrum
234 (45%)
237 (46%)
Body
172 (33%)
166 (32%)
29 (6%)
31 (6%)
Fundus
40 (8%)
46 (9%)
13 (3%)
10 (2%)
Gastro-oesophageal junction
9 (2%)
15 (3%)
Whole gastric
6 (1%)
6 (1%)
Other
12 (2%)
436 (167)
9 (2%)
450 (174)
Nodal status
N0
56 (11%)
47 (9%)
N1
308 (60%)
313 (60%)
N2
151 (29%)
160 (31%)
Data are mean (SD), n (%), or median (IQR). Intention-to-treat population; all patients were Asian.
AJCC/UICC=American Joint Cancer Committee/Union Internationale Contre le Cancer. *Antrum is the lower third,
body the middle third, and fundus the upper third. Includes multiple localisations.
317
Articles
A
Surgery only
Capecitabine and oxaliplatin
10
09
08
07
06
05
04
03
02
01
0
36
39
42
45
48
51
Number at risk
Surgery only 515 442 415 388 353 332 289 255 211 188 148 120 58
Capecitabine 520 462 442 425 409 379 332 295 246 218 166 147 73
and oxaliplatin
25
31
22
30
20
25
6
10
0
0
39
42
45
48
51
Number at risk
Surgery only 515 473 460 453 443 424 380 338 288 238 204 156 112 43
Capecitabine 520 481 468 461 451 425 396 352 306 255 217 169 120 39
and oxaliplatin
34
36
31
33
12
16
0
0
12
15
18
21
24
27
30
33
B
10
09
08
Chance of survival
07
06
05
04
03
02
01
0
0
12
15
18
21 24 27 30
Time (months)
33
36
Figure 2: 3 year disease-free survival (A) and preliminary overall survival (B) in the intention-to-treat population
Results
Figure 1 shows the trial prole. Between June, 2006, and
June, 2009, 1035 patients were randomly assigned to
receive either oxaliplatin and capecitabine (n=520) or
surgery only (n=515; intention-to-treat population). The
trial was stopped because accrual was complete and the
necessary 257 disease-free survival events had occurred.
The median duration of follow-up was 342 months
(254417) in the chemotherapy group and 343 months
(256419) in the surgery only group; 468 patients were
followed up in the chemotherapy group and 463 in the
surgery only group. Patient demographic and baseline
disease characteristics were similar in each group
(table 1).
At the cuto for the interim analysis, 106 patients (20%)
in the chemotherapy group had relapsed, developed a new
318
Articles
125
066 (033132)
South Korea
910
056 (043073)
II
515
055 (036084)
IIIA
377
057 (039082)
IIIB
143
057 (035095)
<65
766
062 (047083)
65
269
048 (030078)
Women
304
083 (054127)
Men
731
049 (036066)
N0
103
090 (041197)
N1
621
062 (044089)
N2
311
045 (031066)
<57
508
063 (046088)
57
527
052 (036075)
1035
058 (045074)
Stage of disease
Age (years)
Sex
Nodal status
Weight (kg)
Overall
Discussion
This study shows that a 6 month course of chemotherapy
after D2 gastrectomy improves 3 year disease-free survival
compared with surgery only. Chemotherapy reduced the
relative risk of disease recurrence, new disease
occurrence, or death compared with surgery alone.
Moreover, a subgroup analysis suggested that adjuvant
capecitabine and oxaliplatin was benecial for all disease
stages (II, IIIA or IIIB). The overall survival data from
our study are not yet mature; however, the data suggest
an improvement in overall survival with capecitabine and
oxaliplatin compared with surgery only. An analysis after
a median follow-up of 5 years is planned to conclusively
establish the overall survival benet of capecitabine and
oxaliplatin in this setting.
We used 3 year disease-free survival as the primary
endpoint because most relapses occur within 3 years of
surgery.26 Although 3 year disease-free survival has not yet
been formally validated as a surrogate measure,
preliminary data from the Global Advanced/Adjuvant
Stomach Tumor Research International Collaboration
(GASTRIC) group indicate that 3 year disease-free survival
is strongly correlated with 5 year overall survival, the
benchmark for judging eectiveness of adjuvant therapy
in gastric cancer.27 The clinical relevance of disease-free
survival in gastric cancer is supported by the GASTRIC
group meta-analysis of 17 trials, which showed an 18%
relative risk reduction for both disease-free survival and
overall survival with adjuvant chemotherapy compared
with surgery only in patients with resectable disease.12
www.thelancet.com Vol 379 January 28, 2012
HR (95% CI)
Country
04
06
1
2
Favours capecitabine and oxaliplatin Favours surgery only
Figure 3: 3 year disease-free survival by stratication and prognostic factors in the intention-to-treat population
All grades
Grade 3 or 4
All grades
Grade 3 or 4
253 (53%)
30 (6%)
490 (99%)
279 (56%)
20 (4%)
0 (0%)
326 (66%)
39 (8%)
1 (<1%)
300 (60%)
107 (22%)
18 (4%)
1 (<1%)
294 (59%)
23 (5%)
0 (0%)
0 (0%)
277 (56%)
12 (2%)
Diarrhoea
52 (11%)
1 (<1%)
236 (48%)
9 (2%)
Vomiting
16 (3%)
0 (0%)
191 (39%)
37 (7%)
Fatigue
4 (<1%)
11 (2%)
0 (0%)
156 (31%)
23 (5%)
Thrombocytopenia
0 (0%)
0 (0%)
130 (26%)
40 (8%)
Handfoot syndrome
0 (0%)
0 (0%)
93 (19%)
5 (1%)
Asthenia
5 (1%)
0 (0%)
87 (18%)
10 (2%)
8 (2%)
Abdominal pain
42 (9%)
2 (<1%)
85 (17%)
Constipation
17 (4%)
0 (0%)
63 (13%)
1 (<1%)
Dizziness
25 (5%)
0 (0%)
64 (13%)
3 (<1%)
3 (<1%)
Stomatitis, all
Weight decreased
Peripheral sensory neuropathy
0 (0%)
0 (0%)
59 (12%)
13 (3%)
2 (<1%)
59 (12%)
1 (<1%)
0 (0%)
0 (0%)
50 (10%)
3 (<1%)
Data are n (%). *Patients who received at least one dose of capecitabine or oxaliplatin.
319
Articles
320
Articles
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
321
Articles
Summary
Lancet 2012; 379: 32234
Published Online
December 1, 2011
DOI:10.1016/S01406736(11)61294-4
See Comment page 292
Oxford University, Department
of Primary Care Health
Sciences, Oxford, UK
(C Heneghan DPhil, A Ward PhD,
R Perera DPhil)
Correspondence to:
Dr Carl Heneghan, Oxford
University, Department of
Primary Care Health Sciences,
2338 Hythe Bridge St, Oxford,
OX1 2ET, UK
carl.heneghan@phc.ox.ac.uk
Background Uptake of self-testing and self-management of oral coagulation has remained inconsistent, despite good
evidence of their eectiveness. To clarify the value of self-monitoring of oral anticoagulation, we did a meta-analysis
of individual patient data addressing several important gaps in the evidence, including an estimate of the eect on
time to death, rst major haemorrhage, and thromboembolism.
Methods We searched Ovid versions of Embase (19802009) and Medline (19662009), limiting searches to randomised
trials with a maximally sensitive strategy. We approached all authors of included trials and requested individual
patient data: primary outcomes were time to death, rst major haemorrhage, and rst thromboembolic event. We did
prespecied subgroup analyses according to age, type of control-group care (anticoagulation-clinic care vs primary
care), self-testing alone versus self-management, and sex. We analysed patients with mechanical heart valves or atrial
brillation separately. We used a random-eect model method to calculate pooled hazard ratios and did tests for
interaction and heterogeneity, and calculated a time-specic number needed to treat.
Findings Of 1357 abstracts, we included 11 trials with data for 6417 participants and 12 800 person-years of follow-up.
We reported a signicant reduction in thromboembolic events in the self-monitoring group (hazard ratio 051; 95% CI
031085) but not for major haemorrhagic events (088, 074106) or death (082, 062109). Participants younger
than 55 years showed a striking reduction in thrombotic events (hazard ratio 033, 95% CI 017066), as did
participants with mechanical heart valve (052, 035077). Analysis of major outcomes in the very elderly
(age 85 years, n=99) showed no signicant adverse eects of the intervention for all outcomes.
Interpretation Our analysis showed that self-monitoring and self-management of oral coagulation is a safe option for
suitable patients of all ages. Patients should also be oered the option to self-manage their disease with suitable
health-care support as back-up.
Funding UK National Institute for Health Research (NIHR) Technology Assessment Programme, UK NIHR National
School for Primary Care Research.
Introduction
Oral anticoagulation with vitamin K antagonists substantially reduces the incidence of thromboembolic
events.14 Although the number of patients receiving
oral anticoagulants has consistently increased, uptake
is limited by requirements to maintain the international
normalised ratio (INR) within a narrow target range,
which includes frequent testing and appropriate dose
adjustments. Benets shown in clinical trials might not
translate into routine practice: namely the risk of major
bleeding could be high in specic populations of
patients, especially in the elderly.2
Introduction of reliable and analytically accurate
point-of-care devices allows self-testing by the patient in
the home setting.1,2 Patients can have their test result
managed by their health-care provider (self-testing) or
they can interpret their INR result, and adjust their own
dose of anticoagulant accordingly (self-management).
Previous systematic reviews46 showed that selfmonitoring is a safe intervention, which gives rise to
signicant reduction in thromboembolic events, while
reducing the risk of death. Additionally, patients spend
more time in the therapeutic range of INR than they
322
Articles
Methods
Statistical analysis
The protocol methods have been previously published.9 We used the same search strategy as for
previous reviews.46 We searched Ovid versions of Embase
(19802009) and Medline (19662009), limiting searches
to randomised trials with a maximally sensitive strategy.10
A list of search terms is shown in webappendix pp 13.
We modied these searches for the Cochrane Central
Register of Controlled Trials, the Cochrane Library,
issue 2, 2009, and Cinahl (19822009). We also searched
for trials that are still underway or unpublished (eg, UK
National Research Register and Trials Central), and handsearched reference lists of retrieved papers.
Data extraction
We approached all authors whose trials met the inclusion
criteria and requested the following data for individual
patients: date of randomisation, age, indication for
treatment, type of care, demographic and psychosocial
characteristics at randomisation including quality-of-life
measures, treatment allocation, time to death, time to
rst major haemorrhage, time to rst thromboembolic
event, and INR measurements.
Data validation
We kept original data on a secure server with a backup copy according to a prespecied data-security-agreement policy. Two researchers (CB and AF) cross-checked
trial details, summary measures, and major outcomes
were cross-checked with prespecied outcome denitions
against published articles. Any inconsistencies were
discussed with the original trialist and corrections were
made when appropriate. Requirements for authorship
were those of the International Committee of Medical
Journal Editors and a representative of each trial was
invited to an investigators meeting before publication to
discuss analysis and results.
www.thelancet.com Vol 379 January 28, 2012
1
(Sc[t]hSc[t])
323
Articles
1313 excluded
23 excluded
8 duplicates
15 not randomised trials
10 excluded
10 did not have data available
(1181 participants)
Results
11 trials included
(6417 participants)
Country
Dates of
Year of
recruitment publication
of primary
results
Age range,
Study
duration years (mean)
(months)
Total
number of
patients
Female
Atrial
brillation
Mechanical
valve
Other
Type of control
Self
manage- group care
ment
Beyth et al36*
USA
199295
2000
6594 (747)
325 (5%)
184 (57%)
54 (17%)
36 (11%)
235 (72%)
No
Primary Care
Cromheecke
et al17
Holland
1998
2000
2271 (423)
49 (1%)
21 (43%)
11 (22%)
23 (47%)
15 (31%)
Yes
Anticoagulation
clinic
Koertke et al18
Germany 199497
2001
24
1777 (597)
930 (14%)
293 (32%)
19982002
2004
20
2085 (600)
139 (2%)
41 (29%)
47 (34%)
82 (59%)
MenndezJndula et al19
Spain
200102
2005
12
1990 (635)
737 (11%)
347 (47%)
296 (40%)
285 (39%)
Vller et al38
Germany 19992001
2005
19
3685 (644)
202 (3%)
53 (26%)
202 (100%)
Yes
Primary care
Fitzmaurice
et al20
UK
200102
2005
12
1887 (651)
617 (10%)
217 (35%)
343 (56%)
97 (16%)
177 (29%)
Yes
Both
Christensen
et al21
Denmark 200203
2006
2178 (507)
100 (2%)
33 (33%)
24 (24%)
35 (35%)
41 (41%)
Yes
Both
Siebenhofer
et al23
Austria
200205
2007
36
6089 (688)
195 (3%)
81 (42%)
89 (46%)
32 (16%)
74 (38%)
Yes
Both
200306
2010
36
2390 (670)
2922 (46%)
51 (2%)
2236 (77%)||
684 (23%)
2 (<1%)
No
Anticoagulation
clinic
Kaatz et al24
USA
199899
2001
12
3087 (641)
201 (3%)
84 (42%)
86 (43%)
39 (19%)
76 (38%)
No
Anticoagulation
clinic
Totals
19922006
200010
1794 (650)
1405 (22%)
3388 (53%)
6417
930 (100%)
10 (7%)
Yes
Primary care
Yes
Primary care
Anticoagulation
clinic
Data are range (mean) or number (%). *Coumatrack monitor.Coaguchek system. Pro time microcoagulation system.Two patients were unclassied for indication. Study stopped early. ||2236 with no
mechanical heart valve, 2422 had atrial brillation, mechanical heart valve, or both.
324
Articles
A
50
Control
Self-monitoring
40
30
20
10
0
0
Number at risk
Control 3151 (84)
Self-monitoring 3266 (43)
Time (years)
2137 (39)
2232 (38)
1214 (17)
1299 (19)
771 (7)
846 (12)
275 (4)
317 (2)
0
0
B
Probability of an event (%)
50
40
30
20
10
0
0
Number at risk
Control 3151 (131)
Self-monitoring 3266 (106)
Time (years)
2109 (56)
2189 (64)
1189 (30)
1273 (31)
742 (24)
820 (21)
251 (2)
296 (1)
0
0
C
50
Probability of an event (%)
40
30
20
10
0
0
Number at risk
Control 3151 (130)
Self-monitoring 3266 (96)
Time (years)
2197 (67)
2271 (76)
1264 (48)
1346 (46)
816 (19)
884 (24)
296 (5)
334 (4)
0
0
Articles
Number needed
to treat (95% CI)
Number of patients
still at risk
Number of patients
still at risk
Thrombosis
Death
Year 1
78 (55 to 253)
4369
Year 1
101 (50 to )
Year 2
47 (33 to 154)
2513
Year 2
56 (28 to )
1921
Year 3
36 (26 to 119)
1617
Year 3
37 (18 to )
1216
Year 4
32 (23 to 104)
592
Year 4
30 (15 to )
413
Year 5
27 (19 to 87)
Year 5
26 (13 to )
Year 1
205 (94 to )
4598
Year 2
123 (59 to )
2462
Year 3
96 (44 to )
1564
Year 4
74 (34 to )
547
Year 5
70 (32 to )
Major Haemorrhage
Death
Year 1
137 (65 to )
4468
Year 2
82 (39 to )
2610
Year 3
55 (26 to )
1700
Year 4
47 (22 to )
630
Year 5
42 (20 to )
55 (41 to 116)
1721
Year 2
37 ( 28 to 78)
589
Year 3
33 ( 25 to 70 )
419
Year 4
31 ( 23 to 65)
186
Year 5
24 (18 to 50)
Year 1
127 (66 to )
1699
Year 2
65 (34 to )
565
Year 3
43 (22 to )
386
Year 4
31 (16 to )
160
Year 5
29 (15 to )
Year 1
156 (67 to )
1763
Year 2
65 (34 to )
618
Year 3
43 (23 to )
442
Year 4
31 (16 to )
202
Year 5
29 (15 to )
Major haemorrhage
Death
185 (85 to )
2386
Year 2
91 (42 to )
1860
Year 3
65 (30 to )
1163
Year 4
54 ( 25 to )
393
Year 5
49 (23 to )
Major haemorrhage
Year 1
675 (79 to )*
2344
Year 2
453 (53 to )*
1824
Year 3
342 (40 to )*
1134
Year 4
268 (32 to )*
373
Year 5
255 (30 to )*
326
Number needed to
treat (95% CI)
2428
1
(Sc[t]hSc[t])
Articles
A
Sex
Male
Female
Age (years)
<55
5564
6574
75
Indication
Mechanical valve
Atrial brillation
Other
Self-monitoring
Self-testing only
Self-management
Type of control group care
Anticoagulation clinic
Primary care
Total
Self-monitoring
Control
OE
Variance
HR (95% CI)
Interaction test
96/2453
18/539
124/2323
28/555
1757
328
5473
1113
040 (020080)
076 (036158)
p=094
11/496
33/895
44/1038
26/488
25/338
34/798
66/1117
27/538
917
217
988
102
837
1659
2706
1303
033 (017066)
080 (045142)
069 (048101)
096 (041226)
p=0052
36/1132
70/1629
8/261
63/1040
69/1623
20/237
1622
006
387
2458
3458
630
052 (035077)
067 (028157)
054 (025118)
p=0032
89/1566
25/1487
89/1557
63/1382
283
1852
4444
2157
074 (030182)
042 (028065)
p=0002
92/1959
16/995
112/1950
27/893
1146
634
5083
1044
044 (017114)
055 (028108)
p=034
114/3053
152/2939
201/2542
29/618
201/2403
43/639
986
515
9983
1756
091 (074110)
075 (047119)
p=045
25/500
66/812
78/1110
61/583
24/340
44/699
101/1185
75/640
271
667
1166
448
1179
2716
4386
3349
079 (045141)
128 (088186)
077 (057103)
087 (062123)
p=020
78/1111
126/1676
26/260
91/1015
119/1677
34/281
1011
235
690
4166
6060
1387
078 (058106)
104 (081134)
061 (036103)
p=013
190/1729
40/1487
200/1719
44/1382
1141
327
9681
2089
084 (064112)
086 (056131)
p=087
182/1959
42/1158
187/1950
46/1055
827
402
9158
2189
091 (074112)
083 (050136)
p=060
230/3216
244/3101
204/2440
43/618
219/2297
55/639
1706
281
10504
2411
085 (070103)
089 (044181)
p=084
14/495
46/874
98/1034
89/583
15/353
46/781
129/1112
84/640
356
284
1544
474
661
2282
5565
4291
058 (022157)
072 (036141)
076 (058099)
112 (083151)
p=019
64/1101
138/1555
45/334
75/1002
156/1549
43/349
966
1432
371
3454
7267
2153
080 (042152)
072 (043120)
119 (078181)
p=023
195/1729
52/1342
201/1719
73/1237
894
1204
9873
3105
091 (075111)
075 (042133)
p=015
159/1934
72/988
173/1926
90/884
1062
1302
8283
4020
082 (052128)
072 (044118)
p=020
247/3071
274/2956
051 (031085)
B
Sex
Male
Female
Age (years)
<55
5564
6574
75
Indication
Mechanical valve
Atrial brillation
Other
Self-monitoring
Self-testing only
Self-management
Type of control group care
Anticoagulation clinic
Primary care
Total
088 (074106)
C
Sex
Male
Female
Age (years)
<55
5564
6574
75
Indication
Mechanical valve
Atrial brillation
Other
Self-monitoring
Self-testing only
Self-management
Type of control group care
Anticoagulation clinic
Primary care
Total
082 (062109)
01
02
05
Favours self-monitoring
10
Favours control
Figure 3: Major outcomes by sex, age, indication, type of monitoring and control group care
Thrombosis (A), major haemorrhage (B), and death (C). OE=observed minus expected.
327
Articles
A
Self-monitoring
Self-management
Christensen21
Cromheecke17
Fitzmaurice20
Koertke18
Menndez-Jndula19
Siebenhofer23
Sunderji37
Vller38
Subtotal
Self-testing only
Kaatz24
Matchar22
Beyth36
Subtotal
Total
Control
0/50
0/25
4/337
12/488
3/368
6/99
0/69
0/101
25/1487
0/50
2/24
3/280
21/442
21/369
13/96
2/70
1/101
63/1382
2/101
87/1465
0/163
89/1566
114/3053
6/100
83/1457
0/162
89/1557
152/2939
013 (001203)
163 (033812)
050 (025100)
027 (012062)
047 (019116)
014 (001218)
017 (000875)
042 (028065)
035 (009141)
098 (073133)
074 (030182)
051 (031085)
01
02
05
10
B
Self-management
Christensen21
Cromheecke17
Fitzmaurice20
Koertke18
Menndez-Jndula19
Siebenhofer23
Sunderji37
Vller38
Subtotal
Self-testing only
Kaatz24
Matchar22
Beyth36
Subtotal
Total
0/50
0/25
5/337
25/488
3/368
6/99
0/69
1/101
40/1487
0/50
1/24
4/280
20/442
7/269
11/96
1/70
0/101
44/1382
013 (000655)
103 (028384)
110 (061198)
054 (016189)
057 (022147)
014 (000692)
701 (01435401)
086(056131)
32/101
147/1465
11/163
190/1729
230/3216
36/100
143/1457
21/162
200/1719
244/3101
081 (050132)
096 (076121)
052(026105)
084 (064112)
088 (074106)
01
02
05
10
C
Self-management
Christensen21
Cromheecke17
Fitzmaurice20
Koertke18
Menndez-Jndula19
Siebenhofer23
Sunderji37
Vller38
Subtotal
Self-testing only
Kaatz24
Matchar22
Beyth36
Subtotal
Total
1/50
0/25
13/337
18/488
5/368
15/99
0/69
0/101
52/1342
0/50
0/24
11/280
36/442
15/369
11/96
0/70
0/101
73/1237
707 (01435658)
2/101
152/1465
41/163
195/1729
247/3071
1/100
157/1457
43/162
201/1719
274/2956
193 (0201860)
091 (073114)
090 (059139)
091 (075111)
082 (062109)
098 (044219)
045 (026077)
046 (019112)
140 (065302)
075 (042133)
01
02
05
Favours self-monitoring
10
Favours control
328
Articles
A
Thrombosis
Sex
Male
Female
Age (years)
<65
65
Total
Major haemorrhage
Male
Female
Age (years)
<65
65
Total
Death
Male
Female
Age
<65
65
Total
Self-monitoring
Variance
Control
OE
28/853
8/211
53/748
10/236
1507
002
1993
440
042 (024072)
100 (039253)
p=015
22/745
14/376
36/1132
32/552
31/473
63/1040
855
694
1300
1084
052 (030089)
053 (029096)
052 (035077)
p=097
65/860
13/226
77/743
14/249
1162
101
3486
642
072 (051100)
117 (054254)
p=025
44/665
34/400
78/1111
36/499
54/484
91/1015
036
878
1925
2131
098 (063153)
066 (043101)
078 (058106)
p=021
56/852
8/232
54/735
21/252
353
538
2711
717
088 (060128)
088 (019400)
p=014
24/701
40/395
64/1101
28/518
47/476
75/1002
657
054
1237
2102
062 (027143)
097 (064149)
080 (042152)
p=016
058 (014237)
071 (025204)
p=049
091 (048173)
089 (044178)
067 (028157)
p=071
105 (081136)
089 (033243)
p=076
135 (082223)
095 (071127)
104 (081134)
p=023
083 (065105)
094 (030297)
p=091
059 (013265)
076 (049119)
072 (043120)
p=053
01
02
05
10
B
Thrombosis
Sex
Male
Female
Age (years)
<65
65
Total
Major haemorrhage
Male
Female
Age (years)
<65
65
Total
Death
Male
Female
Age (years)
<65
65
Total
65/1405
5/122
59/1382
10/129
152
118
3095
345
19/462
51/1060
70/1629
19/425
50/1093
69/1623
085
134
946
2490
119/1462
7/186
107/1438
12/200
257
043
5630
385
37/477
89/1087
126/1676
25/456
94/1120
119/1677
466
224
1537
4484
129/1341
9/186
143/1322
13/200
1263
128
6723
528
32/512
106/1030
138/1555
31/470
125/1067
156/1549
081
1304
1571
5662
01
02
05
Favours
self-monitoring
10
Favours
control
Figure 5: Major outcomes in mechanical valve and atrial brillation by age and sex
Patients with mechanical valve by sex and age (<65 years and 65 years; A) and patients with atrial brillation by sex and age (<65 years and 65 years; B).
OE=observed minus expected.
Articles
Country
7 days
Intervention
30 days
Control
Intervention
90 days
6 months
Control
Intervention
Control
578 (145)
478 (186)
1 year
Intervention
Control
Intervention
Control
Cromheecke
et al17
Holland
Krtke et al18
Germany
492% (429)
569% (413)
712% (287)
557% (355)
782% (220)
596% (325)
830% (186)
617 (308 )
MenndezJndula et al
Spain
667% (374)
625% (447)
662% (267)
682% (362)
662% (186)
685% (272)
679% (140)
688 (200)
Fitzmaurice
et al20
UK
666% (404)
532% (453)
697% (322)
639% (368)
707% (251)
636% (317)
710% (230)
634% (288)
717% (220)
638 (284)
Christensen21
et al
Denmark
802% (243)
617% (441)
783% (246)
712% (337)
775% (206)
677% (309)
755% (189)
673% (245)
755% (189)
674 (242)
Siebenhofer23
et al
Austria
436% (435 )
529% (440)
510% (366)
615% (358)
536% (262)
633% (297)
613% (199)
645 (211)
Matchar et al22
USA
633% (217)
523% (251)
635% (160)
530% (203)
641% (146)
551% (199)
651% (141)
577% (199)
672% (140)
610 (201)
Kaatz et al24
USA
597% (400)
611% (416)
566% (287)
669% (321)
629% (216)
700% (253)
647% (196)
716% (199)
659% (174)
708 (176)
7 days
Number of tests
p value
p value
<0001
77%
0001
72%
005
94%
<0001
79%
011
96%
<0001
94%
055
93%
<0001
30 days
6 months
1 year
Table 4: Mean dierence between self-monitoring and control group in time in therapeutic range and number of tests for participants with a
mechanical valve
7 days
Number of tests
p value
p value
0%
<0001
92%
091
30 days
77%
039
97%
<0001
6 months
79%
010
99%
<0001
1 year
57%
002
98%
<0001
Table 5: Mean dierence between self-monitoring and control group in time in therapeutic range and number of tests for participants with atrial brillation
Discussion
Our study used individual patient data for assessment of
self-monitoring for oral anticoagulation. Overall, we
www.thelancet.com Vol 379 January 28, 2012
Articles
Articles
332
Articles
References
1
Fitzmaurice DA, Gardiner C, Kitchen S, Mackie I, Murray ET,
Machin SJ. An evidence-based review and guidelines for patient
self-testing and management of oral anticoagulation. Br J Haematol
2005; 131: 156165.
2
Szucs TD, Bramkamp M. Pharmacoeconomics of anticoagulation
therapy for stroke prevention in atrial brillation: a review.
J Thromb Haemost 2006; 4: 118085.
3
Poller L, Keown M, Chauhan N, et al. European concerted action on
anticoagulation. Quality assessment of the CoaguChek Mini and
TAS PT-NC point-of-care whole-blood prothrombin time monitors.
Clin Chem 2004; 50: 53744.
4
Christensen TD, Johnsen SP, Hjortdal VE, Hasenkam JM.
Self-management of oral anticoagulant therapy: a systematic
review and meta-analysis. Int J Cardiol 2007; 118: 5461.
5
Heneghan C, Alonso-Coello P, Garcia-Alamino JM, Perera R,
Meats E, Glasziou P. Self-monitoring of oral anticoagulation:
a systematic review and meta-analysis. Lancet 2006; 367: 40411.
6
Garcia-Alamino JM, Ward AM, Alonso-Coello P, et al.
Self-monitoring and self-management of oral anticoagulation.
Cochrane Database Syst Rev 2010; 4: CD003839.
7
Ansell J, Jacobson A, Levy J, Voller H, Hasenkam JM.
Guidelines for implementation of patient self-testing and patient
self-management of oral anticoagulation. International consensus
guidelines prepared by International Self-Monitoring Association
for Oral Anticoagulation. Int J Cardiol 2005; 99: 3745.
8
Lip GY, Rudolf M, Kakar P. Management of atrial brillation:
the NICE guidelines. Int J Clin Pract 2007; 61: 911.
9
Perera R, Heneghan C, Fitzmaurice D. Individual patient
meta-analysis of self-monitoring of an oral anticoagulation protocol.
J Heart Valve Dis 2008; 17: 23338.
10 Dickersin K, Manheimer E, Wieland S, Robinson KA, Lefebvre C,
McDonald S. Development of the Cochrane Collaborations
CENTRAL Register of controlled clinical trials. Eval Health Prof
2002; 25: 3864.
11 Schulman S, Kearon C. Denition of major bleeding in clinical
investigations of antihemostatic medicinal products in non-surgical
patients. J Thromb Haemost 2005; 3: 69294.
12 Tinazzi A, Tierney J, Stewart L, Parmar M, Torri V. Survival
curve and hazard ratio program (SCHARP) for meta-analysis
of individual patient data.
Syst Rev Evid Action Int Cochrane Colloq 6th 1998 Baltim Md 1998;
6: 65.
13 Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR. Practical
methods for incorporating summary time-to-event data into
meta-analysis. Trials 2007; 8: 16.
14 DerSimonian R, Laird N. Meta-analysis in clinical trials.
Control Clin Trials 1986; 7: 17788.
15 Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring
inconsistency in meta-analyses. BMJ 2003; 327: 55760.
16 Altman DG, Andersen PK. Calculating the number needed to
treat for trials where the outcome is time to an event. BMJ 1999;
319: 149295.
17 Cromheecke ME, Levi M, Colly LP, et al. Oral anticoagulation
self-management and management by a specialist anticoagulation
clinic: a randomised cross-over comparison. Lancet 2000;
356: 97102.
18 Krtke H, Minami K, Breymann T, et al. INR self-management
after mechanical heart valve replacement: ESCAT (Early
Self-Controlled Anticoagulation Trial). Z Kardiol 2001;
90 (suppl 6): 11824.
19 Menendez-Jandula B, Souto JC, Oliver A, et al. Comparing
self-management of oral anticoagulant therapy with clinic
management: a randomized trial. Ann Intern Med 2005; 142: 110.
20 Fitzmaurice DA, Murray ET, McCahon D, et al. Self management
of oral anticoagulation: randomised trial. BMJ 2005; 331: 1057.
21 Christensen TD, Maegaard M, Sorensen HT, Hjortdal VE,
Hasenkam JM. Self-management versus conventional management
of oral anticoagulant therapy: a randomized, controlled trial.
Eur J Intern Med 2006; 17: 26066.
22 Matchar DB, Jacobson A, Dolor R, et al. Eect of home testing
of international normalized ratio on clinical events. N Engl J Med
2010; 363: 160820.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
333
Articles
42
43
44
45
334
46
47
48
49
Articles
Summary
Background The interest in neonatal screening for lysosomal storage disorders has increased substantially because of
newly developed enzyme replacement therapies, the need for early diagnosis, and technical advances. We tested for
Gauchers disease, Pompes disease, Fabrys disease, and Niemann-Pick disease types A and B in an anonymous
prospective nationwide screening study that included genetic mutation analysis to assess the practicality and
appropriateness of including these disorders in neonatal screening panels.
Methods Specimens from dried blood spots of 34 736 newborn babies were collected consecutively from January, 2010
to July, 2010, as part of the national routine Austrian newborn screening programme. Anonymised samples were
analysed for enzyme activities of acid -glucocerebrosidase, -galactosidase, -glucosidase, and acid sphingomyelinase
by electrospray ionisation tandem mass spectrometry. Genetic mutation analyses were done in samples with suspected
enzyme deciency.
Findings All 34 736 samples were analysed successfully by the multiplex screening assay. Low enzyme activities were
detected in 38 babies. Mutation analysis conrmed lysosomal storage disorders in 15 of them. The most frequent
mutations were found for Fabrys disease (1 per 3859 births), followed by Pompes disease (1 per 8684), and Gauchers
disease (1 per 17 368). The positive predictive values were 32% (95% CI 1652), 80% (2899), and 50% (793),
respectively. Mutational analysis detected predominantly missense mutations associated with a late-onset phenotype.
Interpretation The combined overall proportion of infants carrying a mutation for lysosomal storage disorders was
higher than expected. Neonatal screening for lysosomal storage disorders is likely to raise challenges for primary
health-care providers. Furthermore, the high frequency of late-onset mutations makes lysosomal storage disorders a
broad health problem beyond childhood.
Funding Austrian Ministry of Health, Family, and Women.
Introduction
Starting in the early 1960s, neonatal screening was the
rst organised nationwide eort to identify inborn
errors of metabolism and endocrine defects at a
presymptomatic stage. It was revolutionised by the
introduction of electrospray ionisation tandem mass
spectrometry (ESI-MS) in 1993,1 which enabled for the
rst time the identication of several disorders by
simultaneous measurement of aminoacids and
acylcarnitines.2 The increased technological capacity
means that expanded neonatal screening programmes
can now identify a broad range of disorders in which
early detection and presymptomatic treatment result in
clinical benet. A further and more controversial benet
of such expanded programmes is the opportunity to
inform parents of future reproductive risks.3
Lysosomal storage disorders are an attractive candidate for an expanded neonatal screening programme.
These disorders result in the accumulation of macromolecular substrates that would normally be degraded
by enzymes involved in lysosomal metabolism.4 Although
individual lysosomal storage disorders are believed to be
rare, their combined incidence has been estimated at
1 per 7700 livebirths for white people,5 and a third of
www.thelancet.com Vol 379 January 28, 2012
Published Online
November 30, 2011
DOI:10.1016/S01406736(11)61266-X
See Comment page 294
335
Articles
Methods
Study design and population
Blood spot samples from 34 736 newborn babies were
collected consecutively from January, 2010, to July, 2010,
during the national Austrian newborn screening
programme. In Austria, neonatal screening is routine
and centralised with more than 99% coverage
(83 million inhabitants, mean 77 496 births per year,
SD 1183 in the past 10 years). We analysed all samples
that were screened successfully with the regular
screening panel for endocrine and metabolic disorders,
including cystic brosis, additionally for four dierent
lysosomal enzyme activities.19 These were acid
-glucocerebrosidase (GBA, decient in Gauchers
disease), -galactosidase (GLA, decient in Fabrys
disease), -glucosidase (GAA, decient in Pompes
disease), and acid sphingomyelinase (ASM, decient in
Niemann-Pick disease types A and B). The samples were
not corrected for white blood cell count because, in
Austria, a second blood spot sample is taken from
preterm babies, those of very low birthweight, and those
who are sick.20 Dried blood spots from potentially
enzyme-decient infants were retested in duplicates,
and positive results were diagnostically conrmed by
subsequent mutation analyses. The ethics committee of
the Medical University of Vienna and Vienna General
Hospital approved the study (EK 478/2009).
Sequence analysis
We isolated genomic DNA directly from the dried blood
spot card using QIAamp DNA micro Kit (Qiagen,
Carlsbad, CA, USA). We amplied exons and intron to
exon boundaries of the genes GBA, GAA, ASM, and
GLA by PCR using JumpStart RED Taq Ready Mix
(Sigma-Aldrich, St Louis, MO, USA) according to the
suppliers instructions. We sequenced PCR using a
3130 Genetic Analyzer (Applied Biosystems, Foster City,
CA, USA) following standard procedures. Primers used
for PCR amplication and sequencing are listed in the
32
30
n=34 732
28 n=34 733
26
24
Enzyme activity in mol/L per h
22
20
18
16
n=34 727
14
12
n=34 735
10
8
6
4
2
n=3*
n=10
n=9 n=13
n=4 n=8
n=1
0
N S P
N P
Gauchers Niemann-Pick A/B
(GBA)
(ASM)
S P
Pompes
(GAA)
S P
Fabrys
(GLA)
Articles
Positive by rst-line
biochemical screening
Statistical analysis
We analysed all mass spectrometry data with Analyst 15
(AB Sciex, Foster City, CA, USA). We used SPSS version
160 for data analysis and we calculated all CIs according
to the Clopper-Pearson condence method.27
Conrmed by
Positive predictive
mutation analysis value % (95% CI)
False-positive
False-positive rate
per million (95% CI)
Gauchers (GBA)
50 (793)
Pompes (GAA)
80 (2899)
30 (1160)
28
32 (1652)
19
550 (330850)
Fabrys (GLA)
Niemann-Pick types A/B (ASM)
Total
60 (10210)
0 (095)
30 (1160)
38
15
40 (2457)
23
660 (420990)
Table 1: Detection of enzyme deciencies and corresponding lysosomal storage disorders (n=34 736)
Sex
Enzyme activity*
Gene
Female
10
GBA
Male
18
Male
cDNA
Protein
Mutation type
Phenotype
604CT
1226AG
Arg202X
Asn409Ser
Nonsense
Missense
Severe
Mild
GBA
680AT
680AT
Asn227IIe
Asn227IIe
Missense
Missense
Mild
Mild
04
GAA
896TC
896TC
Leu299Pro
Leu299Pro
Missense
Missense
Male
07
GAA
3213TG
1551+1GA
V480_I517del
Splicing
Splicing
Potentially mild
Very severe
Male
13
GAA
896TC
896TC
Leu299Pro
Leu299Pro
Missense
Missense
Female
13
GAA
3213TG
3213TG
Splicing
Splicing
Potentially mild
Potentially mild
Gauchers disease
Pompes disease
Fabrys disease
7
Male
05
GLA
335GA
Arg112His
Missense
Mild
Male
07
GLA
1076TC
Ile359Thr
Missense
Not known
Male
11
GLA
427GA
Ala143Thr
Missense
Cardiovascular or renal
10
Male
12
GLA
335GA
Arg112His
Missense
Mild
11
Male
13
GLA
427GA
Ala143Thr
Missense
Cardiovascular or renal
12
Female
14
GLA
427GA
Ala143Thr
Missense
Cardiovascular or renal
13
Female
17
GLA
427GA
Ala143Thr
Missense
Cardiovascular or renal
14
Male
22
GLA
427GA
Ala143Thr
Missense
Cardiovascular or renal
15
Female
25
GLA
427GA
Ala143Thr
Missense
Cardiovascular or renal
GBA=-glucocerebrosidase. GAA=-glucosidase. GLA=-galactosidase. *In mol/L per h. References are provided in webappendix pp 24.
337
Articles
Results
Minimum values for normal activity are above 40 mol/L
per h for Gauchers disease, 20 mol/L per h for Pompes
disease, 28 mol/L per h for Fabrys disease, and
13 mol/L per h for Niemann-Pick disease types A and B,
according to the 01 percentile of 5000 samples of each
enzymes activity. The coecient of variation was less
than 12% for GAA, GBA, and ASM, and less than 15% for
GLA with low and medium quality controls.
The adapted biochemical multiplex screening assay was
successful for all 34 736 samples. The 32 samples from
known aected patients had low enzyme activities that
could be clearly distinguished from the activities in
samples from healthy controls (gure 1). A detailed
overview of single enzyme activities of the controls is
given in the webappendix p 7.
This rst-line screening identied 124 samples with low
enzyme activity: 29 samples for GBA, 25 for GAA, 42 for
45
40
35
10 percentile (31 mol/L per h)
30
25
20
15
n=3
10
n=19
05
n=6
n=6
n=7
0
Normal
Male
Female
Male
Female
Neonatal screening
Aected patients
Fabrys disease
338
Discussion
Lysosomal storage disorders are only one of a new
category of disorders that will confront clinicians with
www.thelancet.com Vol 379 January 28, 2012
Articles
Gauchers disease
Pompes disease
1 per 8684
Fabrys disease
1 per 3859
Articles
Articles
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
341
Articles
Summary
Lancet 2012; 379: 34247
Published Online
January 11, 2012
DOI:10.1016/S01406736(11)61624-3
See Comment page 295
Lihir Medical Centre
International SOS, Newcrest
Mining, Lihir Island,
Papua New Guinea (O Mitj MD,
R Hays MD, A Ipai HEO,
M Penias HEO, R Paru BSc,
D Fagaho BSc); and Barcelona
Centre for International Health
Research, Hospital Clinic,
University of Barcelona,
Barcelona, Spain (O Mitj MD,
E de Lazzari MSc, Q Bassat PhD)
Correspondence to:
Dr Oriol Mitj, Department of
Medicine, Lihir Medical Center,
PO Box 34, Lihir Island, NIP,
Papua New Guinea
oriolmitja@hotmail.com
Background Yawsan endemic treponematosis and, as such, a neglected tropical diseaseis re-emerging in children
in rural, tropical areas. Oral azithromycin is eective for syphilis. We assessed the ecacy of azithromycin compared
with intramuscular long-acting penicillin to treat patients with yaws.
Methods We did an open-label, non-inferiority, randomised trial at Lihir Medical Centre, Papua New Guinea, between
Sept 1, 2010, and Feb 1, 2011. Children aged 6 months to 15 years with a serologically conrmed diagnosis of yaws
were randomly allocated, by a computer-generated randomisation sequence, to receive either one 30 mg/kg oral dose
of azithromycin or an intramuscular injection of 50 000 units per kg benzathine benzylpenicillin. Investigators were
masked to group assignment. The primary endpoint was treatment ecacy, with cure rate dened serologically as a
decrease in rapid plasma reagin titre of at least two dilutions by 6 months after treatment, and, in participants with
primary ulcers, also by epithelialisation of lesions within 2 weeks. Non-inferiority was shown if the upper limit of the
two-sided 95% CI for the dierence in rates was lower than 10%. The primary analysis was per protocol. This trial is
registered with ClinicalTrials.gov, number NCT01382004.
Findings We allocated 124 patients to the azithromycin group and 126 to the benzathine benzylpenicillin group. In the
per-protocol analysis, after 6 months of follow-up, 106 (96%) of 110 patients in the azithromycin group were cured,
compared with 105 (93%) of 113 in the benzathine benzylpenicillin group (treatment dierence 34%; 95% CI
93 to 24), thus meeting prespecied criteria for non-inferiority. The number of drug-related adverse events (all
mild or moderate) was similar in both treatment groups (ten [8%] in the azithromycin group vs eight [7%] in the
benzathine benzylpenicillin group).
Interpretation A single oral dose of azithromycin is non-inferior to benzathine benzylpenicillin and avoids the need
for injection equipment and medically trained personnel. A change to the simpler azithromycin treatment regimen
could enable yaws elimination through mass drug administration programmes.
Funding International SOS and Newcrest Mining.
Introduction
Yawsan endemic treponematosis and, as such, a
neglected tropical diseaseis re-emerging. 40 years after
a worldwide control programme almost eradicated the
disease, it has re-emerged in children in poor, rural, and
marginalised populations in parts of Africa, Asia, and
South America. Yaws is caused by Treponema pallidum
subsp pertenue, and aects mainly skin, bones, and
cartilage. The disease has a natural history in primary,
secondary, and tertiary stages. Unless diagnosed and
treated at an early stage, yaws can become a chronic,
relapsing disease, and can lead to severe deforming bone
lesions in the long term.1
Between 1952, and 1964, WHO and UNICEF led a
worldwide campaign to control and eventually eradicate
yaws and other endemic treponematoses.2 Yaws became
the second disease targeted for eradication, after
smallpox. Control programmes were established in
46 countries and, by the end of 1964, the number of cases
had reduced by 95%, from 50 million to 25 million.
However, control eorts were gradually abandoned in
342
Articles
Methods
Study setting and patients
We undertook a prospective, open-label, non-inferiority,
randomised controlled trial at Lihir Medical Centre in
Papua New Guinea between Sept 1, 2010, and
Feb 24, 2011. The Lihir islands are geographically
remote, and despite being host to a major gold-mining
operation since 1995, the living conditions and sanitation
remain basic in most areas. Yaws is still a substantial
cause of morbidity in Papua New Guinea.18,19 Monthly
reports for monitoring several indicators of infectious
diseases and maternal and child health are being
collected via forms from hospitals, health centres, and
aid posts throughout the country. The national health
department estimated the number of yaws cases to be
17 000 nationwide in 2003, and 23 000 in 2008, of which
5000 were in New Ireland province, where Lihir Island
is located, and another 5000 in the neighbouring
province of West New Britain (unpublished).
All patients examined in the outpatient medical
department and suspected to have primary-stage or
secondary-stage yaws were assessed for possible inclusion
in the study. Eligible patients were children aged 6 months
to 15 years with a rapid plasma reagin titre of at least 1 in
16 and a reactive T pallidum haemagglutination test.
Exclusion criteria were known allergy to penicillin or
macrolide antibiotics, use of antibiotics active against
T pallidum during the preceding month, and known or
suspected coexisting diseases for which additional
antibiotic treatment with drugs eective against T pallidum
would be needed (use of quinolones, sulphonamides,
trimethoprim, and metronidazole was allowed).
www.thelancet.com Vol 379 January 28, 2012
Procedures
The primary endpoint was serological cure, dened as
a decrease in the rapid plasma reagin titre by at least
two dilutions at the 6-month follow-up examination,
compared with the titre at time of treatment. For ulcers,
improvement of lesions in 2 weeks after treatment was
also needed. Secondary endpoints were the individual
components of the primary endpoint, cure rate 3 months
after treatment, and cure rates according to stage of yaws,
rapid plasma reagin titre at treatment, and history of
household exposure.
To guarantee timely follow-up of participants, we
implemented a community-based follow-up strategy.
A eld team, consisting of a physician, a laboratory
technician, and a local health worker, located patients
twice a week (for follow-up visit) at their residence by
tracking detailed locator information. All participants
were re-examined 2 weeks after treatment to assess
clinical resolution. Photographic documentation of skin
lesions was obtained at diagnosis and at the 14-day followup visit for comparison over time. Patients with
worsening ulcers were retreated with benzathine
benzylpenicillin (at the same dose). We assessed all
participants at 3 months and 6 months after treatment. A
343
Articles
Statistical analysis
255 patients enrolled
5 ineligible
3 declined to participate
2 allergic to benzathine
benzylpenicillin
250 patients underwent
randomisation
12 lost to follow-up
2 adverse events
13 lost to
follow-up
110 completed
113 completed
92 (37)
84 (33)
Male sex
54 (44%)
59 (47%)
Exposure*
26 (21%)
20 (16%)
Primary stage
50 (40%)
56 (44%)
Secondary stage
74 (60%)
70 (56%)
Persisting ulcer
18 (15%)
15 (12%)
16 (13%)
11 (9%)
Arthralgias
68 (55%)
64 (51%)
12 (10%)
10 (8%)
Clinical presentation
47 (38%)
60 (48%)
1 in 64
77 (62%)
66 (52%)
Data are mean (SD) or number (%). RPR=rapid plasma regain. *Household
exposure to other children with open skin ulcers during the previous 3 months.
Results
Figure 1 shows the trial prole. 250 patients with
serologically conrmed yaws were randomly assigned
to receive either azithromycin or benzathine benzylpenicillin. Baseline clinical and serological characteristics
of the two treatment groups were similar (table 1). Mean
age of the participants was 88 years (SD 36; range
8 months to 15 years). 42% of patients had primary yaws
(table 1). The rapid plasma reagin titre was less than 1 in
32 in 107 (43%) participants and 1 in 64 or more in
143 (57%; table 1).
www.thelancet.com Vol 379 January 28, 2012
Articles
5 cm
5 cm
Figure 2: Ulcers in patients with primary-stage or secondary-stage yaws who were re-examined 2 weeks
after treatment
(A, B) Red, moist, bedded, 5 cm ulcer on the left leg of a 9-year-old patient with primary yaws. (C, D) Partially
epithelialised tumour 2 weeks after treatment with azithromycin.
Azithromycin %
(95% CI)
Risk dierence %
Benzathine
benzylpenicillin % (95% CI)
(95% CI)
100% (965100)
Cure at 3 months
PP=per protocol. ITT=intention to treat. *Including only patients with complete follow-up and study endpoint.
Including all randomised patients; we regarded patients with missing data as having treatment failure.
Azithromycin %
(95% CI)
Benzathine
Risk dierence %
benzylpenicillin % (95% CI) (95% CI)
909% (788964)
100% (945100)
891% (782949)
966% (883991)
929% (830972)
930% (833972)
100% (908100)
944% (866-978)
100% (851100)
955% (889985)
100% (816100)
920% (850959)
345
Articles
Discussion
Our ndings show that azithromycin was non-inferior to
benzathine benzylpenicillin for the primary composite
endpoint of serological cure at 6 months and healing of
ulcers. Furthermore, the two treatment groups had
similar rates of cure at 3 month follow-up and in
subgroups dened according to demographic and
biological characteristics. These results add to previous
evidence of the suitability of use of a single dose of a drug
such as azithromycin to treat various infectious diseases
(panel). Rates of serologically dened cure at 6 months
were substantially higher than expected for both
treatments, which validates our non-inferiority hypothesis
for the estimated penicillin cure rate. All participants had
a lower rapid plasmin reagin titre at 6 months than at
3 months, including 150 (67%) patients who
seroconverted. Additionally, we did not identify any
clinical or serological relapse after cure at 6 month followup. The azithromycin regimen did not resolve active
primary lesions in four patients. However, in the three
cases of failure that could be investigated, an
immunoperoxidase stain from a skin biopsy specimen
was negative for spirochaetes; therefore, we could not
conrm the biological treatment failure.
Azithromycin was well tolerated and no major adverse
eects occurred. Of participants who were treated with
azithromycin and interviewed, 8% reported mild to
moderate side-eects that were mainly gastrointestinal.
Only two children vomited within 30 min of oral
azithromycin administration, thus negligible drug
absorption would have occurred. These children were
then re-treated with benzathine benzylpenicillin. The
small number of participants vomiting after administration emphasises the suitability of azithromycin for
mass treatment programmes.
Our study had several limitations. First, diagnostic
criteria for inclusion in the study of primary lesions
did not include a microbiological test (eg, darkeld
microscopy); therefore, the non-healing ulcers could have
had post-treatment infection by other pathogens.
However, darkeld microscopy is rarely used to diagnose
treponemal infections because rapid serological tests are
available. Second, the imprecise denition of serological
cure, which could lead to overestimations in true rates of
cure, is a major issue aecting all research on the
treatment of treponematoses. Because laboratory technicians in this study were unaware of participants
treatment assignments, this drawback should not have
biased the comparison of cure rates between the groups.
Third, 6 months of follow-up might not be sucient to
assess the results after antibiotic treatment for yaws. Four
participants in the benzathine benzylpenicillin group did
not achieve serological cure. This nding could represent
a slower than usual decline in non-treponemal test titres
346
Articles
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
347
Seminar
The cryoglobulinaemias
Manuel Ramos-Casals, John H Stone, Maria C Cid, Xavier Bosch
Lancet 2012; 379: 34860
Published Online
August 24, 2011
DOI:10.1016/S01406736(11)60242-0
Josep Font Laboratory of
Autoimmune Diseases
(M Ramos-Casals MD) and
Vasculitis Research Unit,
Department of Autoimmune
Diseases (M C Cid MD), and
Department of Internal
Medicine (X Bosch MD),
Institut Clnic de Medicina I
Dermatologia, Hospital Clnic,
Institut dInvestigacions
Biomdiques August Pi i
Sunyer, University of
Barcelona, Barcelona, Spain;
and Rheumatology Unit,
Massachusetts General
Hospital, Boston, MA, USA
(J H Stone MD)
Correspondence to:
Dr Xavier Bosch, Department
of Internal Medicine,
Hospital Clinic, Villarroel 170,
Barcelona 08036, Spain
xavbosch@clinic.ub.es
348
Cryoglobulins are immunoglobulins that precipitate in vitro at temperatures less than 37C and produce organ
damage through two main pathways: vascular sludging (hyperviscosity syndrome, mainly in type I
cryoglobulinaemia) and immune-mediated mechanisms (principally vasculitis, in mixed cryoglobulinaemia).
Cryoglobulinaemia is associated with many illnesses, which can be broadly grouped into infections, autoimmune
disorders, and malignancies; the most common cause is infection with hepatitis C virus. Mixed cryoglobulinaemic
syndrome is diagnosed when a patient has typical organ involvement (mainly skin, kidney, or peripheral nerve)
and circulating cryoglobulins. Cutaneous purpura is the most common manifestation of cryoglobulinaemic
vasculitis. The most frequently aected internal organs are the peripheral nerves, kidneys, and joints. The course
varies widely and prognosis is inuenced by both cryoglobulinaemic damage to vital organs and by comorbidities
associated with underlying diseases. More than 90% of cases of cryoglobulinaemia have a known underlying
cause; therefore treatment is focused on the cause of the disorder rather than merely symptomatic relief. Studies
suggest that both combined or sequential antiviral therapies and targeted biological treatments might be more
eective than monotherapy.
Causes
Infections
The discovery of the hepatitis C virus (HCV) in 19898
radically changed the focus of research from essential to
HCV-related cryoglobulinaemia.9,10 Ferri and colleagues11
conrmed the detection of circulating HCV-RNA in
nearly 90% of Italian patients with mixed
cryoglobulinaemia, although later studies found wide
geographical variations.1215 HCV is predominantly
associated with type II cryoglobulinaemia. The hepatitis
B virus is reported to be associated with mixed
cryoglobulinaemia.7,16 In patients infected with HIV, the
percentage with cryoglobulinaemia ranges from 7% to
17% but rises to between 35% and 64% in those coinfected with HCV.1721 Highly-active antiretroviral therapy
lowers the frequency of cryoglobulinaemia in HIV.22 Case
reports have associated cryoglobulinaemic syndrome
with a wide range of other infectious agents (table).
Autoimmune diseases
Patients with systemic autoimmune diseases can present with complications of mixed cryoglobulinaemia.
Seminar
Type I
Type II
Type III
Monoclonal lg
Most frequent
causes
Less frequent
causes
Infrequent
causes
Infections
Hepatitis C
virus
Streptococcus spp; Brucella spp; Coxiella spp; Klebsiella spp; Leishmania spp; Chlamydia spp;
Mycobacterium tuberculosis; leprosy; hepatitis A virus; cytomegalovirus; parvovirus B-19;
chikungunya virus; Epstein-Barr virus; hantavirus; plasmodium; amoebaiasis; toxoplasmosis
Autoimmune
diseases
Sjgrens
syndrome
Systemic lupus
erythematosus;
Rheumatoid arthritis
Cancer
B-cell
lymphoma
Multiple myeloma
Other causes
Alcoholic cirrhosis
Malignancy
B-cell lymphoproliferative diseases are the major cause
of cryoglobulinaemia associated with malignancy.
Type I cryoglobulinaemia is reported predominantly in
patients with Waldenstrms macroglobulinaemia,
multiple myeloma, or chronic lymphocytic leukaemia.30
Mixed cryoglobulinaemias occur mainly in B-cell
lymphomas.31 Cryoglobulins can be detected in patients
with solid cancers.32
Essential cryoglobulinaemia
Nearly 10% of cases of mixed cryoglobulinaemia are
regarded as idiopathic or essential,7 a percentage that
rises to 25% in HCV-negative patients.32 The possibility
349
Seminar
19%
USA
0%
32%
44%
37%
China
Japan
Sweden
India
patients.12 Conversely, between 12% and 56% of HCVinfected patients have circulating cryoglobulins, again
with the highest frequency in Mediterranean patients
(gure 2). The reasons for this geographic variation are
not clear. Genetic studies have identied few consistent
risk factors. The risk of mixed cryoglobulinaemia has a
well-described relation with the duration of HCV
infection, with an annual incidence of cryoglobulinaemia
of 3%.15 Patients infected with HCV who have cryoglobulinaemia have a mean disease duration roughly
twice that of patients without cryoglobulinaemia,35 with
an odds ratio for cirrhosis of 487.36
51%
Pathophysiology
Lithuania
15%
19%
Ireland
Generation of cryoglobulins
55%
UK
28%
Poland
Germany
56%
France
37%
43%
55%
Spain
Italy
Bulgaria
17%
46%
Turkey
Greece
12%
29%
Israel
Egypt
16%
Iran
30%
Kuwait
Epidemiology
The prevalence of cryoglobulinaemia remains unknown.34
Reasons include the careful laboratory technique needed
to isolate and identify cryoglobulins and the absence of a
standard clinical assessment of patients with possible
cryoglobulinaemia. Even so, because HCV infects more
than 170 million individuals worldwide, the number of
patients at risk for the complications of mixed
cryoglobulinaemia is substantial.
The prevalence of HCV infection in patients with
mixed cryoglobulinaemia ranges from 30% to nearly
100%, with the highest prevalence in Mediterranean
350
Precipitation of cryoglobulins
Reversible precipitation on exposure to low temperatures
permits laboratory detection of cryoglobulins, but the
biochemical mechanisms of this process are not fully
understood.49 However, temperature is probably not the
only factor aecting the solubility of cryoglobulins. Some
cryoglobulins precipitate in vivo at temperatures well
above the 4C at which they are precipitated in the
www.thelancet.com Vol 379 January 28, 2012
Seminar
Clinical manifestations
The percentage of patients with circulating cryoglobulins
who develop symptoms varies from 2% to 50%.7 The
most common presentation, the triad of purpura,
arthralgia, and weakness, is reported in 80% of patients
at disease onset.7,16 The development of cryoglobulinaemic
symptoms is aected by age, underlying illness (such as
HCV infection) and the characteristics of the cryoglobulins
(type II subclass, high serum concentrations).7,53
Hyperviscosity syndrome
Hyperviscosity syndrome develops mainly in patients
with type I cryoglobulinaemia associated with haematological neoplasia,54 and is very uncommon in patients
with mixed cryoglobulinaemia (<3%).16,55 The key
symptoms are neurological (headache, confusion), ocular
(blurry vision, visual loss), and rhino-otological (epistaxis,
hearing loss). Massive intratubular cryoprecipitation
leading to rapidly progressive renal failure has been
reported. The physical examination should include
funduscopy to exclude hyperviscosity-related retinal
changes, especially haemorrhages. In patients in whom
hyperviscosity syndrome is suspected, it is helpful to
www.thelancet.com Vol 379 January 28, 2012
Cryoglobulinaemic vasculitis
Flares of cryoglobulinaemic vasculitis are often, but not
always, accompanied by general symptoms such as fever,
weakness, myalgia, and arthralgia. These symptoms,
particularly when combined with purpura, strongly
suggest vasculitis. Fevers of unknown origin can occur.
Articular involvement consists mainly of joint pain in the
hands, knees, and wrists, without clinical signs of
inammation (4471%).7,16,55,58 Arthritis is reported in
fewer than 10% of patients.59 Radiographs show no
evidence of bone erosions and anticitrullinated antibodies
are negative, by contrast with what is seen in rheumatoid
arthritis.59 Weakness and fatigue are reported by more
than 50% of patients. Fibromyalgia and endocrine
processes, such as hypothyroidism and diabetes, should
be investigated because of their higher frequency in
cryoglobulinaemic patients.6062
Cutaneous purpura is probably the most characteristic manifestation of cryoglobulinaemic vasculitis
(5482%).7,16,55,58,63 The typical presentation consists of
many small petechial lesions in the legs (gure 3A);
bullous or vesicular lesions are uncommon. Purpura
can extend to the abdominal region and, less frequently,
to the upper limbs and thorax (gure 3B). Involvement
at other sites is rare. Isolated purpura has a good
prognosis, and patients usually recover spontaneously,
often in less than 1 week, leaving discoloured skin
351
Seminar
Seminar
Diagnosis
Cryoglobulin testing
There are no standardised or validated diagnostic or
classication criteria for cryoglobulinaemic vasculitis.83
Diagnosis is based on clinical, laboratory, and histopathological data. For most patients, cryoglobulinaemic
disease is diagnosed by the presence of typical organ
involvement (mainly skin, kidney or peripheral nerve)
and circulating cryoglobulins.
The diagnosis of cryoglobulinaemia requires demonstration of the presence of cryoglobulins in serum
(panel 1). Appropriate sample collection and handling is
crucial.84 Blood should be collected in prewarmed syringes
and tubes, transported, clotted, and centrifuged at
3740C, ensuring that the temperature never falls
below 37C. The serum should then be stored at 4C for
up to 7 days. Precipitation of type I cryoglobulins usually
occurs within hours. By contrast, mixed cryoglobulins,
particularly type III, can need days to precipitate.85
Expert laboratory interpretation that considers the
patient in the appropriate clinical context is essential.
Some healthy individuals have low concentrations of
cryoglobulins (<006 g/L),86,87 and mixed polyclonal
cryoglobulins often occur transiently during infection.49
On the other hand, a negative test for cryoglobulins does
not exclude cryoglobulinaemia because of the possibility
of false-negative results caused by improper sample
collection or inconsistent laboratory techniques.49
Moreover, cryoglobulin concentrations can uctuate,
depending on their in-vivo precipitation in target vessels.
Cryoglobulin should be assayed serially when there is a
high degree of suspicion of disease.49,85
Cryoglobulin concentration can be assayed indirectly
by measurement of the total protein concentration within
the cryoprecipitate and by estimation of the cryocrit.49,75
Cryoglobulin concentration is usually more than 5 g/L in
type I cryoglobulinaemia, but generally lower in types II
and III.49 Quantication of the cryocrit is important
because the amount of serum cryoprotein can correlate
with the severity of symptoms and is useful in monitoring
the response to treatment, particularly in patients with
hyperviscosity syndromes. Immunoxation of the
redissolved cryoprecipitate, if feasible, allows identication of the type of cryoglobulin. If the cryoprecipitate
is too small, the type of cryoglobulin can be estimated
indirectly by serum immunoxation. The monoclonal
component detected usually corresponds to the existing
monoclonal cryoglobulin.
Laboratory tests
Laboratory tests are a key instrument for evaluating
functional results of visceral involvement and causal
factors. Biochemical tests for renal and liver involvement
are mandatory. Low complement levels (particularly C4)
and raised titres of serum rheumatoid factor are usually
observed in mixed cryoglobulinaemias49,51,85 and can
correlate with some clinical symptoms.88
www.thelancet.com Vol 379 January 28, 2012
Histopathological ndings
Histopathological examination is usually appropriate to
conrm the diagnosis in an aected organ. Precipitated
cryoglobulins appear in vivo as hyaline thrombi that
occlude small blood vessels, including the glomerular
tuft and endoneural microvessels. Hyaline thrombi are
especially likely when the monoclonal component in
type I and II cryoglobulinaemias is abundant.67,89
Vasculitis, the typical pathological lesion of mixed
cryoglobulinaemia, consists of mixed inammatory
inltrates involving small and, less often, medium sized
vessels. Fibrinoid necrosis can occur. The most commonly
aected organs are the skin, kidneys, and peripheral
nerves.51 Skin biopsies of purpuric lesions reveal
353
Seminar
Mild/moderate disease
Severe disease
Induction phase
Antiviral therapy
+/
Corticosteroids
Maintenance
phase
Purpura, articular,
general features
Mild neuropathy
GN without
renal failure
Antiviral therapy
Cutaneous ulcers,
ischaemia
Severe neuropathy
GN with renal failure/
nephrotic syndrome
GI involvement
Refractory
Rituximab
+
Corticosteroids
Antiviral therapy
Life-threatening
Rapidly progressive GN
CNS involvement
Intestinal ischaemia
Alveolar haemorrhage
Plasma exchanges
+
Corticosteroids pulses
Refractory
+
Cyclophosphamide
or
Rituximab
Antiviral therapy
Figure 5: Proposed therapeutic algorithm for patients with HCV-related cryoglobulinaemic syndrome
according to disease severity
GN=glomerulonephritis, GI=gastrointestinal. Limitations: there is little evidence for treatment of
cryoglobulinaemic features (available therapeutic data relies mainly on uncontrolled studies and case series). There
is no standardised classication of disease severity. Recommended strategies are based on personal experience and
expert opinions.1,94,98,105,106 Corticosteroids should be used at minimum dose and period necessary; short courses and
low doses (<30 mg/day) can be used. There are no available data on immunosuppressive maintenance therapy;
similar regimens and doses to those used in other autoimmune diseases (lupus, vasculitis) are suggested. Best dose
and therapeutic regimen of rituximab is unknown. Timing between rituximab and antiviral therapy administration
(simultaneous, or 14 weeks after starting rituximab, or after 46 months) is not clear.
Outcome
The evolution of cryoglobulinaemic disease varies
widely. Roughly half of patients have chronic disease
with no involvement of vital organs. A third of patients
have moderate-to-severe disease, with chronic renal
failure or cirrhosis, and nearly 15% present with sudden
life-threatening disease.16 Patients with cryoglobulinaemic disease have a worse 10-year survival rate
compared with the rate in the general population.16,55
Risk factors for poor outcomes include male sex, age
more than 60 years, glomerulonephritis, gastrointestinal
or pulmonary involvement, chronic HCV infection and
type II cryoglobulinaemia.16,55,63,94,95
Prognosis is inuenced heavily by both cryoglobulinaemic damage to vital organs and by underlying diseases
and comorbidities. Chronic HCV infection resulting in
liver cirrhosis and cancer and haematological neoplasia are
associated with a poor prognosis. Therapeutic interventions
(glucocorticoids, immunosuppressive agents, plasma
exchange) raise the risk of infection, especially in patients
with chronic renal or liver failure. Patients with a
combination of severe processes (chronic organ failure,
infection, and vasculitis or neoplasia) are often seen and
need an integrated, multidisciplinary approach.96,97
Life-threatening cryoglobulinaemia
Glomerulonephritis results in acute renal failure in 10%
of patients67 or can evolve progressively to chronic renal
failure.16,67,94 10-year survival rates of glomerulonephritis
range between 33% and 49%,16,94 although a study reported
in 2007 recorded nearly 80% survival after 10 years67
resulting from improved therapeutic management. The
outlook is poor in men and patients with HCV infection,
high cryocrit values, low C3 values, and raised creatinine
at diagnosis.6769,94 Mortality rates associated with intestinal
ischaemia and alveolar haemorrhage are very high
(>80%),94 although a 2010 study77 suggested improved
prognosis for gastrointestinal involvement.
Risk of neoplasia
Mixed cryoglobulinaemic syndrome is traditionally
regarded as a crossroads between autoimmune disease
and cancer.98,99 Expansion of the peripheral B-lymphocyte
pool and lymphoid inltrates within the liver and bone
marrow, characteristic of overt B-cell lymphoproliferative
www.thelancet.com Vol 379 January 28, 2012
Seminar
Treatment
Conventional immunosuppression
Treatment should be modulated according to the
underlying aetiopathogenesis (hyperviscosity vs vasculitis)
and the severity of clinical presentation (gure 5).1,93,98,105,106
There are three broad strategies in the treatment of
cryoglobulinaemia: conventional immunosuppression,
antiviral treatments, and biological therapies (panel 2).
The immunosuppressive approaches used in cryoglobulinaemic vasculitis, based on high-dose glucocorticoids and cyclophosphamide, were derived mainly
from strategies used in other systemic vasculitides before
it was understood that most cases result from HCV
infection. No clinical trial has studied these agents in
cryoglobulinaemic vasculitis. However, they remain
essential to control severe disease quickly and can help to
alter the course of disease if used shrewdly as a bridge to
antiviral or biological agents. In patients with severe
cryoglobulinaemic vasculitis, one goal of therapy should
be to discontinue conventional immunosuppressive
agents within 23 months, once the major end-organ
eects have been controlled.
Patients with moderate to severe manifestations of
cryoglobulinaemic vasculitis can be treated with
glucocorticoids to control inammation rapidly.107
Regimens of methylprednisolone (0510 g/day) for
3 days followed by prednisone (10 mg/kg daily, not to
exceed 80 mg/day) are appropriate in the setting of skin
ulceration, sensorimotor neuropathy, glomerulonephritis, and other severe vasculitic manifestations.
When disease control is achieved, glucocorticoids should
be tapered to the minimum dose necessary to suppress
activity and discontinued altogether if possible.
Cyclophosphamide is used in the most severe cases of
cryoglobulinaemic vasculitis, either daily (oral, 2 mg/kg
daily) or intermittent (intravenous, 750 mg/m monthly)
regimens. Azathioprine (2 mg/kg daily) and mycophenolate mofetil (1 g twice daily) are often used instead
www.thelancet.com Vol 379 January 28, 2012
Apheresis
Both plasma exchange and plasmapheresis remove
cryoglobulins from the circulation, thereby interrupting
the immune-complex-mediated pathogenesis of cryoglobulinaemic vasculitis. These interventions are useful in
patients with immediately life-threatening disease and for
those with hyperviscosity syndrome. However, apheresis
does not treat the underlying disease and can lead to
rebound in which cryoglobulin production increases after
the cessation of apheresis. Cyclophosphamide for up to
6 weeks might be needed to prevent this rebound.
Antiviral treatments
The most robust antiviral approach to the treatment of
HCV-related cryoglobulinaemic vasculitis involves the
use of both pegylated interferon alfa and ribavirin.109111
Patients with HCV genotypes 2 and 3 respond most
favourably to this regimen, with between 75% and 90%
achieving sustained virological responses at 24 weeks,
whereas those with genotypes 1 and 4 have a lower
likelihood of achieving sustained virological responses
(4552%).112114 The recommended regimen is shown in
panel 3.114 The dose of interferon should be carefully
adjusted in cirrhotic patients, whereas the dose of
ribavirin should be adjusted in patients with renal failure
(glomerular ltration rate <60 mL/min per m).
355
Seminar
Biological therapies
The most-promising biological approach to cryoglobulinaemia used thus far is B-cell depletion with
rituximab. Peripheral B-lymphocyte depletion should
lead to a reduction in the B-cell clones that produce
356
Future perspectives
A new era in our understanding of and therapeutic
approach to cryoglobulinaemia began 20 years ago with
the discovery of the HCV. The progressive introduction
of antiviral therapies, with eradication of HCV currently
regarded as the therapeutic gold standard, has improved
survival rates.77,134 However, many aspects of the disease
www.thelancet.com Vol 379 January 28, 2012
Seminar
remain unresolved.1 The causes of essential cryoglobulinaemia are not known. Disease management
remains dicult in many patients, because many do not
respond or are intolerant to antiviral therapies. Few data
are available on the treatment of patients with essential131
or type I135,136 cryoglobulinaemia. The role of new antiviral
agents in HCV-related cryoglobulinaemia, such as
telaprevir,137 remains to be dened. Similarly, international eorts are needed to develop a set of criteria for
a homogeneous classication of patients in future
epidemiological and therapeutic studies.138
The new century has seen the emergence of B-celltargeted therapies, whose use is currently limited by the
absence of specic licensing. Two studies reported
in 2010128,129 suggest greater benet when rituximab is
added to antiviral therapy, although some researchers
support rituximab monotherapy.139 Dierences in study
designs and therapeutic schedules do not allow denitive
recommendations on the relative timing of antiviral
and B-cell-depletion strategies. B-cell-activating factor of
the family of tumour-necrosis-factor blocking agents140 and
Toll-like receptor agonists141 might be promising therapies.
Cryoglobulinaemia is characterised by a wide range of
causes, symptoms, and outcomes, with various aetiopathogenic pathways involved in organ damage. This
complex scenario results in equally complex management
considerations. The present emphasis on approaching
the dierent causes in a simultaneous or closely
sequenced manner could produce improved therapeutic
results for patients with this challenging disorder.
Contributors
All authors contributed to the literature search and writing of the
Seminar. MR-C, JHS, MCC contributed to the gures.
Conicts of interest
MCC has received consultancy fees, research grants and speakers fees
from Centocor; and speakers fees from Bristol-Myers Squibb and
Roche. JS has received consultancy fees from Genentech. XB and MR-C
declare that they have no conicts of interest.
10
11
12
13
14
15
16
17
18
19
20
21
22
Acknowledgments
We thank David Buss for his technical assistance.
References
1
Ferri C. Mixed cryoglobulinaemia. Orphanet J Rare Dis 2008; 3: 25.
2
Wintrobe MM, Buell MV. Hyperproteinemia associated with
multiple myeloma. Bull Johns Hopkins Hosp 1933; 52: 15665.
3
Lerner AB, Watson CJ. Studies of cryoglobulins I: unusual
purpura associated with the presence of a high concentration
of cryoglobulin (cold precipitable serum globulin). Am J Med Sci
1947; 214: 41015.
4
Meltzer M, Franklin EC. Cryoglobulinaemia: a study of 29 patients.
I: IgG and IgM cryoglobulins and factors eecting
cryoprecipitability. Am J Med 1966; 40: 82836.
5
Brouet JC, Clauvel JP, Danon F, Klein M, Seligmann M. Biologic
and clinical signicance of cryoglobulins: a report of 86 cases.
Am J Med 1974; 57: 77588.
6
Tissot JD, Schierli JA, Hochstrasser DF, et al. Two-dimensional
polyacrylamide gel electrophoresis analysis of cryoglobulins and
identication of an IgM-associated peptide. J Immunol Meth 1994;
173: 6375.
7
Trejo O, Ramos-Casals M, Garca-Carrasco M, et al.
Cryoglobulinaemia: study of etiologic factors and clinical and
immunologic features in 443 patients from a single center.
Medicine (Baltimore) 2001; 80: 25262.
23
24
25
26
27
28
Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M.
Isolation of a cDNA clone derived from a blood borne non-A, non-B
viral hepatitis genome. Science 1989; 244: 35962.
Bambara LM, Biasi D, Caramaschi P, Carletto A, Pacor ML.
Cryoglobulinaemia and hepatitis C virus (HCV) infection.
Clin Exp Rheumatol 1991; 9: 9697.
Ferri C, Greco F, Longombardo G, et al. Antibodies to hepatitis C
virus in patients with mixed cryoglobulinaemia. Arthritis Rheum
1991; 34: 160610.
Ferri C, Greco F, Longombardo G, et al. Association between
hepatitis C virus and mixed cryoglobulinemia. Clin Exp Rheumatol
1991; 9: 62124.
Sansonno D, Carbone A, De Re V, Dammacco F. Hepatitis C virus
infection, cryoglobulinaemia, and beyond. Rheumatology (Oxford)
2007; 46: 57278.
Tedeschi A, Barat C, Minola E, Morra E. Cryoglobulinaemia.
Blood Rev 2007; 21: 183200.
Persico M, De Marino FA, Di Giacomo Russo G, et al. Prevalence
and incidence of cryoglobulins in hepatitis C virus-related chronic
hepatitis patients: a prospective study. Am J Gastroenterol 2003;
98: 8848.
Adinol LE, Utili R, Attanasio V, et al. Epidemiology, clinical
spectrum and prognostic value of mixed cryoglobulinaemia in
hepatitis C virus patients: a prospective study. Ital J Gastroenterol
1996; 28: 19.
Ferri C, Sebastiani M, Giuggioli D, et al. Mixed cryoglobulinaemia:
demographic, clinical, and serologic features and survival in
231 patients. Semin Arthritis Rheum 2004; 33: 35574.
Cohen P, Roulot D, Ferrire F, et al. Prevalence of cryoglobulins
and hepatitis C virus infection in HIV-infected patients.
Clin Exp Rheumatol 1997; 15: 52327.
Dimitrakopoulos AN, Kordossis T, Hatzakis A, Moutsopoulos HM.
Mixed cryoglobulinaemia in HIV-1 infection: the role of HIV-1.
Ann Intern Med 1999; 130: 22630.
Fabris P, Tositti G, Giordani MT, et al. Prevalence and clinical
signicance of circulating cryoglobulins in HIV-positive patients
with and without co-infection with hepatitis C virus. J Med Virol
2003; 69: 33943.
Ramos-Casals M, Forns X, Brito-Zern P, et al. Cryoglobulinaemia
associated with hepatitis C virus: inuence of HCV genotypes,
HCV-RNA viraemia and HIV coinfection. J Viral Hepat 2007;
14: 73642.
Bonnet F, Pineau JJ, Taupin JL, et al. Prevalence of
cryoglobulinaemia and serological markers of autoimmunity
in human immunodeciency virus infected individuals:
a cross-sectional study of 97 patients. J Rheumatol 2003;
30: 200510.
Kosmas N, Kontos A, Panayiotakopoulos G, Dimitrakopoulos A,
Kordossis T. Decreased prevalence of mixed cryoglobulinaemia in
the HAART era among HIV-positive, HCV-negative patients.
J Med Virol 2006; 78: 125761.
Ramos-Casals M, Trejo O, Garca-Carrasco M, Cervera R, Font J.
Mixed cryoglobulinaemia: new concepts. Lupus 2000; 9: 8391.
Tzioufas AG, Boumba DS, Skopouli FN, Moutsopoulos HM. Mixed
monoclonal cryoglobulinaemia and monoclonal rheumatoid factor
cross-reactive idiotypes as predictive factors for the development of
lymphoma in primary Sjogrens syndrome. Arthritis Rheum 1996;
39: 76772.
Brito-Zern P, Ramos-Casals M, Bove A, Sentis J, Font J. Predicting
adverse outcomes in primary Sjgrens syndrome: identication of
prognostic factors. Rheumatology (Oxford) 2007; 46: 135962.
Baimpa E, Dahabreh IJ, Voulgarelis M, Moutsopoulos HM.
Hematologic manifestations and predictors of lymphoma
development in primary Sjgren syndrome: clinical and
pathophysiologic aspects. Medicine (Baltimore) 2009; 88: 28493.
Ramos-Casals M, Loustaud-Ratti V, De Vita S, et al, for the SS-HCV
Study Group. Sjgren syndrome associated with hepatitis C virus:
a multicenter analysis of 137 cases. Medicine (Baltimore) 2005;
84: 8189.
Garca-Carrasco M, Ramos-Casals M, Cervera R, et al.
Cryoglobulinaemia in systemic lupus erythematosus: prevalence
and clinical characteristics in a series of 122 patients.
Semin Arthritis Rheum 2001; 30: 36673.
357
Seminar
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
358
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
Seminar
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
359
Seminar
360
Seminar
Acne vulgaris
Hywel C Williams, Robert P Dellavalle, Sarah Garner
Acne is a chronic inammatory disease of the pilosebaceous unit resulting from androgen-induced increased sebum
production, altered keratinisation, inammation, and bacterial colonisation of hair follicles on the face, neck, chest,
and back by Propionibacterium acnes. Although early colonisation with P acnes and family history might have important
roles in the disease, exactly what triggers acne and how treatment aects the course of the disease remain unclear.
Other factors such as diet have been implicated, but not proven. Facial scarring due to acne aects up to 20% of
teenagers. Acne can persist into adulthood, with detrimental eects on self-esteem. There is no ideal treatment for
acne, although a suitable regimen for reducing lesions can be found for most patients. Good quality evidence on
comparative eectiveness of common topical and systemic acne therapies is scarce. Topical therapies including
benzoyl peroxide, retinoids, and antibiotics when used in combination usually improve control of mild to moderate
acne. Treatment with combined oral contraceptives can help women with acne. Patients with more severe inammatory
acne usually need oral antibiotics combined with topical benzoyl peroxide to decrease antibiotic-resistant organisms.
Oral isotretinoin is the most eective therapy and is used early in severe disease, although its use is limited by
teratogenicity and other side-eects. Availability, adverse eects, and cost, limit the use of photodynamic therapy. New
research is needed into the therapeutic comparative eectiveness and safety of the many products available, and to
better understand the natural history, subtypes, and triggers of acne.
Introduction
Acne is a disease of the pilosebaceous unithair follicles
in the skin that are associated with an oil gland (gure 1).2
The clinical features of acne include seborrhoea (excess
grease), non-inammatory lesions (open and closed
comedones), inammatory lesions (papules and pustules),
and various degrees of scarring. The distribution of acne
corresponds to the highest density of pilosebaceous units
(face, neck, upper chest, shoulders, and back). Nodules
and cysts comprise severe nodulocystic acne. This Seminar
summarises information relating to the clinical aspects of
common acne (acne vulgaris). Acne classication,
scarring, acne rosacea, chloracne, acne associated with
polycystic ovary syndrome, infantile acne, acne inversa,
and drug-induced acne have been reviewed elsewhere.310
361
Seminar
Figure 1: Normal sebaceous follicle (A) and comedo (B), and inammatory acne lesion with rupture of follicular wall and secondary inammation (C)
Reproduced, with permission, from reference 2.
Cause
Risk factors and genes associated with acne prognosis
and treatment are unclear.31,32 Twin studies have pointed
to the importance of genetic factors for more severe
scarring acne.33 A positive family history of acne doubled
the risk of signicant acne in a study of 1002 Iranian
16-year-olds,14 and the heritability of acne was 78% in rstdegree relatives of those with acne in a large study of
Chinese undergraduates.34 Acne appears earlier in girls,
but more boys are aected during the mid-teenage years.35
Acne can occur at a younger age and be more comedonal
in black children than in white children, probably from
earlier onset of puberty.36 A study of 1394 Ghanaian
schoolchildren found that acne was less common in rural
locations, but the reasons for this are unclear.37
Although earlier observational studies suggested an
inverse association between smoking and acne,38 subsequent studies have shown that severe acne increases
with smoking.19,39 Increased insulin resistance and high
serum dehydroepiandrosterone might explain the
presence of acne in polycystic ovary syndrome.40,41
Occlusion of the skin surface with greasy products
(pomade acne),42 clothing, and sweating can worsen acne.
Drugs such as anti-epileptics typically produce a
362
Disease mechanisms
Four processes have a pivotal role in the formation of
acne lesions: inammatory mediators released into the
skin; alteration of the keratinisation process leading to
comedones; increased and altered sebum production
under androgen control (or increased androgen receptor
sensitivity); and follicular colonisation by P acnes.27 The
exact sequence of events and how they and other factors
interact remains unclear.
www.thelancet.com Vol 379 January 28, 2012
Seminar
anxiety, psychosomatic symptoms, shame, embarrassment, and social inhibition,66 which improve with eective
treatment.67 Anger inversely correlates with quality of life
in acne and satisfaction with acne treatment.68 Patients
might not volunteer depressive symptoms and need
prompting during consultation. UK teenagers with acne
twice as often scored in the borderline or abnormal range
on an age-appropriate validated questionnaire of emotional
wellbeing than did those who did not have acne, and had
higher levels of behavioural diculties.69 The presence of
acne was associated with unemployment in a case-control
study of young men and women.70 One community study
of 1417-year-old Australian students reported no
association between acne and subsequent psychological
or psychiatric morbidity, a surprising nding perhaps
explained by eective treatments or personality traits.71
Acne severity and degree of psychological impairment
do not necessarily correspondmild disease in one
person can cause high degrees of psychological disability,
whereas another with more severe disease can seem less
bothered by their acne.12 Most studies assessing
psychological morbidity in acne have been cross-sectional,
and therefore unable to establish causal direction. Few
studies report the direct and indirect costs of acne.72,73
Treatment guidelines
The many over-the-counter and prescription treatments
for acne allow for a large number of potential combination treatments. A comprehensive systematic review
363
Seminar
Sebum excretion
Keratinisation
Inammation
Follicular
Proprionibacterium
acnes
Benzoyl peroxide
(+)
+++
(+)
Retinoids
++
(+)
Clindamycin
(+)
++
Antiandrogens
++
Azelaic acid
++
++
Tetracyclines
++
Erythromycin
++
Isotretinoin
+++
++
(++)
++
+++=very strong eect. ++=strong eect. +=moderate eect. (+)=indirect/weak eect. =no eect.
Benzoyl peroxide
Benzoyl peroxide is a safe and eective81 over-the-counter
preparation that has several mechanisms of action, and
should be applied to all the aected area.82 Single-agent
benzoyl peroxide works as well as oral antibiotics or a
topical antibiotic combination that included benzoyl
peroxide for people with mild-to-moderate facial acne.64 It
has greater activity than topical (iso)tretinoin against
inammatory lesions;83,84 the results of two further
underpowered trials were equivocal.85,86 Further studies
are needed, especially as combination therapy might be
better.86 Benzoyl peroxide causes initial local irritation.
Patients need to be counselled to expect irritation but
discontinue treatment if it becomes severe. Irritation will
decrease in most cases, especially if patients start applying
it every other day and then increase the frequency. Low
strength (25% or 5%) benzoyl peroxide is recommended,
since it is less irritating and there is no clear evidence that
stronger preparations are more eective.87
Topical retinoids
Treatment with tretinoin, adapalene, and isotretinoin
require medical prescriptions. Tazarotene is not licensed
in the UK for acne. All retinoids are contraindicated in
pregnancy, and women of childbearing age must use
eective contraception. Topical retinoids act on abnormal
keratinisation and are also anti-inammatory, so they
work for both comedonal and inammatory acne. Many
placebo-controlled or non-inferiority studies citing better
tolerability exist, but few trials guide practice. More trials
comparing retinoids against each other and against other
therapies are needed. Randomised controlled trials
(RCTs) have shown that higher-strength preparations
might have greater activity than lower-strength ones, but
at the expense of more irritation. All topical retinoids
induce local reactions, and should be discontinued if
severe. They do not seem to cause temporary worsening
of acne lesions,88 but can increase the sensitivity of skin
to ultraviolet light.
Topical treatments
Topical agents when used alone or in combination
eectively treat mild acne consisting of open and closed
comedones with a few inammatory lesions.77 The many
treatment options oer dierent modes of action. Although
364
Topical antibiotics
How topical antibiotics improve acne has not been
clearly dened, but they seem to act directly on P acnes
and reduce inammation. Topical antibiotics have less
www.thelancet.com Vol 379 January 28, 2012
Seminar
Mild acne
Mainly blackheads
Mixture of lesions
Start 25% BP
Build up over weeks until tolerance to irritation develops. If not enough benet when assessed at 68 weeks:
Continue with topical therapy as long as benet continues. If not, progress to oral treatments as for moderate acne
Moderate acne
(or back acne or mild acne that fails to respond to topical therapy)
Severe acne
(or moderate acne not responding to oral therapy)
If results are not good or are not sustained with the above treatments
eg, two 8-week courses of dierent oral antibiotics without signicant benet:
Proceed to oral isotretinoin early before scarring occurs (avoid wasting time
with several prolonged courses of oral antibiotics if ineective)
Counsel for adverse eects and ensure adequate contraception for women of
child-bearing potential
Figure 2: Suggested algorithm for treatment of mild, moderate, and severe acne based on our appraisal of current clinical evidence and uncertainties
Figures reproduced with permission from DermNet NZ. BP=benzoyl peroxide. *Topical combination could be benzoyl peroxide plus topical antibiotic, or topical
benzoyl peroxide plus topical retinoid.
Seminar
Combination topicals
There is accumulating information that combinations of
topical treatments with dierent mechanisms of action
work better than single agents.94 Few combinations have
been tested properly against the relevant monotherapy.
The trials tend to be methodologically awed by factors
such as suboptimal dose or frequency of monotherapy.82
Compliance can be increased with once-daily combination
products because of their convenience and faster speed
of onset,95,96 although individual generic preparations
used concomitantly might be more cost-eective.64
Benzoyl peroxide inactivates tretinoin, and the two agents
should not therefore be applied simultaneously; if used
in combination one should be applied in the morning
and one at night.
Oral contraceptives
Oral treatments
Oral antibiotics
Oral antibiotics are usually reserved for more severe
acne, acne predominantly on the trunk, acne unresponsive
to topical therapy, and in patients at greater risk of
scarring. Although antibiotics have shown eectiveness
in terms of reducing the number of inammatory lesions,
none clear acne completely. Most patients seek acne
366
Seminar
Oral isotretinoin
When given for around 20 weeks, oral isotretinoin is the
most eective medication resulting in clinical cure in
around 85% of cases.76,94,104 Relapse rates are around 21%
and are dose-dependent, the best responses being seen
with daily doses of 1 mg/kg per day or a total of 150 mg/kg
over the treatment duration. Isotretinoin is usually
reserved for severe nodulocystic scarring acne or acne
resistant to other therapies (gure 3). Research is needed
to investigate whether isotretinoin could be benecial if
used sooner for moderate cases. Isotretinoin causes
cheilitis, dry skin, nose bleeds, secondary infection,
temporary worsening of lesions, photosensitivity, and
increased serum lipids, but these are rarely severe
enough to cause treatment withdrawal.105 Other less
common side-eects might include an increased risk of
ulcerative colitis.106 Due to teratogenicity, isotretinoin
should be given with adequate contraception for women
of childbearing age. Strict regulation in the USA has
decreased legal isotretinoin prescriptions and increased
illegal buying over the internet.107 A possible link between
isotretinoin and depression is discussed later.
Figure 3: Before (A) and after (B) view of a woman with severe acne treated with a course of isotretinoin
Reproduced with permission from Amy Derick; full patient consent was received.
Seminar
Natural history
Longitudinal studies that document the natural history
of acne are needed, especially with a view to identifying
risk factors for persistent disease. It is unknown whether
early treatment (eg, at prepubertal years) can alter the
natural history of P acnes colonisation and subsequent
inammatory acne.
Seminar
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Contributors
All authors contributed to searching published works and took part in
writing the rst and subsequent drafts.
26
Conicts of interest
We declare that we have no conicts of interest.
27
References
1
Grindlay D, Williams HC. Systematic reviews on acne vulgaris
since 1999mapping by topic. http://www.library.nhs.uk/skin/
ViewResource.aspx?resID=280039&tabID=289&catID=8275
(accessed Sept 13, 2010).
2
Degitz K, Placzek M, Borelli C, Plewig G. Pathophysiology of acne.
J Dtsch Dermatol Ges 2007; 5: 31623.
3
Shalita AR. Acne: clinical presentations. Clin Dermatol 2004;
22: 38586.
4
Jacob CI, Dover JS, Kaminer MS. Acne scarring: a classication
system and review of treatment options. J Am Acad Dermatol 2001;
45: 10917.
5
van Zuuren EJ, Gupta AK, Gover MD, Graber M, Hollis S. Systematic
review of rosacea treatments. J Am Acad Dermatol 2007; 56: 10715.
6
Rosas Vazquez E, Campos Macias P, Ochoa Tirado JG,
Garcia Solana C, Casanova A, Palomino Moncada JF. Chloracne
in the 1990s. Int J Dermatol 1996; 35: 64345.
7
Roseneld RL. What every physician should know about polycystic
ovary syndrome. Dermatol Ther 2008; 21: 35461.
8
Lucky AW. A review of infantile and pediatric acne. Dermatology
1998; 196: 9597.
9
Meixner D, Schneider S, Krause M, Sterry W. Acne inversa.
J Dtsch Dermatol Ges 2008; 6: 18996.
28
29
30
31
32
33
34
35
369
Seminar
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
370
Do JE, Cho SM, In SI, Lim KY, Lee S, Lee ES. Psychosocial aspects
of acne vulgaris: a community-based study with Korean adolescents.
Ann Dermatol 2009; 21: 12529.
Hogewoning AA, Koelemij I, Amoah AS, et al. Prevalence and risk
factors of inammatory acne vulgaris in rural and urban Ghanaian
schoolchildren. Br J Dermatol 2009; 161: 47577.
Mills CM, Peters TJ, Finlay AY. Does smoking inuence acne?
Clin Exp Dermatol 1993; 18: 10001.
Klaz I, Kochba I, Shohat T, Zarka S, Brenner S. Severe acne vulgaris
and tobacco smoking in young men. J Invest Dermatol 2006;
126: 174952.
Ehrmann DA. Polycystic ovary syndrome. N Engl J Med 2005;
352: 122336.
Chen MJ, Chen CD, Yang JH, et al. High serum
dehydroepiandrosterone sulfate is associated with phenotypic acne
and a reduced risk of abdominal obesity in women with polycystic
ovary syndrome. Hum Reprod 2011; 26: 22734.
Plewig G, Fulton JE, Kligman AM. Pomade acne. Arch Dermatol
1970; 101: 58084.
Melnik B, Jansen T, Grabbe S. Abuse of anabolic-androgenic
steroids and bodybuilding acne: an underestimated health problem.
J Dtsch Dermatol Ges 2007; 5: 11017.
Tucker SB. Occupational tropical acne. Cutis 1983; 31: 7981.
Green J, Sinclair RD. Perceptions of acne vulgaris in nal year
medical student written examination answers. Aust J Dermatol 2001;
42: 98101.
Magin P, Pond D, Smith W, Watson A. A systematic review
of the evidence for myths and misconceptions in acne
management: diet, face-washing and sunlight. Fam Pract 2004;
22: 6270.
Spencer EH, Ferdowsian HR, Barnard ND. Diet and acne: a review
of the evidence. Int J Dermatol 2009; 48: 33947.
Grant JD, Anderson PC. Chocolate and acne: a dissenting view.
Missouri Med 1965; 62: 45960.
Cordain L, Lindeberg S, Hurtado M, Hill K, Eaton SB,
BrandMiller J. Acne vulgaris: a disease of Western civilization
[comment]. Arch Dermatol 2002; 138: 158490.
Thiboutot DM, Strauss JS. Diet and acne revisited. Arch Dermatol
2002; 138: 159192.
Abulnaja KO. Changes in the hormone and lipid prole of obese
adolescent Saudi females with acne vulgaris. Braz J Med Biol Res
2009; 42: 50105.
Smith RN, Mann NJ, Braue A, Makelainen H, Varigos GA.
A low-glycemic-load diet improves symptoms in acne vulgaris
patients: a randomized controlled trial. Am J Clin Nutr 2007;
86: 10715.
Tsai MC, Chen W, Cheng YW, Wang CY, Chen GY, Hsu TJ. Higher
body mass index is a signicant risk factor for acne formation in
schoolchildren. Eur J Dermatol 2006; 16: 25153.
Jeremy AH, Holland DB, Roberts SG, Thomson KF, Cunlie WJ.
Inammatory events are involved in acne lesion initiation,
J Invest Dermatol 2003; 121: 2027.
Kurokawa I, Danby FW, Ju Q, et al. New developments in our
understanding of acne pathogenesis and treatment. Exp Dermatol
2009; 18: 82132.
Zouboulis CC, Bohm M. Neuroendocrine regulation of sebocytes
a pathogenetic link between stress and acne. Exp Dermatol 2004;
13: 3135.
Zouboulis CC, Baron JM, Bohm M, et al. Frontiers in sebaceous
gland biology and pathology. Exp Dermatol 2008; 17: 54251.
Alestas T, Ganceviciene R, Fimmel S, Muller-Decker K,
Zouboulis CC. Enzymes involved in the biosynthesis of
leukotriene B4 and prostaglandin E2 are active in sebaceous glands.
J Mol Med 2006; 84: 7587.
Papakonstantinou E, Aletras AJ, Glass E, et al. Matrix
metalloproteinases of epithelial origin in facial sebum of patients
with acne and their regulation by isotretinoin. J Invest Dermatol
2005; 125: 67384.
Brown SK, Shalita AR. Acne vulgaris. Lancet 1998; 351: 187176.
Iinuma K, Sato T, Akimoto N, et al. Involvement of
propionibacterium acnes in the augmentation of lipogenesis in
hamster sebaceous glands in vivo and in vitro. J Invest Dermatol
2009; 129: 211319.
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
Seminar
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
371
Seminar
134 Hamilton FL, Car J, Lyons C, Car M, Layton A, Majeed A. Laser and
other light therapies for the treatment of acne vulgaris: systematic
review. Br J Dermatol 2009; 160: 127385.
135 Sakamoto FH, Lopes JD, Anderson RR. Photodynamic therapy for
acne vulgaris: a critical review from basics to clinical practice: part I.
Acne vulgaris: when and why consider photodynamic therapy?
J Am Acad Dermatol 2010; 63: 18393.
136 Skidmore R, Kovach R, Walker C, et al. Eects of subantimicrobialdose doxycycline in the treatment of moderate acne. Arch Dermatol
2003; 139: 45964.
137 Toossi P, Farshchian M, Malekzad F, Mohtasham N,
Kimyai-Asadi A. Subantimicrobial-dose doxycycline in the
treatment of moderate facial acne. J Drugs Dermatol 2008;
7: 114952.
372
138 Kim J. Acne vaccines: therapeutic option for the treatment of acne
vulgaris? J Invest Dermatol 2008; 128: 235354.
139 Ingram JR, Grindlay JC, Williams HC. Problems in the reporting
of acne clinical trials: a spot check from the 2009 annual evidence
update on acne vulgaris. Trials 2010; 11: 77.
140 Barratt H, Hamilton F, Car J, Lyons C, Layton A, Majeed A.
Outcome measures in acne vulgaris: systematic review.
Br J Dermatol 2009; 160: 13236.
141 UK National Health Service. UK Database of Uncertainties about
the Eects of Treatments. http://www.library.nhs.uk/DUETs/
SearchResults.aspx?searchText=acne (accessed Sept 13, 2010).
Seminar
Thalassaemia
Douglas R Higgs, James Douglas Engel, George Stamatoyannopoulos
Thalassaemia is one of the most common genetic diseases worldwide, with at least 60 000 severely aected individuals
born every year. Individuals originating from tropical and subtropical regions are most at risk. Disorders of
haemoglobin synthesis (thalassaemia) and structure (eg, sickle-cell disease) were among the rst molecular diseases
to be identied, and have been investigated and characterised in detail over the past 40 years. Nevertheless, treatment
of thalassaemia is still largely dependent on supportive care with blood transfusion and iron chelation. Since 1978,
scientists and clinicians in this specialty have met regularly in an international eort to improve the management of
thalassaemia, with the aim of increasing the expression of unaected fetal genes to improve the deciency in adult
-globin synthesis. In this Seminar we discuss important advances in the understanding of the molecular and cellular
basis of normal and abnormal expression of globin genes. We will summarise new approaches to the development of
tailored pharmacological agents to alter regulation of globin genes, the rst trial of gene therapy for thalassaemia, and
future prospects of cell therapy.
Introduction
Abnormalities in the structure and synthesis of the -like
and -like globin chains that form tetramers of
haemoglobin (22) lead to the most common forms of
inherited anaemias.1 In thalassaemia, there are defects in
the production of either the -like (-thalassaemia2) or
the -like (-thalassaemia3) globin chains. From the 1970s,
these diseases, which specically aect red blood cells,
were among the rst to be analysed with the use of
molecular biology. Their detailed characterisation has
established many of the general principles supporting
our understanding of human molecular genetics.4
Furthermore, research into globin genes has greatly
contributed to the understanding of how human gene
expression is activated and silenced during dierentiation
and development.58 Despite these advances, manipulation
of globin gene expression to ameliorate or potentially
cure the common disorders of these genes is not yet
possible. Every 2 years since 1978, leading research
groups have met at the Hemaglobin Switching
Conference to report and discuss progress. In this
Seminar we will summarise the understanding of the
molecular and cellular pathophysiology, epidemiology,
and management of -thalassaemia, which is the main
clinical problem in this specialty.1 We also review the
developments reported at the 17th Hemaglobin Switching
Conference in Oxford, UK, which oer renewed hope for
novel approaches to treat these disorders.
Published Online
September 12, 2011
DOI:10.1016/S01406736(11)60283-3
Medical Research Council
Molecular Haematology Unit,
Weatherall Institute of
Molecular Medicine, and
National Institute for Health
Research Biomedical Research
Centre, John Radclie Hospital,
Oxford, UK (Prof D R Higgs MD);
Cell and Developmental Biology,
University of Michigan Medical
School, Ann Arbor, MI, USA
(Prof J D Engel MD); and
Department of Medicine and
Department of Genome
Sciences, University
of Washington Medical School,
Seattle, WA, USA
(Prof G Stamatoyannopoulos MD)
Correspondence to:
Prof Douglas R Higgs, Medical
Research Council Molecular
Haematology Unit, Weatherall
Institute of Molecular Medicine,
John Radclie Hospital,
Headington, Oxford OX3 9DS, UK
doug.higgs@imm.ox.ac.uk
373
Seminar
Site of
erythropoiesis
A
Yolk sac
Liver
Spleen
Bone marrow
100
60
20
0
0
3
6
Post-conceptual age (months)
Birth
12
Postnatal age (months)
Adult
B
Chromosome 11
HS
-LCR
3 2
22
Fetus
22
Adult
22
Embryo
22 22
Chromosome 16
HS-40
Telomere
Promoter
Regulatory element
374
Seminar
-LCR
Stem cell
-LCR
BFUe
-LCR
RNA
-LCR
-RNA
Intermediate
erythroblast
Adult
Fetal
Promoter
Regulatory element
Histones acetylated
Histones unacetylated
DNA methylated
DNA unmethylated
Seminar
Point mutations
causing HPFH
HS
-LCR
3 2
About 200
point mutations
1
Small deletions
removing the gene
Large deletions causing
thalassaemia and HPFH
Upstream deletions
Relocation of enhancer
Trans-acting factors
identied in GWAS
and twin studies
BCL11A and HBS1L-MYB
Promoter
Regulatory element
Trans-acting mutations
TFIIH
GATA1
EKLF
BCL11A
HBS1L-MYB
DRED TRF2/TRF4
HS
-LCR
3 2
Promoter
SOX6
KLF1
GATA1
NFE2
BCL11A
NFE4
KLF1
KLF1
GATA1
NFE2
GATA1
MBD2
PRMT3
Regulatory element
Ikaros-PYR
BRG1
376
Epidemiology
WHO has estimated that about 15% of the worlds
population might be carriers of -thalassaemia (/T)
and that about 60 000 severely aected infants are born
every year.44 These individuals mostly originate from the
Mediterranean, Middle East, central Asia, India, and
southern China, which suggests that there could be a
selective advantage to carrying such a mutation in these
areas. Similar observations have been made for
thalassaemia,1 which is even more widely distributed
and more frequent than -thalassaemia. Findings from
extensive microepidemiological and case-controlled
studies45 strongly suggest that individuals with either
-thalassaemia or -thalassaemia trait are somewhat
protected in areas where Falciparum malaria is or has
been endemic, thus explaining the high carrier
frequency via natural selection. Frequent consanguinity
might also contribute to the prevalence of thalassaemia
in many of these areas.1,46
www.thelancet.com Vol 379 January 28, 2012
Seminar
A
genes
genes
mRNA
mRNA
globin
globin
HbA
Anaemia
Bone deformities
Bone-marrow expansion
Hypermetabolic state
Iron accumulation
377
Seminar
Seminar
Mobilisation
C
Viral vector
CD34+ cell
purification
PBMCs
Non-ablative myelosuppression
Quality control
Figure 6: An outline of the protocol used for gene therapy in patients with thalassaemia
(A) The patient is rst treated with G-CSF to mobilise HSCs into the peripheral blood. (B) Peripheral blood mononuclear cells are then collected and CD34+ cells
(enriched for HSCs) are harvested. (C) These cells are cultured with a viral vector that is designed to express high levels of normal human globin. (D) The patient
then undergoes non-ablative myelosuppression and is engrafted with the virally-modied HSCs (lled circles). PBMC=peripheral blood mononuclear cell.
G-CSF=granulocyte colony-stimulating factor. HSC=haemopoietic stem cell.
379
Seminar
Som
Somatic
matic cel
cell
Transplant
nspllant
gene-corrected
e-co
orreected
cellss
B
E
Reprogramme by viral
transduction with key
transcription factors
Dierentiate to HSC
Patient-specic
iPS cell cones
Figure 7: An outline of the proposed use of iPS cells for therapy in patients with thalassaemia
Somatic cells from the patient (A) are reprogrammed by expression of key transcription factors that are associated
with pluripotency (B) to make patient-specic iPS cells (C). In thalassaemia, the defect in the -globin gene is then
corrected (D). These corrected iPS cells are then dierentiated into haemopoietic stem cells (E) and transplanted
(F) back into the patient after myelosuppression. iPS cell=induced pluripiotent stem cell.
Seminar
Conclusions
Despite intensive clinical and scientic investigation of
thalassaemiaa molecular disease that is perhaps better
understood than any otherattempts to improve its
management and to develop targeted drug therapy have
not yielded a clear breakthrough. Stem-cell transplantation is an eective cure but still has a substantial
risk of mortality and morbidity. Supportive results
from cord-blood transplantation should encourage the
development of cord-blood banking to address this issue.
Gene therapy is now entering early-phase clinical trials,
and cell therapy with iPS cells is an exciting prospect
with many challenges to overcome. Even if these
approaches are successful, such therapy might not be
immediately applied to most patients in low-income
countries. Developments in the understanding of the
molecular circuitry involved in the transition from fetal
to adult globin-gene expression might facilitate the
development of new drugs to manipulate this switch.
Even so, the cost of what would presumably be lifelong
medication could rule out therapy for all but those who
www.thelancet.com Vol 379 January 28, 2012
381
Seminar
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
382
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
Seminar
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
383
Case Report
384