Correspondence

patients, including non-EVD patients Laos

requiring transfer from an ETU. Ubon Champasak
Thailand Ratchatani
KKA declares a speaker fee from Biocartis. All other
authors declare no competing interests. Phusing

*Robert Colebunders, Shevin T Jacob,

Kevin K Ariën, Anja De Weggheleire,
Tom Decroo Bangkok

robert.colebunders@uantwerpen.be
Pailin
Global Health Institute (RC), Department of
Biomedical Science (KKA), University of Antwerp, Cambodia
Antwerp 2610, Belgium; Division of Allergy and Vietnam
Infectious Diseases, University of Washington,
Seattle, WA, USA (STJ); and Department of
Biomedical Sciences (KKA), Department of Clinical
Sciences (ADW, TD), Institute of Tropical Medicine,
Antwerp, Belgium Binh Phuoc

1 Waxman M, Aluisio AR, Rege S, Levine AC.

Characteristics and survival of patients with
Ebola virus infection, malaria, or both in
Sierra Leone: a retrospective cohort study.
Lancet Infect Dis 2017; 17: 654–60.
2 Ingelbeen B, Bah EI, Decroo T, et al.
Mortality among PCR negative admitted Ebola
suspects during the 2014/15 outbreak in
Conakry, Guinea: a retrospective cohort study.
PLoS One 2017; 12: e0180070.
3 Brown CS, Mepham S, Shorten RJ. Ebola virus
disease: an update on epidemiology,
symptoms, laboratory findings, diagnostic
issues, and infection prevention and control
issues for laboratory professionals.
Clin Lab Med 2017; 37: 269–84.
4 O’Shea M, Clay KA, Craig DG, et al. Diagnosis of
febrile illnesses other than Ebola virus disease
at an Ebola treatment unit in Sierra Leone.
Clin Inf Dis 2015; 61: 795–98.
5 Elston JWT, Cartwright C, Ndumbi P, Wright J.
The health impact of the 2014–15 Ebola
outbreak. Public Health 2017; 143: 60–70.
Spread of a single
multidrug resistant
malaria parasite lineage
(PfPailin) to Vietnam
The spread of artemisinin resistance
in Plasmodium falciparum and
the subsequent loss of partner
antimalarial drugs in the Greater
Mekong subregion1 presents one of
the greatest threats to the control and
elimination of malaria. Artemisinin
resistance is associated with mutations
in the PfKelch gene. Initially multiple
independent Kelch mutations were
observed,1 but in a recent sinister
development, a single dominant
artemisinin-resistant P falciparum
C580Y mutant lineage has arisen in
0 125 250 500 km
Figure: Transnational spread of multidrug resistant PfPailin
The artemisinin resistant Plasmodium falciparum C580Y lineage (PfPailin) was detected first in Pailin,
Western Cambodia, in 2008.2 It later acquired piperaquine resistance and spread east. 8 years later it has
now reached the south of Vietnam encompassing all four countries of the Eastern Greater Mekong
subregion.
western Cambodia, outcompeted the (figure).4 Microsatellite typing of 86 of
other resistant malaria parasites, and 152 P falciparum isolates from the Binh
subsequently acquired resistance to Phuoc locality in 2016 shows the same
piperaquine.2 Cambodia had adopted flanking sequence surrounding the
dihydroartemisinin-piperaquine as PfKelch C580Y gene as that observed
first-line antimalarial treatment, but in parasites from the affected areas
has now been forced to switch its of the other three Greater Mekong
first line artemisinin combination subregion countries.2 The evolution
treatment back to artesunate- and subsequent transnational
mefloquine as a consequence3. This spread of this single fit multidrug-
dominant multidrug-resistant parasite resistant malaria parasite lineage is of
lineage, identified first in Pailin in international concern.
western Cambodia and tentatively We declare no competing interests. This study was
denoted as PfPailin, then spread to supported by Mahidol University, Thailand and the
Wellcome Trust. Some of the samples were from the
northeastern Thailand and southern TRAC study supported by the UK Department for
Laos2. We now find that the PfPailin International Development (DFID).
lineage, with associated piperaquine
Mallika Imwong, Tran T Hien,
resistance (evidenced by amplification
Nguyen T Thuy-Nhien,
in the PfPlasmepsin2 gene), has spread Arjen M Dondorp, *Nicholas J White
to the south of Vietnam where it is nickw@tropmedres.ac
responsible for alarming rates of failure
Department of Molecular Tropical Medicine and
of dihydroartemisinin-piperaquine— Genetics (MI) and Mahidol-Oxford Tropical Medicine
the National first-line treatment Research Unit (AMD, NJW), Faculty of Tropical

1022 www.thelancet.com/infection Vol 17 October 2017

 Correspondence

Medicine, Mahidol University, Bangkok 10400,
Thailand; Oxford University Clinical Research
Unit – Hospital for Tropical Diseases, Ho Chi Minh
City, Vietnam (TTH, NTT-N); and Centre for Tropical
Medicine and Global Health, Nuffield Department of
Medicine, University of Oxford, Oxford, UK
(AMD, NJW)
1 Ashley EA, Dhorda M, Fairhurst RM, et al.
Tracking Resistance to Artemisinin
Collaboration (TRAC). Spread of artemisinin
resistance in Plasmodium falciparum malaria.
N Engl J Med 2014; 371: 411–23.
2 Imwong M, Suwannasin K, Kunasol C, et al.
The spread of artemisinin-resistant
Plasmodium falciparum in the Greater Mekong
Subregion: a molecular epidemiology
observational study. Lancet Infect Dis 2017;
17: 491–97.
3 World Health Organization. Status report on
artemisinin and ACT resistance (April 2017).
http://www.who.int/malaria/publications/
atoz/artemisinin-resistance-april2017/en
(accessed July 11, 2017).
4 Thanh NV, Thuy-Nhien N, Tuyen NT, et al.
Rapid decline in the susceptibility of Plasmodium
falciparum to dihydroartemisinin-piperaquine in
the south of Vietnam. Malar J 2017; 16: e27.
WHO’s recommendation
for surgical skin
antisepsis is premature
The new WHO guidelines on
prevention of surgical site infections1
recommend chlorhexidine-alcohol
rather than aqueous povidone-iodine
or povidone-iodine with alcohol
for surgical skin preparation. This
recommendation was provided as a
“strong recommendation” with “low
to moderate” quality of evidence.
One of us (AFW) was a member
of the guidelines development
group that formulated the WHO
recommendations. However, we are
now concerned with the completeness
and quality of the evidence that
led to the chlorhexidine-alcohol
recommendation.
It is clear that alcohol-based
antiseptics with either chlorhexidine
or iodine are better in terms of clinical
and antimicrobial effectiveness
than are aqueous ones. 1,2 WHO’s
recommendation now effectively
no longer supports use of povidone-
iodine-alcohol for surgical site
preparation. However, povidone-
iodine-alcohol has been the standard
of care in European hospitals for
decades and associated with low
surgical infection rates. A worldwide
change of practice, as advocated by
WHO, however, should be supported
by strong and high-quality evidence.
The final meta-analysis that led to
WHO’s recommendation included
six trials of chlorhexidine-alcohol
versus iodine-alcohol preparations
(rated as overall “low” quality) that
showed significance in favour of
chlorhexidine-alcohol. However, one
trial included a solution with only
23% isopropanol in the iodine-alcohol
group, which is clearly below the
established microbicidal concentration
range (about 50–90%, depending on
alcohol species). Two other trials had
unknown (and irretrievable) alcohol
concentrations in their antiseptic
preparations, and two further trials
(adding up to five trials) had small
sample sizes (n=100 each), leading
to only one surgical site infection.
Another trial, published in 2016 after
WHO’s literature inclusion period, was
“exceptionally included”, stating that
the committee was confident that
no additional relevant trial had been
published.
Clearly, with unknown active
ingredient concentrations or
concentrations below the active
range, participation in the microbicidal
process simply cannot be assumed and
therefore the information coming
from such trials is of uncertain
usefulness. In a previous meta-
analysis,2 we excluded such trials.
Furthermore, one published trial3 of
substantial size from 2015 was not

Within Appropriate Included Included Study Chlorhexidine- Iodine-alcohol Weight Risk ratio
WHO ingredients† in WHO in this alcohol (n/N)‡ (n/N)‡ (95%)§
inclusion analysis analysis
period*

No No Yes No Berry et al (1982) 44/453 61/413 NA 0·66 (0·46–0·95)

Yes Yes No Yes Ostrander et al (2005)¶ 1/40 0/40 0·3% 3·00 (0·13–71·51)
Yes No Yes No Veiga et al (2008) 0/125 4/125 NA 0·11 (0·01–2·04)
Yes No Yes No Cheng et al (2009) 0/25 0/25 NA Not estimable
Yes Yes Yes Yes Saltzman et al (2009) 0/50 0/50 0% Not estimable
Yes Yes Yes Yes Savage et al (2012) 1/50 0/50 0·3% 3·00 (0·13–71·92)
No Yes No Yes Ngai et al (2015)¶ 21/474 21/463 13·8% 0·98 (0·54–1·76)
No No No No Salama et al (2016)¶ 7/189 21/173 NA 0·31 (0·13–0·70)
No Yes Yes Yes Tuuli et al (2016) 23/572 42/575 27·3% 0·55 (0·34–0·90)
No Yes No Yes Broach et al (2017)¶|| 82/392 90/396 58·3% 0·92 (0·71–1·20)

Overall** 128/1578 153/1574 100·0% 0·84 (0·68–1·04)

Heterogeneity I2=16%
Test for overall effect p=0·12 0·1 0·2 0·5 1 2 5 10

Favours chlorhexidine-alcohol Favours iodine-alcohol

Figure: Meta-analysis of randomised clinical trials of the effect of chlorhexidine-alcohol versus iodine-alcohol preparations for surgical skin antisepsis on surgical site infections
Relevant details of the WHO meta-analysis and the appropriateness of antiseptic ingredients are shown. Further details and references are available in the appendix. NA=not applicable.
*WHO had specified an inclusion period between 1990 and Aug 15, 2014. †We assessed appropriateness of antiseptics on the basis of whether or not the concentrations of all active
ingredients were known, within published microbicidal ranges, and consistent with product information. ‡Alcohol was ethanol or isopropanol. §Risk ratios are fixed-effects Mantel-Haenszel
risk ratios. ¶Identified through additional literature searches (cutoff date of May 1, 2017). ||This article was published after the WHO guideline. **Only data from studies included in this analysis
are included.

www.thelancet.com/infection Vol 17 October 207 1023