R

EiL

RMGFa

b

c

d

e

a

A

KSSRLPM

C

0h

Vaccine 30S (2012) G11– G17

Contents lists available at SciVerse ScienceDirect

Vaccine

jou rn al h om epa ge: www.elsev ier .com/ locate /vacc ine

eview

pidemiologic characteristics, serotypes, and antimicrobial susceptibilities ofnvasive Streptococcus pneumoniae isolates in a nationwide surveillance study inebanon

ima Hanna-Wakima,b,1, Hiba Chehaba,b,1, Imane Mahfouza,b,1, Farah Nassarc, Maysa Baroudb,arwa Shehabb, Guillermo Pimenteld, Momtaz Wasfyd, Brent Housed, George Arajb,e,hassan Matarb,c,1, Ghassan Dbaiboa,b,∗,1,or the Lebanese Inter-Hospital Pneumococcal Surveillance Program2

Department of Pediatrics and Adolescent Medicine, American University of Beirut, Beirut, LebanonThe Center for Infectious Diseases Research, American University of Beirut, Beirut, LebanonDepartment of Experimental Pathology, Immunology and Microbiology, American University of Beirut, Beirut, LebanonThe United States Naval Medical Research Unit 3, Cairo, EgyptDepartment of Pathology and Laboratory Medicine, American University of Beirut, Lebanon

r t i c l e i n f o

rticle history:

eywords:. pneumoniaeerotypeesistanceebanonediatriciddle East

a b s t r a c t

Invasive pneumococcal disease (IPD) associated with Streptococcus pneumonia is a major public healthproblem worldwide for all age groups, including in Lebanon. Prevention through vaccination remainsthe most valuable tool to decrease the burden of disease. Pneumococcal conjugate vaccine 7 (PCV7),marketed internationally including in the Middle East and North Africa region for the prevention of IPD,was introduced in Lebanon in 2006, followed by PCV10 and PCV13 in 2010. However, none of these iscurrently part of the Extended Program of Immunization schedule and published data on IPD incidence,pneumococcal serotypes and vaccine coverage in the region are lacking. The Lebanese Inter-HospitalPneumococcal Surveillance Program is a surveillance system set up to determine the burden of IPD andthe prevalent serotypes responsible. The aim of this prospective 6-year study carried out in 78 hospitalsthroughout Lebanon was to obtain such data to help health authorities make informed decisions on theimplementation of pneumococcal vaccination at the national level. A total of 257 isolates of culture-confirmed Streptococcus pneumoniae were evaluated. Considering all age groups, vaccine coverage was41.4%, 53.9%, and 67.2% for PCV7, PCV10, and PCV13 serotypes, respectively; for patients <2, 2–5, and>60 years of age, PCV7 coverage was 50%, 51%, and 35%, respectively; PCV10 coverage was 53%, 74%,45%, respectively; and PCV13 coverage was 63%, 80%, and 68%, respectively. Overall, 17.4% of these iso-lates were penicillin-G non-susceptible using the latest established breakpoints and mortality occurredin 23.5% of the patients with non-susceptible isolates. In addition, 10.9% of isolates were multi-drug-

resistant. The highest mortality rates were observed in the eldest (>60 years of age) and youngest (<2years of age) patients. The most prevalent invasive serotypes identified were those found in currentlyavailable pneumococcal conjugate vaccines, emphasizing the importance of implementing the vaccine inthe routine immunization schedule at the national level. Continuation of current surveillance practices will help assess the impact of vaccine implementation on IPD epidemiology, serotype distribution andantibiotic resistance patterns.

© 2012 Elsevier Ltd. All rights reserved.

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author at: Department of Pediatrics and Adolescent Medicine, AmericPO Box 11-0236, Beirut, Lebanon. Tel.: +961 1 350000x5752; fax: +961 1370781.

E-mail address: gdbaibo@aub.edu.lb (G. Dbaibo).1 These authors contributed equally to this work.2 See Appendix A for contributors of Lebanese Inter-Hospital Pneumococcal Surveillanc

264-410X/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.vaccine.2012.07.020

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G12

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G12

an University of Beirut Medical Center, Cairo Street, Room C-649,

e Program.

G12 R. Hanna-Wakim et al. / Vaccine 30S (2012) G11– G17

3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G133.1. Descriptive epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G133.2. Serotype distribution and vaccine coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G133.3. Antimicrobial susceptibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G133.4. Statistics by age group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G13

4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G135. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G15

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G16Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G16

. . . . . .

1

tseamydtmUbmfmpm

bvccssj1aa2tbGdysa(yctfmPrtlr

Mdc

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. Introduction

Streptococcus pneumoniae is a major cause of morbidity and mor-ality worldwide. It can cause a spectrum of diseases that range ineverity from simple otitis media to invasive pneumococcal dis-ase (IPD), including sepsis and meningitis [1]. Both children anddults suffer from the burden of pneumococcal disease. It is esti-ated that 14.3 IPD cases occurred per 100,000 in the USA in the

ear 2009 [2]. The WHO estimates that, worldwide, 1.6 millioneaths are attributed to pneumococcal disease annually. Most ofhese deaths occur in children younger than 5 years of age (0.7–1

illion deaths annually) and in underdeveloped countries. In theSA and Europe, the most common cause of community-acquiredacterial pneumonia in adults is S. pneumoniae [3]. Hence, pneu-ococcal infections are a major public health problem worldwide

or all age groups. Moreover, identification of appropriate treat-ent strategies is a challenge for physicians due to the rise in

enicillin-resistant S. pneumoniae isolates and the emergence ofultiple drug-resistant strains [4].Prevention remains the most valuable tool to help decrease the

urden of disease. The development of pneumococcal conjugateaccines has had a significant impact on the burden of disease inountries in which they were introduced [5]. These vaccines includeapsular polysaccharides of the most prevalent disease-causingerotypes among more than 90 different known pneumococcalerotypes conjugated to a carrier protein. In 2000, the 7-valent con-ugated vaccine (PCV7), containing the serotypes 4, 6B, 9V, 14, 18C,9F, and 23F, which were the most prevalent in the United Statesnd Europe, was introduced. PCV10, adding serotypes 1, 5, and 7F,nd PCV13, adding serotypes 3, 6A, and 19A, were introduced in009 and 2010, respectively [6]. Conjugated vaccines have proveno be safe, immunogenic, and effective in preventing IPD causedy serotypes contained in the vaccine. In a study done in Atlanta,eorgia in the USA during the 2003–2004 season, there was a 96%ecrease in the rate of IPD due to PCV7 serotypes in children <5ears old and 68–72% reduction for adult age groups in compari-on to the pre-vaccine period [5]. In another study from the USA,

significant decline was also noted in penicillin non-susceptiblePNS) IPD rates by 64% and 45% from 1998 to 2008 in children <5ears and adults 65 years and older respectively [7]. Since then,onjugated pneumococcal vaccines have been marketed interna-ionally, reaching the Middle East and North Africa (MENA) region,or the prevention of IPD. PCV7 was introduced into the Lebanese

arket, largely driven by the private sector, in 2006, followed byCV10 and PCV13 in 2010. However, none of these vaccines is cur-ently part of Extended Program of Immunization (EPI) schedule inhe country. Published data about the incidence of IPD, the preva-ent pneumococcal serotypes, and vaccine coverage from the MENAegion, including Lebanon, are largely lacking.

In December 2004, the World Health Organization-Easternediterranean Regional Office (WHO-EMRO) held a meeting to

iscuss the status of pneumococcal disease and prevention inountries of the MENA region. As a recommendation from that

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G16

meeting, MENA countries were asked to set up surveillance sys-tems for IPD in order to determine the burden of disease caused by S.pneumoniae and the prevalent serotypes responsible for the major-ity of the cases; thus the Lebanese Inter-hospital PneumococcalSurveillance Program (LIPSP) was initiated [8].

The aim of this surveillance study was to obtain data aboutthe epidemiologic characteristics, serotypes, and antibiotic sus-ceptibilities of S. pneumoniae isolates causing invasive disease inLebanon. These data will be useful in helping health authoritiesmake decisions about the importance of vaccinations and changesin antibiotic treatment regimens in order to decrease the diseaseburden at the national level.

2. Methods

A prospective 6-year (October 2005–December 2011) program,the LIPSP was carried out in 78 hospitals geographically distributedall over Lebanon. Active surveillance was initiated by our centerthrough weekly or bi-weekly calls to all participating hospitalsto enroll qualifying cases. The study population included patientsof all age groups, with culture-confirmed IPD. IPD was defined asthe isolation of S. pneumoniae from a normally sterile site, such asblood, cerebrospinal fluid (CSF), and/or pleural fluid [2]. Epidemio-logical data collected for each sample included age, gender, sourceof the sample, clinical presentation, and outcome. The study wasapproved by the American University of Beirut (AUB) InstitutionalReview Board. Statistical analysis was done using the chi-squaretest of independence and the Pearson chi-square test. Statisticalsignificance was based on p-values less than 0.05.

Preliminary susceptibility testing and phenotypic identificationwere performed at the enrolled medical center laboratories. Bloodagar plates growing S. pneumoniae were shipped at room temper-ature from corresponding hospitals to the Center for InfectiousDiseases Research (CIDR) laboratory facility at AUBMC, where theywere sub-cultured onto blood agar plates for purification, S. pneu-moniae growth was confirmed based on colony morphology andoptochin sensitivity. Purified cultures were then stored in Bru-cella Broth at −80 ◦C for further use. All isolates were serotypedusing a multiplex-PCR serotype deduction assay [9]. Samples fromstored cultures that were non-viable or contaminated upon sub-culturing were subjected to DNA extraction from Brucella Brothand were serotyped. A total of 41 primer pairs were used andgrouped into seven multiplex PCR reactions. DNA extractions wereperformed using the illustra bacteria genomicPrep Mini Spin Kit(GE Healthcare, UK Limited Little Chalfont Buckinghamshire). PCRswere run on the PCR Sprint Thermal Cycler (Thermo Fisher Scien-tific, Waltham, MA, USA).

The stored live samples were sub-cultured onto blood agarplates, then onto chocolate agar slants, and periodically shipped,

in batches of 25 samples, to the US Naval Medical ResearchUnit No. 3 (NAMRU-3) in Cairo, Egypt, where all isolates weretested for antimicrobial susceptibility and serotype identification.Antimicrobial susceptibility testing was performed by the disc

Vaccine 30S (2012) G11– G17 G13

drtSc(((wScactirsisuI

3

3

cfpL(1(((mpAn(t(alw

3

cstPVpat

3

mrGo

Table 1Antimicrobial non-susceptibility by age group: percentage (count). CLSI 2009 break-points were used.

Age group PenicillinNon-susceptibility

CeftriaxoneNon-susceptibility

ErythromycinResistance

<2 years 29.1% (16) 18.3% (11) 40.7% (24)2–5 years 13.2% (5) 12.8% (5) 30.8% (12)6–20 years 3.4% (1) 3.4% (1) 24.1% (7)

R. Hanna-Wakim et al. /

iffusion and E-test (AB Biodisk, Solna, Sweden) methods asecommended by the Clinical and Laboratory Standards Insti-ute (CLSI, 2009) [10,11]. Quality control was carried out using. pneumoniae ATCC 49619 as the control strain. The antimi-robial agents tested included: penicillin (E-test), ceftriaxoneE-test), erythromycin (15 mcg), levofloxacin (5 mcg), vancomycin30 mcg), and trimethoprim–sulfamethoxazole (TMP/SMX)1.25/23.75 mcg). The inoculum was prepared in 0.85% saline,ith turbidity adjusted to the equivalent of a 0.5 McFarland

tandard. Meningitis isolates with a penicillin minimum inhibitoryoncentration (MIC) ≤ 0.06 mcg/ml were considered susceptiblend those with MIC ≥ 0.12 were resistant. Also, isolates witheftriaxone MIC ≤ 0.5 mcg/ml were considered susceptible andhose with MIC ≥ 2 were resistant. For cases other than meningitis,solates were defined as penicillin susceptible if MIC was ≤2 andesistant if MIC was ≥8. Also, isolates were defined as ceftriaxoneusceptible if MIC was ≤1 and resistant if MIC was ≥4. Isolatesn the intermediate and resistant groups were defined as non-usceptible. Pneumococcal isolates were additionally serotypedsing latex agglutination and the Quellung reaction (Statens Serum

nstitute, Copenhagen, Denmark) or PCR.

. Results

.1. Descriptive epidemiology

During the 6-year study period, 257 isolates of culture-onfirmed S. pneumoniae were identified, all of which were isolatedrom patients who met the definition criteria of IPD. Most sam-les were collected from Beirut (64%, n = 165), followed by Northebanon (24%, n = 61). Males constituted 56% of the patientsn = 145). Age distribution was as follows: 24.1% <2 years (n = 62),5.2% between 2 and 5 years (n = 39), 11.3% between 6 and 20 yearsn = 29), 16.3% between 21 and 60 years (n = 42), and 33.1% >60 yearsn = 85). Samples were collected from the blood (78.2%, n = 201), CSF13.6%, n = 35), pleural fluid (3.5%, n = 9), other sites (4.7%, n = 12):

iddle ear fluid (n = 3), mastoid abscess, eye discharge, urine (n = 2),eri-sternal abscess, peritoneal fluid, and synovial fluid (n = 1).

total of 46.5% of the patients were diagnosed with pneumo-ia (n = 119), 21.5% primary bacteremia (n = 55), 17.2% meningitisn = 44), 14.8% other diagnoses (n = 38): mastoiditis (n = 11), oti-is media (n = 6), bronchitis, urinary tract infection, eye infectionn = 3), sinusitis, osteomyelitis, gastroenteritis (n = 2), pharyngitis,bdominal abscess, peritonitis, pyelonephritis, septic arthritis, cel-ulits (n = 1). Of the 224 patients for whom outcome information

as available, 13.4% resulted in death (n = 30).

.2. Serotype distribution and vaccine coverage

Serotype distribution is summarized in Fig. 1. We assumedross-coverage among serotypes 6A, 6B, and 6C [12], and betweenerotypes 9V and 9A [13]. When all age groups were consideredogether, IPD caused by vaccine serotypes was: 41.4% (n = 106) forCV7, 53.9% (n = 138) for PCV10, and 67.2% (n = 162) for PCV13.accine coverage per each age group is represented in Fig. 2. Foratients <2, 2–5, and >60 years of age, PCV7 coverage was 50%, 51%,nd 35%, respectively; PCV10 coverage was 53%, 74%, 45%, respec-ively; and PCV13 coverage was 63%, 80%, and 68%, respectively.

.3. Antimicrobial susceptibilities

Using the latest CLSI breakpoints, which differentiate between

eningeal and non-meningeal isolates, we obtained the following

esults: 82.6% (n = 200) of the isolates were susceptible to penicillin-, 17.4% (n = 42) were non-susceptible. Mortality occurred in 23.5%f patients with penicillin non-susceptible isolates as compared

21–60 years 23.7% (9) 12.8% (5) 21.1% (8)>60 years 13.4% (11) 13.1% (11) 26.2% (22)

to 11.9% in the susceptible group (p-value 0.073). Of strains non-susceptible to penicillin, serotype 19F (n = 17) was most frequent,followed by 6 and 14 (n = 5 each). PCV7 serotypes accounted for71.4% of the PNS strains, while PCV13 serotypes constituted 78.6%.

A total of 86.9% of the isolates were susceptible to ceftriaxone(n = 218), 9.5% had intermediate resistance (n = 24), 3.6% were resis-tant (n = 9). Susceptibility to erythromycin was 70.7% (n = 176) andresistance was 29.3% (n = 73). Susceptibility to levofloxacin was99.5% in the 192 isolates tested. Susceptibility to TMP/SXT was33.9% (n = 76), while 12.1% (n = 27) showed intermediate resistance,and 54% (n = 21) were resistant. All of the 237 isolates tested withvancomycin were susceptible and 28 isolates (10.9%) were multi-drug-resistant (MDR), defined as resistant to 2 or more differentanti-microbial classes. Among the isolates that were MDR, the mostprevalent serotypes were: 19F (17 isolates) and 14 (5 isolates).

3.4. Statistics by age group

Mortality of IPD was significantly associated with age group(p < 0.01). It was highest among the eldest (>60 years, n = 18/72,25%) and the youngest (<2 years, 6/55, 10.9%) patients. The clinicalpresentation in each age group is summarized in Fig. 3. Suscep-tibility by age group is described in Table 1. The most commonserotypes in each age group were the following: <2 years: serotypes14 (n = 12), 19F (n = 9), and 6 (n = 7); 2–5 years: serotypes 5 (n = 6),14 (n = 5), 6 (n = 4), and 19F (n = 4); 6–20 years: serotypes 1 (n = 4),19F (n = 4), and 6 (n = 3); 21–60 years: serotypes 19F (n = 7), 1 (n = 4),6 (n = 4), and 19A (n = 4); and: >60 years: serotypes 3 (n = 13), 9V/A(n = 9), 19F (n = 7), and 19A (n = 6).

4. Discussion

This study describes the epidemiologic characteristics, antibi-otic susceptibility, and serotype prevalence of invasive pneumococ-cal infections in Lebanon, one of the countries in the MENA region,with an estimated population of 4,000,000 [14]. This is the firstreport in the literature describing serotype prevalence in IPD fromLebanon, as all previous studies focused exclusively on antibioticsusceptibility [4,15–21].

Although the study initially included a total of 78 hospitals, wereceived isolates from only 31 hospitals, collecting a total of 257samples from October 2005 to December 2011. Most of the sam-ples were from the capital Beirut and the second largest city inLebanon, Tripoli. This is probably the result of the difference inpractice between the capital and remote areas, where there is agreater tendency in remote areas to start antibiotic therapy empiri-cally without taking any cultures, thus explaining the lower numberof isolates from these areas. Moreover, the diagnosis of pneumo-coccal pneumonia in our study was based on positive blood cultureand imaging studies. The literature reports that only about 5–30%

of cases of pneumococcal pneumonia have a positive blood culture[22]. Therefore, reporting the incidence of IPD in Lebanon basedon our data would significantly underestimate the actual diseaseburden.

G14 R. Hanna-Wakim et al. / Vaccine 30S (2012) G11– G17

31

23

18

139 9 8

16

106

1815

7 6 5 5 4 4 4 3 3 3 2 2 2 2

1216

0

5

10

15

20

25

30

35

19F 6 14

9V/9A

23F 4

18C 1 5 7F 3

19A

22F

33F

11A/D 9N 10A

12F 8

15A

15B/C

16F

23A 29 35B 38

Other

Untypeable

PCV7 PCV1 0 PCV1 3 Non -Vaccine

Num

bero

fCases

Serotype

Fig. 1. Serotype distribution by vaccine coverage. Cross-coverage is assumed for serotypes belonging to serogroup 6, and for serotypes 9V/9A. Serogroup 6 includes thefollowing: 6B (n = 10), 6A (n = 1), 6A/B (n = 8), 6A/B/C (n = 3), 6C (n = 1). Serotypes 9V/9A include: 9V (n = 8), 9V/9A (n = 2), 9A (n = 3). “Other” infrequent serotypes include: 10F,13, 17F, 21, 22A, 24F, 28A, 31, 34,35F, 42, and Sg18 (n = 1 each).

50 51

41

33 35

53

74

62

4845

63

80

6264

68

0

10

20

30

40

50

60

70

80

90

<2y 2y-5y 6y-20 y 21 y-60 y >60 y

PCVCo

verage

(%)

Age Grou p

PCV7

PCV10

PCV13

Fig. 2. Vaccine serotypes (PCV7, PCV10, and PCV13) by age group. Vaccine coverage for patients 5 years old or younger was: 50.5% for PCV7, 61.4% for PCV10, and 69.3% forPCV13.

1619

12

20

52

16

9

4

13

6

16

6 74

22

14

96 5 4

0

10

20

30

40

50

60

<2y 2y-5y 6y-20 y 21 y-60 y >60 y

Num

bero

fCases

ge G

Pneumonia

Meningi�s

Primary Bacteremia

Other

senta

cbwos

A

Fig. 3. Clinical pre

There are no official data in Lebanon about the number ofhildren receiving the pneumococcal vaccine. With 72,000 annualirths [14], it is estimated that approximately 18.5% of the children

ho are 2 years old or younger have received an average of 3 doses

f the conjugated pneumococcal vaccine, based on drug companyales in the years 2010 and 2011 (Personal communication).

rou p

tion by age group.

In our study population, most of the S. pneumoniae sampleswere from patients older than 60 years or younger than 2 years.Robinson et al. reported that 53% of IPD cases occurred in these age

groups in the USA in 1998 [23]. A male predominance (56%) wasnoted in our study, a finding consistent with what was reportedin other studies in Lebanon, where 62% of the patients were males

Vaccin

[lcipmob[sbee((l

bbomwh3[oaa27sPt1tibiat1s

ooptt1[s5anb[MairdioClt

R. Hanna-Wakim et al. /

16,18] and in the USA (55% males) [23]. Our samples were col-ected mostly from blood, followed by CSF. Blood was also the mostommon source for invasive pneumococcal isolates in other stud-es from the MENA region and USA [24,25]. Most of our patientsresented with pneumonia, followed by primary bacteremia andeningitis. According to the CDC ABC Surveillance report, 70.4%

f IPD cases were pneumonia with bacteremia, followed by 19.2%acteremia without focus, and 4.8% meningitis in 2009 in the USA2]. The mortality in our patient population reached 13.4%, a valueimilar to the mortality caused by IPD in the Kingdom of Saudi Ara-ia (KSA) (12%) [26] and in the USA during the pre-vaccinationra (19%) as compared to that during the post-vaccinationra (5%) [27]. Outcome was significantly associated with agep < 0.01); the highest mortality was in the extreme age groups>60 years followed by <2 years), which is consistent with theiterature [28].

The most prevalent serotypes in our study were those coveredy the available pneumococcal vaccines. In all age-groups com-ined, the most common 6 serotypes/groups were in decreasingrder: 19F, 6, 3, 14, 1, and 19A. In children less than 2 years, theost common serotypes/groups were 14, 19F, and 6. Serotype 3as, surprisingly, the highest in the age group >60 years, whichad the highest mortality. Rodriguez et al. reported that serotype

was the most common serotype associated with patient death29]. Studies in the MENA region describe similar results in termsf serotype/group prevalence with 6, 14, 19, and 23F reporteds the most common [25,30–33]. Pneumococcal vaccine cover-ge seems to be inconsistent in MENA countries, ranging from6.7% to 55.2% for PCV7, 28.4% to 65% for PCV10, and 30.2% to3% for PCV13 [25,30,32,34]. Shibl et al. reported that circulatingerotypes/serogroups causing IPD and potentially covered by theCV7 vaccine ranged from 49% to 83% in children <2 years, and 61%o 69% in children ≤5 years in the Arabian Peninsula and Egypt from990 to 2007 [31]. Vaccine serotypes contributed to a high propor-ion of the serotypes prevalent in our study population, especiallyn children ≤5 years, as compared to values reported in the USAefore the introduction of the PCV7. Introduction of the vaccine

n the USA in 2000 caused a significant decrease in the percent-ge of IPD cases due to PCV7 serotypes through 2004 from 81–84%o 16% in children, and from 53–62% to 26% in adults. Serotype9A accounted for the majority of cases of IPD caused by non-PCV7erotypes [5].

Appropriate antimicrobial therapy is essential for the treatmentf IPD. Drug resistance is an obstacle to effective therapy. Using thelder CLSI breakpoints, our study showed that non-susceptibility toenicillin was present in 46% of S. pneumoniae isolates identified inhe current LIPSP surveillance study covering the period from 2005o 2011, as compared to 56% in the years 1996–1998 [4], 50–69% in997–2004 [16], 69.9% in 2000–2001 [19], 50–60% in 2000–200417], and 54% in 2005–2009 [18]. However, CLSI 2009 breakpointshow an increase in penicillin susceptibility of our isolates from4% to 82.6%. PNS in the MENA region varies from one study tonother. Reports from Saudi Arabia and Tunisia describe penicillinon-susceptibilities between 44% and 54% [24,25,34]. One studyy Memish et al. reports penicillin non-susceptibility reaching 73%32]. The Antibiotic Resistance Surveillance and Control in the

editerranean Region (ARMed) investigated the epidemiology ofntimicrobial resistance in IPD, and concluded that PNS was presentn 26% of the total isolates; the highest rates were observed in Alge-ia (44%) and Lebanon (40%) [35]. We could not find any studiesescribing PNS in IPD according to the new CLSI 2009 breakpoints

n the MENA region with which to compare our data. In 2008, 10%

f the isolates causing IPD in 32 countries in Europe were PNS (newLSI) according to the European Antimicrobial Resistance Surveil-

ance System (EARSS) [36]. In the USA, Hampton et al. reportedhat PNS (CLSI 2009) strains caused 6–14% of IPD cases from 1998

e 30S (2012) G11– G17 G15

to 2008 [7]. The CDC ABC Surveillance described 10.2% PNS ratesin 2009 [2]. The rates of PNS in Europe and USA are lower thanthose reported in our study (17.4% PNS) using the CLSI 2009 break-points. This could be due to higher PNS among vaccine serotypes [7],which contribute to the majority of IPD in our study, as opposed tothe higher proportion of non-vaccine serotypes in the developedworld. We have found no association between PNS and outcome(p = 0.073), as was described by Kaplan et al. [37]. Of the strains non-susceptible to penicillin, serotype 19F (n = 17) was most frequent,followed by 6 and 14 (n = 5 each). Mokaddas et al., in a study inKuwait, reported that the most common penicillin-resistant strainsbelonged to serotypes 23F, 19F, 6B, 14, and 9A [33]. PCV7 serotypesaccounted for 71.43% of our PNS strains. The serotypes included inPCV13, but not in PCV7, constituted 7.14% of our PNS isolates. In theUSA, serotypes in PCV7 caused 0–11.1% of PNS IPD, depending onthe age group, while the additional serotypes in PCV13 (excludingPCV7 serotypes) caused 78–97% of PNS IPD in the year 2007–2008[7]. The above finding could be explained by the implementation ofPCV7 in the routine vaccine schedule in the year 2000 in the USA,leading to the rise in non-PCV7 serotypes.

Ceftriaxone non-susceptibility in our study was encounteredin 13.1% of the isolates (3.6% fully resistant). Other studies inLebanon reported non-susceptibilities ranging from 0% [17,20] to13% [17]. Studies in Egypt and Saudi Arabia reported resistance toceftriaxone/cefotaxime ranging from 0% to 6% [24,30,38–40]. TheCDC ABC Surveillance reports cefotaxime non-susceptibility of 8.1%(1.9% resistance) [2], which remains lower than ceftriaxone non-susceptibility in our study. The high �-lactam non-susceptibilityrates in Lebanon are probably due to the widespread use of antibi-otics, which is one of the most important risk factors for emergenceof drug-resistant pneumococci [36]. Resistance to erythromycinin our study was 29.3%. Previous studies in Lebanon described anincrease in erythromycin resistance from 11.5% in 1997 to 43% in2004 [16]. ARMed reported that non-susceptibility to erythromycinwas highest in Malta (46%) and Tunisia (39%) [35]. Worldwide,erythromycin resistance rates were highest (80%) in the Far East,followed by South Africa (54%) and Southern Europe (37%), and low-est in Latin America (15%), Australia (18%), and Northern Europe(18%) [36]. Erythromycin resistance rates in the USA in 2009 were24.9%, a finding in line with our results [2]. In our study, sus-ceptibility to levofloxacin was 99.5%, in line with all the previousstudies conducted in Lebanon, which reported 100% levofloxacinsusceptibility [16,19]. This high levofloxacin susceptibility was alsoobserved in KSA (98.7%) [32] and in the USA (99.7%) [2]. A totalof 10.9% of our isolates were MDR, which is consistent with otherstudies from Algeria, Tunisia, Jordan and Turkey, all reporting MDRrates of more than 5% [35].

5. Conclusions

IPD continues to be a burden for both children and adults indeveloping countries, including Lebanon, where surveillance stud-ies are lacking. The most prevalent invasive serotypes in our studyare those found in the commercially available conjugated pneu-mococcal vaccines. This may be a consequence of the low numberof vaccinated children in Lebanon, which in turn emphasizes theimportance of implementing the conjugated pneumococcal vac-cines in the routine immunization schedule on the national level.The higher rates of antimicrobial resistance described in our study,in comparison to developed countries, are probably due to theunrestricted over-the-counter use of antibiotics in Lebanon, neces-

sitating more control over antibiotic intake. Continuing the currentsurveillance study would help assess the on-going changes in epi-demiology of IPD, serotype prevalence, and antibiotic resistance inLebanon.

G Vaccin

A

ap

t

Past

A

PRHHpHmSsMNIWAlGdNtDH

R

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

16 R. Hanna-Wakim et al. /

cknowledgements

Thanks are due to Dr. Bernard Beall, Centers for Disease Controlnd Prevention (CDC), Atlanta, Georgia, for provision of the Multi-lex PCR serotyping procedure and control S. pneumoniae strains.

Thanks for Ms. Caline Balaa for her aid in the statistical analysis.Start up funding for LIPSP was provided by PneumoADIP. Addi-

ional funding was provided by MSD and Pfizer.Conflict of interest: G. Dbaibo is on Advisory Boards for GSK and

fizer, has received honoraria for lectures from GSK, MSD, Pfizer,nd Hikma, and has received research funding through his univer-ity from GSK, Pfizer, and MSD. All of the other authors have noneo declare.

ppendix A.

Contributors of Lebanese Inter-Hospital Pneumococcal Surveillancerogram: Tamima Jisr, Makassed General Hospital; Rita Feghali,afik Hariri University Hospital; Gilbert Kara Yacoub, Haykalospital; Monzer Hamzeh, Nini Hospital; Ricardo Sarraf, Monlaospital; Jacques Mokhbat, University Medical Center-Rizk Hos-ital; Antoine Haddad, Sacré Coeur Hospital; Salam Samad, Centreospitalier Du Nord; Wassim Serhal, Sahel General Hospital; Ray-ond Rohban and Raymond Mikhael, Hopital Saint-Joseph des

oeurs de la Croix; Mohammad Zaatari, Hammoud Hospital Univer-ity Medical Center; Ziad Daoud, Saint Georges Hospital Universityedical Center; Naziha Makhlouf, Tal Chiha Hospital Zahle; Malakaboulsi, Islamic Hospital; Antoine Abi Nasr, Saint Louis Hospital;

brahim Ahmad, Al-Rasoul Al-Aazam Hospital; Rami Caracalla, Ain Zain Hospital; Tony Faddoul, Saint Charles Hospital; Hiam Matta,

l-Arz Hospital; Salam Monzer, Rayak Hospital; Mohamad Abdal-ah, Akkar Rahal Hospital; Mohamed Abdallah, Al-Rayan Hospital;eorges Abdel Nour, Centre Hospitalier Universitaire Notre Damees Secours; Edmond Abboud, Middle East Institute of Health;adim Azar, Najjar Hospital; Hadi El Amin, Al Yousef Medical Cen-

er; Mohammad Haidar, Bahman Hospital; Ghaith Makhoul, Notreame de la Paix Hospital; Farida Saadeh, Notre Dame du Libanospital; Antoine Zablit, Al Hayat Hospital.

eferences

[1] WHO. Weekly epidemiological record. Geneva: World Health Organi-zation; 2003. Available from: http://www.who.int/biologicals/vaccines/pneumococcal/en/ [accessed 03.05.12].

[2] Centers for Disease Control and Prevention. Active Bacterial Core Surveil-lance Report, Emerging Infections Program Network, Streptococcus pneumoniae,2009. Available from: http://www.cdc.gov/abcs/reports-findings/survreports/spneu09.html; 2010 [accessed 03.05.12].

[3] WHO. International travel and health: pneumococcal disease. WorldHealth Organization; 2012. Available from: http://www.who.int/ith/diseases/pneumococcal/en/ [accessed 03.05.12].

[4] Araj GF, Bey HA, Itani LY, Kanj SS. Drug-resistant Streptococcus pneumoniae inthe Lebanon: implications for presumptive therapy. Int J Antimicrob Agents1999;12:349–54.

[5] Albrich WC, Baughman W, Schmotzer B, Farley MM. Changing characteristicsof invasive pneumococcal disease in Metropolitan Atlanta, Georgia, after intro-duction of a 7-valent pneumococcal conjugate vaccine (1997–2004). Clin InfectDis 2007;44:1569–76.

[6] Centers for Disease Control and Prevention. Licensure of a 13-valent pneu-mococcal conjugate vaccine (PCV13) and recommendations for use amongchildren—Advisory Committee on Immunization Practices (ACIP), 2010.MMWR 2010;59:258–61.

[7] Hampton LM, Farley MM, Schaffner W, Thomas A, Reingold A, Harrison LH,et al. Prevention of antibiotic-nonsusceptible Streptococcus pneumoniae withconjugate vaccines. J Infect Dis 2012;205(February (3)):401–11 [Epub 2011December 7].

[8] World Health Organization, Regional Office for the Eastern Mediterranean. The

Work of WHO in the Eastern Mediterranean Region: Annual Report of theRegional Director 1 January–31 December 2005.

[9] Pai R, Gertz RE, Beall B. Sequential multiplex PCR approach for determin-ing capsular serotypes of Streptococcus pneumoniae isolates. J Clin Microbiol2006;44:124–31.

[

e 30S (2012) G11– G17

10] Jacobs MR, Bajaksouzian S, Appelbaum PC, Bolmström A. Evaluation of theE-test for susceptibility testing of pneumococci. Diagn Microbiol Infect Dis1992;15:473–8.

11] Clinical and Laboratory Standards Institute. Performance standards forantimicrobial susceptibility testing; 19th informational supplement. Wayne,PA: Clinical and Laboratory Standards Institute; 2009 [CLSI documentM100-S19].

12] Cooper D, Yu X, Sidhu M, Nahm MH, Fernsten P, Jansen KU. The 13-valent pneu-mococcal conjugate vaccine (PCV13) elicits cross-functional opsonophagocytickilling responses in humans to Streptococcus pneumoniae serotypes 6C and 7A.Vaccine 2011;29:7207–11.

13] Braconier JH, Myhre EB, Odeberg H. Cross-reacting opsonic antibodies to clini-cally important pneumococcal serotypes after pneumococcal vaccination. EurJ Clin Microbiol 1983;2(5):453–8.

14] Ministry of Health Statistical Bulletin; 2010.15] Uwaydah M, Jradeh M, Shihab Z. Antimicrobial resistance of clinical isolates

of Streptococcus pneumoniae in Lebanon. J Antimicrob Chemother 1996;38:283–6.

16] Karam Sarkis D, Hajj A, Adaime A. Evolution of the antibiotic resistance of Strep-tococcus pneumoniae from 1997 to 2004 at Hotel-Dieu de France, a universityhospital in Lebanon. Pathol Biol 2006;54:591–5 [in French].

17] Daoud Z, Cocozaki A, Hakime N. Antimicrobial susceptibility patterns ofHaemophilus influenzae and Streptococcus pneumoniae isolates in a Beirutgeneral university hospital between 2000 and 2004. Clin Microbiol Infect2006;12:86–90.

18] Daoud Z, Kurani M, Saab R, Abi Nader M, Hajjar M. Resistance of Streptococcuspneumoniae isolated from Lebanese patients between 2005 and 2009. Rev EspQuimioter 2011;24(2):84–90.

19] Uwaydah M, Mokhbat JE, Karam-Sarkis D, Baroud-Nassif R, Rohban T. Penicillin-resistant Streptococcus pneumoniae in Lebanon: the first nationwide study. IntJ Antimicrob Agents 2006;27:242–6.

20] Shaar TJ, Al-Hajjar R. Antimicrobial susceptibility patterns of bacteria at theMakassed General Hospital in Lebanon. Int J Antimicrob Agents 2000;14:161–4.

21] Naba MR, Araj GF, Baban TA, Tabbarah ZA, Awar GN, Kanj SS. Emergence offluoroquinolone-resistant Streptococcus pneumoniae in Lebanon: a report ofthree cases. J Infect Public Health 2010;3:113–7.

22] Hughes JR, Sinha DP, Cooper MR, Shah KV, Bose SK. Lung taps in childhood:bacteria, viruses and mycoplasmas in acute lower respiratory tract infections.Pediatrics 1969;44:477–85.

23] Robinson KA, Baughman W, Rothrock G, Barrett NL, Pass M, Lexau C, et al.Epidemiology of invasive Streptococcus pneumoniae infections in the UnitedStates, 1995–1998: opportunities for prevention in the conjugate vaccine era.JAMA 2001;285:1729–35.

24] Shibl AM. Distribution of serotypes and antibiotic resistance of invasive pneu-mococcal disease isolates among children aged 5 years and under in SaudiArabia (2000–2004). Clin Microbiol Infect 2008;14:876–9.

25] Al-Mazrou Y, Twum-Danso K, Al Zamil F, Kambal A. Streptococcus pneumo-niae serotypes/serogroups causing invasive disease in Riyadh, Saudi Arabia:extent of coverage by pneumococcal vaccines. Ann Saudi Med 2005;25:94–9.

26] Memish Z, Al-Otaibi B, Al Shaalan M, Alola S, Althaqafi AH. A retrospectiveepidemiological study of invasive pneumococcal disease in children aged 0–5years in the Kingdom of Saudi Arabia. In: Proceedings of the 6th interna-tional symposium on pneumococci and pneumococcal diseases (ISPPD). 2008[Abstract no. 130].

27] Tsigrelis C, Tleyjeh IM, Lahr BD, Nyre LM, Virk A, Baddour LM. Decreases incase-fatality and mortality rates for invasive pneumococcal disease in OlmstedCounty, Minnesota, during 1995–2007: a population-based study. Clin InfectDis 2008;47:1367–71.

28] Lipsky BA, Boyko EJ, Inui TS, Koepsell TD. Risk factors for acquiring pneumo-coccal infections. Arch Intern Med 1986;146:2179–85.

29] Rodríguez MA, González AV, Gavín MA, Martínez FM, Marín NG, BlázquezBR, et al. Invasive pneumococcal disease: association between serotype,clinical presentation and lethality. Vaccine 2011;29(34):5740–6 [Epub 2011June 15].

30] Wasfy MO, Pimentel G, Abdel-Maksoud M, Russell KL, Barrozo CP, Klena JD,Earhart K, et al. Antimicrobial susceptibility and serotype distribution of Strep-tococcus pneumoniae causing meningitis in Egypt, 1998–2003. J AntimicrobChemother 2005;55:958–64.

31] Shibl A, Memish Z, Pelton S. Epidemiology of invasive pneumococcal diseasein the Arabian Peninsula and Egypt. Int J Antimicrob Agents 2009;33(410):e1–9.

32] Memish ZA, Balkhy HH, Shibl AM, Barrozo CP, Gray GC. Streptococcus pneu-moniae in Saudi Arabia: antibiotic resistance and serotypes of recent clinicalisolates. Int J Antimicrob Agents 2004;23(January (1)):32–8.

33] Mokaddas EM, Rotimi VO, Albert MJ. Implications of Streptococcus pneumoniaepenicillin resistance and serotype distribution in Kuwait for disease treatmentand prevention. Clin Vaccine Immunol 2008;15(February (2)):203–7 [Epub2007 December 12].

34] Smaoui H, Amri J, Hajji N, Kechrid A. Antimicrobial susceptibility and serotype

distribution of Streptococcus pneumoniae isolates in children in Tunis. ArchPediatr 2009;16(March (3)):220–6 [in French, Epub 2009 February 11].

35] Borg MA, Tiemersma E, Scicluna E, van de Sande-Bruinsma N, de KrakerM, Monen J, et al. Prevalence of penicillin and erythromycin resistanceamong invasive Streptococcus pneumoniae isolates reported by laboratories

Vaccin

[

[

[

R. Hanna-Wakim et al. /

in the southern and eastern Mediterranean region. Clin Microbiol Infect2009;15(March (3)):232–7.

36] Linares J, Ardanuy C, Pallares R, Fenoll A. Changes in antimicrobial resistance,

serotypes and genotypes in Streptococcus pneumoniae over a 30-year period.Clin Microbiol Infect 2010;16(May (5)):402–10 [Epub 2010 February 2].

37] Kaplan SL, Mason Jr EO, Barson WJ, Wald ER, Arditi M, Tan TQ, et al. Three-year multicenter surveillance of systemic pneumococcal infections in children.Pediatrics 1998;102:538–45.

[

[

e 30S (2012) G11– G17 G17

38] Al-Tawfiq JA. Antibiotic resistance of pediatric isolates of Streptococcus pneu-moniae in a Saudi Arabian hospital from 1999 to 2004. Med Sci Monit2006;12(November (11)):CR471–5.

39] Kambal AM, Abdullah AM. Childhood pneumococcal bacteraemia in Riyadh,Saudi Arabia. Ann Trop Paediatr 1997;17(September (3)):245–51.

40] Kattan R, Abu Rayyan A, Zheiman I, Idkeidek S, Baraghithi S, Rishmawi N,et al. Serotype distribution and drug resistance in Streptococcus pneumoniae,Palestinian Territories. Emerg Infect Dis 2011;17(January (1)):94–6.