International Journal of Infectious Diseases 80 (2019) 28–33

Contents lists available at ScienceDirect

International Journal of Infectious Diseases

journal homepage: www.elsevier.com/locate/ijid

Effectiveness of 10-valent pneumococcal conjugate vaccine

against vaccine-type invasive pneumococcal disease in Pakistan
Atif Riaza , Syed Mohiuddina , Sara Husaina , Mohammad Tahir Yousafzaia ,
Muhammad Sajida , Furqan Kabira , Najeeb ur Rehmana , Waseem Mirzab , Basit Salamb ,
Naila Nadeemb , Khatidja Pardhana,c , Khalid Mehmood A. Khanc , Syed Jamal Razac ,
Fehmina Arifd , Khalid Iqbale , Hassan Khalid Zuberif , Cynthia G. Whitneyg , Saad B. Omerh ,
Anita K.M. Zaidia , Asad Alia,* , Pakistan Pneumococcal Vaccine Study Group
a
Department of Paediatrics and Child Health, The Aga Khan University Hospital, Karachi, Pakistan
b
Department of Radiology, Aga Khan University Hospital, Karachi, Pakistan
c
National Institute of Child Health, Karachi, Pakistan
d
Civil Hospital, Karachi, Pakistan
e
Kharadar General Hospital, Karachi, Pakistan
f
Sindh Government Children Hospital, Nazimabad, Karachi, Pakistan
g
Centers for Disease Control and Prevention, Atlanta, GA, USA
h
Departments of Global Health, Epidemiology, Pediatrics, and Emory Vaccine Center, Emory University, Atlanta, GA, USA

A R T I C L E I N F O A B S T R A C T

Article history: Objective: To assess the effectiveness of 10-valent pneumococcal conjugate vaccine (PCV10) against
Received 13 August 2018 invasive pneumococcal disease (IPD) due to vaccine serotypes of Streptococcus pneumoniae post
Received in revised form 6 December 2018 introduction of the vaccine into the routine immunization program in Pakistan.
Accepted 11 December 2018
Methods: A matched case–control study was conducted at 16 hospitals in Sindh Province, Pakistan.
Corresponding Editor: Eskild Petersen, Aar-
hus, Denmark
Children aged <5 years (eligible to receive PCV10) who presented with radiographically confirmed
pneumonia and/or meningitis were enrolled as cases. PCR for the lytA gene was conducted on blood (for
radiographic pneumonia) and cerebrospinal fluid (for meningitis) samples to detect S. pneumoniae. The
Keywords:
Pneumococcal conjugate vaccine
proportion of IPD due to vaccine serotypes (including vaccine-related serogroups) was determined
Effectiveness through serial multiplex PCR. For each case, at least five controls were enrolled from children hospitalized
Invasive pneumococcal disease at the same institution, matched for age, district, and season.
Results: Of 92 IPD patients enrolled during July 2013 to March 2017, 24 (26.0%) had disease caused by
vaccine serotypes. Most case (87.5% of 24) and control (66.4% of 134) children had not received any PCV10
doses. The estimated effectiveness of PCV10 against vaccine-type IPD was 72.7% (95% confidence interval
(CI) 7.2% to 92.6%) with at least one dose, 78.8% (95% CI 11.9% to 96.0%) for at least two doses, and 81.9%
(95% CI 55.7% to 97.9%) for all three doses of vaccine.
Conclusions: The vaccine effectiveness point estimates for PCV10 were high and increased with increasing
number of doses. However, vaccine effectiveness estimates did not reach statistical significance, possibly
due to low power. The findings indicate the likely impact of vaccine in reducing the burden of vaccine-
type IPD if vaccine uptake can be improved.
© 2018 The Authors. Published by Elsevier Ltd on behalf of International Society for Infectious Diseases.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).

Introduction proportion of pneumonia and meningitis in low and middle-

income settings is attributed to Streptococcus pneumoniae, often
Pneumococcal disease is a major cause of childhood morbidity presenting as invasive pneumococcal disease (IPD) (O’Brien et al.,
and mortality worldwide (O’Brien et al., 2009). A significant 2009; Zar et al., 2013). The pneumococcal conjugate vaccine (PCV)
has contributed significantly in reducing vaccine-related child-
hood morbidity and mortality in the developed world (Hsu et al.,
* Corresponding author at: Department of Paediatrics and Child Health, Aga Khan
2009; Loo et al., 2014). Several middle- to low-income countries
University, Stadium Road, Karachi, Pakistan. have introduced PCV into their routine immunization schedules
E-mail address: asad.ali@aku.edu (A. Ali). with the assistance of Gavi—The Vaccine Alliance (IVAC, 2014).

https://doi.org/10.1016/j.ijid.2018.12.007
1201-9712/© 2018 The Authors. Published by Elsevier Ltd on behalf of International Society for Infectious Diseases. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 A. Riaz et al. / International Journal of Infectious Diseases 80 (2019) 28–33 29

There are, however, few reports on post-vaccine presented with signs of fast respiratory rate (>60 breaths per
introduction assessment available from these countries, espe- minute for children younger than 2 months, >50 per minute for
cially those in Asia. Pakistan was the first country in South Asia those aged 2–11 months, and >40 per minute for those aged 12–
to introduce the 10-valent PCV (PCV10) into its national 59 months), fever (>38.3  C), and at least one of three danger
expanded program on immunization (EPI). A scientific assess- signs, including lower chest wall indrawing, central cyanosis,
ment of vaccine effectiveness could help policymakers under- and inability to drink (Scott et al., 2012). Similarly, children
stand the performance of a routine vaccination program in were considered to have suspected meningitis if they presented
Pakistan and provide information to guide decisions on with fever (>38.3  C) along with one of the following signs: neck
introducing or sustaining PCV programs in other countries in stiffness, bulging fontanelle (in children aged <12 months),
the region. The primary objective of this study was to assess the altered consciousness and irritability, or convulsions. Children
effectiveness of PCV10 on IPD due to vaccine serotypes of S. eligible for recruitment into the study underwent chest
pneumoniae. This appears to be the first such evaluation of any radiography, blood specimen collection, and lumbar puncture,
pneumococcal vaccine in the region. as determined by the treating physician.
All chest radiographs were screened for substantial alveolar
Methods consolidation by a physician at each sentinel site (reader 1) and
scanned using a VIDAR high-resolution digitizer for further
Setting interpretation. The scanned radiographs were sent to two
independent radiologists trained in World Health Organization
Pakistan is a lower middle-income country with a population (WHO) protocols (readers 2 and 3) (World Health Organization,
of more than 200 million (Pakistan Bureau of Statistics, 2017) and 2001; O’Grady et al., 2010) for X-ray interpretation in a blinded
a gross domestic product (GDP) per capita of US$ 1468.20 (World fashion. Radiographs with discordant end-points between reader 2
Bank, 2019). Globally, Pakistan has the third highest burden of and 3 were interpreted by an additional senior radiologist, with
child mortality, and progress towards reducing child mortality two out of three agreements considered the final diagnosis for
has been slow (Bhutta et al., 2013). Infectious diseases, radiographically proven pneumonia. For quality assurance, every
particularly vaccine-preventable illnesses, contribute a signifi- fourth negative radiograph identified from each sentinel site was
cant burden to childhood mortality, with Pakistan identified as sent to readers 2 and 3 for verification.
one of the six countries globally with a high burden of S. Lumbar puncture specimens of cerebrospinal fluid (CSF)
pneumoniae-related mortality (WHO, 2000). Sentinel site surveil- obtained from children with suspected meningitis were analyzed
lance has been established at several public and private secondary for cell count; visibly cloudy specimens or those with a white blood
and tertiary care hospitals serving low and middle-income cell count >10  106/l were considered diagnostic of bacterial
populations in eight districts of the province of Sindh. Sindh is meningitis.
the second most populous province in Pakistan, with a population All children with radiographically proven pneumonia or
of 48 million. bacterial meningitis were considered eligible for study enroll-
PCV10 was introduced into the Sindh EPI on April 1, 2013, with a ment. Blood samples from children with radiographically proven
schedule of three doses given at 6, 10, and 14 weeks of age and no pneumonia and CSF samples from bacterial meningitis case
opportunity for catch-up vaccination (Ali et al., 2016). All vaccines patients were sent to the Infectious Disease Research Laboratory
are provided free of cost to children of eligible age through at Aga Khan University (AKU-IDRL) in Karachi at a controlled
designated government EPI centers and outreach facilities. temperature (2–8  C) for the detection of S. pneumoniae by lytA-
Twenty-nine percent of children in Sindh were reported to be targeted real-time PCR (Maria da Gloria et al., 2007). A case met
fully vaccinated before the initiation of this study (Pakistan the criteria for IPD if S. pneumoniae was detected in either blood or
Demographic and Health Survey, 2013). CSF specimens on PCR. Serotype determination was done using
Sixteen healthcare facilities (nine public and seven private) serial multiplex PCR. Details of laboratory methods for blood PCR
were included in the study. These sites were established based on are provided elsewhere (Kabir et al., 2017). IPD cases were
the availability of pediatric inpatient care, including the diagnosis, classified as vaccine-type (VT-IPD) if the serotype matched those
treatment, and management of childhood pneumonia and present in the PCV10 composition (1, 4, 5, 6B, 7F, 9V, 14, 18C, 19F,
meningitis, the availability of chest radiography facilities, and and 23F) or if they were from the same serogroups (e.g. 6A/B/C/D,
the willingness of the hospital administration to participate in the 7A/F, 18A/B/C/F).
study. A list of participating hospitals by district is provided in the Children were excluded if they had been enrolled previously,
Supplementary material (Appendix S1). were not eligible to receive vaccines, resided outside the study
catchment area, or if parental consent was not obtained.
Study design and enrollment Five controls were enrolled from the same hospital for each case
of IPD, irrespective of serotype and vaccination status. Children
A matched case–control study design was used. Case enroll- were excluded from enrollment as controls if they had signs of
ment began in July 2013 and was concluded in March 2017. suspected pneumonia, meningitis, or fever of unknown origin, if
Children admitted to inpatient units at the sentinel hospitals were they lived outside the catchment area, or if their parents did not
screened for eligibility prior to enrollment in the study. A research provide consent. Controls were matched to case patients by age,
team (comprising a medical officer and/or nurse) at each sentinel district, and season. Controls were age-matched within 8 weeks of
site maintained inpatient lists and assessed each individual child’s case patient age for case patients aged <1 year and within 12 weeks
eligibility for enrollment. for case patients aged 1 year. Controls were randomly identified
Children born on or after February 15, 2013 (who would from the inpatient wards and emergency rooms and enrollment
have been eligible to receive at least one dose of PCV10) and was completed within 2 months of enrollment of the respective
who were less than 5 years of age at the time of the study, who IPD case. In some instances, where an enrolled control was
had a provisional diagnosis of pneumonia and/or meningitis, vaccinated but did not have a clear vaccination history (i.e., lacked
and who were residents of study catchment districts were a verifiable vaccination record), an additional control was enrolled.
screened for enrollment as a case patient. Children were All additional controls were included in the calculation of vaccine
considered to have suspected severe pneumonia if they effectiveness (VE).
30 A. Riaz et al. / International Journal of Infectious Diseases 80 (2019) 28–33

Figure 1. Study profile.

Data collection cooking smoke, use of natural gas for cooking, at least one other
child <5 years of age in the house, and crowding (>2 people sleeping
Parents/caretakers of all enrolled children were interviewed in in the same room as the child). Propensity scores for significant
person by trained research staff using a standardized question- confounders and risk factors including sex, age, and paternal
naire. Data were collected on demographic characteristics and education, use of natural gas for cooking, crowding, cigarette
household information, including the number of people residing in smoking, and exposure to smoke (p-values 0.25) were used for the
the household, index child’s exposure to smoke, treatment history final adjusted multivariable model.
and outcome of disease for the current illness, and vaccination
history. Where available, vaccination cards were used to verify Results
vaccination history. These cards were reviewed by two indepen-
dent research staff in a blinded manner. In the case of discrepancy A total 10 124 children with suspected pneumonia and 2567
between the two readers, vaccination status was verified with the children with suspected meningitis were screened at the sentinel sites
EPI center where the child was vaccinated, as identified by the using the eligibility criteria. Of 6554 eligible children, 2530 (38.6%)
parents. When parents indicated that the child had not received were identified as having suspected pneumonia by study physicians at
any vaccination, the oral history was used. the sentinel sites and underwent chest radiography. Of these, 1426
(56.3%) were diagnosed with radiographically confirmed pneumonia
Sample size by radiologists at subsequent levels of X-ray interpretation.
Among 1726 eligible children with suspected meningitis, 213
The sample size was calculated using NCSS PASS v. 11 software (12.3%) were diagnosed with bacterial meningitis and their CSF
(NCSS, 2019) for a matched case–control study design. Initially the samples were submitted for PCR testing.
sample size was calculated to assess VE against overall IPD. However, A total of 92 case patients with IPD were enrolled, including 63
it was recalculated in 2014 to detect a VE of 73% for VT-IPD with at cases of pneumococcal meningitis and 29 cases of pneumococcal
least three doses of PCV10 (Domingues et al., 2014). Using an alpha pneumonia (Figure 1). Among these, a serotype could be
error of 0.05, 80% power, 50% vaccine coverage for PCV10, and a determined for 60 (65.2%). Twenty-four (26%) were identified as
correlation coefficient for vaccination between matched cases and vaccine types (VT-IPD), as defined above. Thirty-six cases (39.1%)
controls of 0.20, it was estimated that 28 VT-IPD cases with five were non-vaccine types (NVT) and 32 (34.7%) could not be typed
matched controls for each case would be needed. with the available panel of multiplex PCR reactions employed.
Table 1 provides the general characteristics of the 24 VT-IPD case
Data quality and statistical analysis patients. The mean age of the VT-IPD case patients was 6.5 months;

All case registration forms were double-entered using Visual Table 1

Foxpro v. 6.0 (Microsoft Corporation, USA). Enrollment logs (for cases Characteristics of enrolled children with vaccine-serotype IPD.
and controls) were reviewed on a weekly basis by the primary
Characteristics Case patients
investigator. Data were analyzed using Stata software version 12.0 (n = 24), n (%)
(Hamilton, 2013). The Chi-square test and t-test were applied for the
Age in months, mean  SD 6.5  4.1
comparison of characteristics between case patients and controls. Clinical syndrome
Standard analysis was conducted using logistic regression to Pneumococcal meningitis 17 (70.8)
calculate the odds ratio (OR) of PCV10 vaccination versus no Pneumococcal pneumonia 7 (29.2)
vaccination in VT-IPD cases compared with controls (Pearce, 2016). Medical care
Admitted to hospital 24 (100)
VE was calculated with the formula, VE = (1 OR)  100%. For VE
Duration of hospitalization in days, mean  SD 9.3  6.7
analysis, the vaccine dose was considered valid if received before Death (outcomes)
hospitalization. Confounders and risk factors were assessed by Overall 5 (20.8)
including additional variables such as sex, age, wasting, stunting, In cases of pneumonia 2 (40)
underweight, paternal education, large family size (5 family In cases of meningitis 3 (60)

members), exposure to cigarette smoke in the house, exposure to IPD, invasive pneumococcal disease; SD, standard deviation.
 A. Riaz et al. / International Journal of Infectious Diseases 80 (2019) 28–33 31

Figure 2. Vaccination status by serotype for vaccine serotype cases (n = 24).

Table 2
Comparison of characteristics of case patients and controls.a

Case patients Controls

(n = 24), n (%) (n = 134), n (%)
Sex, male 7 (29.2) 81 (60.5)
Age (months), mean  SD 6.5  4.1 7.1  4
Wasting (weight-to-height Z-score <2) 9 (39.1) 31 (26.7)
Stunting (height-to-age Z-score <2) 12 (52.2) 64 (58.2)
Underweight (weight-to-age Z-score <2) 14 (66.7) 72 (60)
Illiterate (paternal education) 18 (75) 55 (41)
Large family size (5 family members) 14 (58.3) 91 (67.9)
Person smoking cigarettes in the house 7 (29.2) 38 (28.4)
Exposure to cooking smoke (2 meters or closer to cooking smoke) 2 (8.3) 4 (3.0)
Use of natural gas for cooking 15 (62.5) 116 (86.6)
At least 1 other child <5 years old in the house 15 (62.5) 98 (73.1)
Crowding (>2 people sleeping in the same room as the child) 22 (91.7) 130 (97)
At least one dose of pentavalent vaccineb 7 (29.2) 59 (44)
Number of doses of PCV10
Zero doses 21 (87.5) 89 (66.4)
1 dose 1 (4.2) 12 (9.0)
2 doses 1 (4.2) 9 (6.7)
3 doses 1 (4.2) 24 (17.9)
1 dose 3 (12.5) 45 (33.6)
2 doses 2 (8.7) 33 (27.1)

SD, standard deviation; PCV10, 10-valent pneumococcal conjugate vaccine.

a
p-Values for the comparison of proportions between the study groups were calculated using the Chi-square test; p-values to compare means across the study groups were
calculated using the unpaired t-test.
b
Comprising five vaccine antigens: diphtheria, pertussis, tetanus, hepatitis B, and Haemophilus influenzae type B.

six of 24 were <14 weeks of age, too young to have been eligible for controls were found in sex, parental education, exposure to
three PCV10 doses. Seventeen (71%) had meningitis and seven cigarette smoke in the home, use of natural gas for cooking, and
(29%) had pneumonia. The average duration of hospitalization for vaccination status. While 44% of controls compared to 29% of case
these case patients was 9.3 days. Of the 24 VT-IPD case patients, patients had received at least a single dose of pentavalent vaccine
five died during their hospital stay. (comprising five vaccine antigens: diphtheria, pertussis, tetanus,
Among the 24 cases of VT-IPD, seven (29%) were serotype 19F, hepatitis B, and Haemophilus influenzae type B), this difference was
with 14 and 23F as other predominant serotypes. Furthermore, 21 not statistically significant. The majority of both case patients
out of 24 cases occurred in children who had not received even a (87.5%) and controls (66.4%) had not received any doses of PCV10.
single dose of PCV10. Only one case each of serotypes 23F and 5 Table 3 provides unadjusted and adjusted VE against VT-IPD
were identified: one in a child who had received two doses of with 95% confidence intervals (CI). Estimated VE against VT-IPD
PCV10 and one in a child who had received a single dose; only one was 72.7% (95% CI 7.2% to 92.6%) for at least one dose of PCV10
child with serotype 1 disease had received all three doses of and 78.8% (95% CI 11.9% to 96.0%) for at least two doses. Only 4.2%
PCV10 (Figure 2). of case patients had received all three doses of PCV10, compared to
VE for PCV10 was calculated for the 24 VT-IPD cases along with 17.9% of controls, resulting in VE of 81.9% (95% CI 55.7% to 97.9%).
134 age-matched controls. Table 2 presents a comparison of the Point estimates of VE were lower for those receiving exactly two
demographic characteristics of enrolled case patients and controls. doses of PCV10 (71.4%) and one dose of PCV10 (51.5%), compared to
There was no difference in mean age, nutritional status, family size, no vaccination. Effectiveness against non-PCV10 serotypes was not
or crowded living conditions between case patients and controls. significantly different from zero for different vaccination statuses
Statistically significant differences between case patients and (Table 3).
32 A. Riaz et al. / International Journal of Infectious Diseases 80 (2019) 28–33

Table 3
Unadjusted and adjusted PCV10 effectiveness against vaccine-type and non-vaccine type invasive pneumococcal disease outcomes in case patients and controls.

Vaccination status Cases Controls Unadjusted VE (%) Adjusted VE (%)a

(n = 24), n (%) (n = 134), n (%) (95% CI) (95% CI)
VE for PCV10 serotypes
Zero PCV doses 21 (87.5) 89 (66.4) Reference Reference
1 dose 1 (4.2) 12 (9) 64.7 ( 186.9 to 95.7) 51.5 ( 343.7 to 94.7)
2 doses 1 (4.2) 9 (6.7) 52.9 ( 292.3 to 94.3) 71.4 ( 227.6 to 97.5)
3 doses 1 (4.2) 24 (17.9) 82.3 ( 38.0 to 97.7) 81.9 ( 55.7 to 97.9)
1 dose 3 (12.5) 45 (33.6) 71.7 ( 0.2 to 92.0) 72.7 ( 7.2 to 93.1)
2 doses 2 (8.7) 33 (27.1) 74.3 ( 15.6 to 94.3) 78.8 ( 11.9 to 96.0)

Vaccination status Cases Controls Unadjusted VE (%) Adjusted VE (%)a

(n = 67), n (%) (n = 345), n (%) (95% CI) (95% CI)
VE for non-PCV10 serotypes
Zero PCV doses 38 (56.7) 203 (58.8) Reference Reference
1 dose 29 (43.3) 142 (41.2) 9.1 ( 85.1 to 35.7) 44.9 ( 154.6 to 17.6)
2 doses 23 (37.7) 115 (36.2) 6.8 ( 88.2 to 39.3) 54.1 ( 183.7 to 16.3)

PCV10, 10-valent pneumococcal conjugate vaccine; VE, vaccine effectiveness; CI, confidence interval.
a
Adjusted for sex, age, paternal education, use of natural gas for cooking, crowding and exposure to smoke based on propensity scores, VE = (1 OR)  100%.

Discussion to drop, both because of direct protection among children who

have received PCV10 doses and the decreased transmission of
The study findings suggest that PCV10 is protective against IPD vaccine serotype strains from vaccinated children to unvaccinated
(mainly pneumococcal meningitis) in the study population, for children (indirect or herd effects). At these coverage levels,
disease caused by vaccine types among children in South Asia. The however, indirect protection will probably not be as good as
study also showed that VE was associated with the number of expected with higher coverage (Loughlin et al., 2014). The most
doses received, as the point estimate for VE increased with each common PCV10 serotype found in this study was 19F, which
subsequent dose, from 51.5% among those receiving a single dose accounted for a higher proportion of the isolates than was found in
to 81.9% among those receiving all three doses, a trend that has also a previous study in the same setting (Shakoor et al., 2014).
been identified in other studies (Domingues et al., 2014; Palmu Whether this represents a relative shift because other vaccines
et al., 2013; Conklin et al., 2014). The study provides a useful insight serotypes are less common after vaccine introduction is unknown.
into VE and the distribution of vaccine serotypes in South Asia. The In addition to the lack of coverage with PCV10, other factors
estimated results of VE are encouraging and data from this study found to be associated with IPD included parental illiteracy and
could be helpful in studies performing a pooled analysis of PCV natural gas used for cooking. The odds of having an illiterate parent
effectiveness. were significantly higher among case patients than among
The confidence intervals for the VE estimates crossed the null controls, suggesting that parental education may play a role in
for all vaccination statuses examined, i.e., 1 dose, 2 doses, and knowledge about vaccines and other health services or in the
three doses PCV10. The wide confidence intervals can be attributed family’s economic status and overall well-being. The use of natural
to the small sample size and poor coverage of PCV10 within the gas for cooking was more prevalent for control patients than for
catchment population, where only 4.2% of case patients and 17.9% case patients; while the findings suggest that there may be a direct
of controls had received all three doses of PCV10. At the outset of linkage between exposure to indoor smoke and the subsequent
vaccine rollout, the EPI had planned to achieve 72% coverage of development of IPD, data to control for confounding factors are
PCV10 within 6 years of vaccine introduction (EPI, 2014). An insufficient.
independent survey conducted during 2016 in three rural districts In conclusion, the VE point estimates for PCV10 were high and
of Sindh indicated the highest coverage for all three doses of PCV10 increased progressively with increasing doses of PCV10. The VE
to be 38.5% (Anon, 2017). The sample size calculation for this study estimates did not reach statistical significance in this study, but
employed an assumption of at least 50% coverage of PCV10 within this may be due to the low power, as explained above. The study
3 years of vaccine rollout. However, the required coverage levels results suggest that the introduction of PCV10 with adequate
were not achieved within the study duration. This may have coverage is likely to decrease the IPD burden among children less
affected the power of the study. In order to achieve the optimal than 5 years of age in Pakistan.
impact of PCV10 at the population level, vaccine coverage needs to
increase significantly through investments in vaccine promotion, Ethical approval
addressing vaccine hesitancy among the local population and
enhancing coverage. The study received approval from the Ethics Review Committee
An interesting feature of this study was the relatively small (ERC) of Aga Khan University, Karachi, Pakistan. Written informed
proportion of VT-IPD out of the total IPD detected. Prior to the consent was obtained from the parents of all study participants.
introduction of the vaccine into the routine immunization program
in Pakistan, a hospital-based surveillance study from Sindh Funding source
indicated that 66.7% of all IPD cases among children aged less
than 5 years were PCV10 type serotypes (Shakoor et al., 2014). The The study was funded by Gavi—The Vaccine Alliance (grant
proportion of VT-IPD cases in this study was only 26%. This is likely number PP33961213A1).
an indication of the altering bacteriological milieu with the
introduction of PCV10. With the level of coverage measured among Conflict of interest
the controls in this study and in the independent surveys, the
proportion of IPD caused by vaccine serotypes would be expected All authors declare no conflict of interest.
 A. Riaz et al. / International Journal of Infectious Diseases 80 (2019) 28–33 33

Disclaimer invasive pneumococcal disease in Brazil: a matched case-control study. Lancet

Respir Med 2014;2(6):464–71.
EPI. Comprehensive multi-year plan: Ministry of National Health Services,
The findings and conclusions in this report are those of the Regulation and Coordination;2014-2018. 2014 Islamabad.
authors and do not necessarily represent the official position of the Hamilton LC. Statistics with Stata: Version 12. Eighth Edition 2013. https://www.
funding agency or the Centers for Disease Control and Prevention. stata.com/bookstore/statistics-with-stata/.
Hsu HE, Shutt KA, Moore MR, et al. Effect of pneumococcal conjugate vaccine on
pneumococcal meningitis. N Engl J Med 2009;360(3):244–56.
Contributions International Vaccine Access Center (IVAC). Vaccine Information Management
System (VIMS) report: global vaccine introduction. 2014 https://www.jhsph.
edu/research/centers-and-institutes/ivac/view-hub/IVAC-VIMS-Report-2014-
AA supervised all aspects of the study including manuscript Mar.pdf. [Accessed 8 December 2016].
preparation. AKMZ designed the study and contributed to Kabir F, Muneer S, Kalam A, Sami A, Qureshi S, Hotwani A, et al. Experience with the
manuscript revision. AR, SM, and SH participated in study quantitative lytA gene real-time polymerase chain reaction for the detection of
Streptococcus pneumoniae from pediatric whole blood in Pakistan. Biomed
oversight, data review, and manuscript writing. MTY, NR, and
Biotechnol Res J 2017;1(1):71.
SM contributed to the data analysis and manuscript revision. FK Loo JD, Conklin L, Fleming-Dutra KE, Deloria KM, Park DE, Kirk J, et al. Systematic
supervised the laboratory testing and contributed to manuscript review of the effect of pneumococcal conjugate vaccine dosing schedules on
revision. WM, BS, and NM interpreted the radiographs and prevention of pneumonia. Pediatr Infect Dis J 2014;33(Suppl. 2):S140–51, doi:
http://dx.doi.org/10.1097/INF.0000000000000082.
contributed to manuscript revision. CGW and SBO contributed Loughlin AM, Hsu K, Silverio AL, Marchant CD, Pelton SI. Direct and indirect effects
to the data interpretation and manuscript revision. KP, KMAK, SJR, of PCV13 on nasopharyngeal carriage of PCV13 unique pneumococcal serotypes
FA, KI, and HKZ participated in the data collection and manuscript in Massachusetts’ children. Pediatr Infect Dis J 2014;33(5):504–10.
Maria da Gloria SC, Tondella ML, McCaustland K, Weidlich L, McGee L, Mayer LW,
revision. et al. Evaluation and improvement of real-time PCR assays targeting lytA, ply,
and psaA genes for detection of pneumococcal DNA. J Clin Microbiol 2007;45
Members of the Pakistan Pneumococcal Vaccine Study Group (8):2460–6.
NCSS. Statistical software. PASS Sample size version 11.0. 2019 https://www.ncss.
include com/software/pass/.
O’Brien KL, Wolfson LJ, Watt JP, Henkle E, Deloria-Knoll M, McCall N, et al. Burden of
Dr Salma Sheikh, Dr Yasmin Memon, Dr Rehman Siyal (Civil disease caused by Streptococcus pneumoniae in children younger than 5 years:
global estimates. Lancet 2009;374(9693):893–902.
Hospital Hyderabad-Sindh); Dr Heeramani Lohana (Aga Khan
O’Grady K-A, Taylor-Thomson DM, Chang AB, et al. Rates of radiologically confirmed
Maternal and Child Care Centre, Hyderabad); Dr Shaheen Rizvi pneumonia as defined by the World Health Organization in Northern Territory
(Bhittai Hospital, Hyderabad-Sindh); Dr Ghulam Murtaza, Dr Indigenous children. Med J Aust 2010;192(10):592–5.
Pakistan Bureau of Statistics. Provisional summary results of 6th population and
Muzzamil Ejaz (Civil Hospital, Karachi); Dr Lal Mir Shah (District
housing census-. Government of Pakistan; 2017 http://www.pbs.gov.pk/
Headquarter Thatta, Sindh); Dr Khalid Masood, Dr Samra Baig content/provisional-summary-results-6th-population-and-housing-census-
(Sindh Government Qatar Hospital-Karachi-Sindh); Dr Agha 2017-0. [Accessed February 2018].
Ashafaque (Project Director Expanded Program of Immuniza- Pakistan Demographic and Health Survey 2012–13. Islamabad, Pakistan, and
Calverton. Maryland, United States of America: National Institute of Population
tion-Sindh), Abdul H. Ansari (Taluqa Headquarter Hospital-Hala- Studies and ICF International; 2013 http://www.nips.org.pk/abstract_files/
Sindh); Dr Abdul Jabbar, Dr Pushpa (Taluqa Headquarter Hospital- PDHS%20Final%20Report%20as%20of%20Jan%2022-2014.pdf. [Accessed 28 June
Matiari-Sindh). 2018].
Palmu AA, Jokinen J, Borys D, Nieminen H, Ruokokoski E, Siira L, et al. Effectiveness
of the ten-valent pneumococcal Haemophilus influenzae protein D conjugate
Appendix A. Supplementary data vaccine (PHiD-CV10) against invasive pneumococcal disease: a cluster
randomised trial. Lancet 2013;381(9862):214–22.
Pearce N. Analysis of matched case-control studies. BMJ 2016;352:i969.
Supplementary data associated with this article can be found, in Scott JA, Wonodi C, Moïsi JC, Deloria-Knoll M, DeLuca AN, Karron RA, et al. The
the online version, at https://doi.org/10.1016/j.ijid.2018.12.007. definition of pneumonia, the assessment of severity, and clinical standardiza-
tion in the Pneumonia Etiology Research for Child Health study. Clin Infect Dis
2012;54(Suppl. 2):S109–16.
References Shakoor S, Kabir F, Khowaja AR, Qureshi SM, Jehan F, Qamar F, et al. Pneumococcal
serotypes and serogroups causing invasive disease in Pakistan, 2005-2013. PLoS
Ali A, Husain S, Riaz A, Khawar H. Status of introduction of pneumococcal conjugate One 2014;9(6)e98796.
vaccine in Pakistan. Pediatr Infect Dis 2016;8(2):64–6. WHO. Estimated Hib and pneumococcal deaths for children under 5 years of age.
Anon. Routine vaccination coverage in selected districts of Sindh: a cross sectional 2000.
survey. 2017 [submitted for publication]. The World Bank: Data. GDP per capita (current US$). 2019. http://data.worldbank.
Bhutta ZA, Hafeez A, Rizvi A, Ali N, Khan A, Ahmed F, et al. Reproductive, maternal, org/indicator/NY.GDP.PCAP.CD.
newborn, and child health in Pakistan: challenges and opportunities. Lancet World Health Organization. Standardization of interpretation of chest radiographs
2013;381(9884):2207–18. for the diagnosis of pneumonia in children. World Health Organization
Conklin L, Loo JD, Kirk J, Fleming-Dutra KE, Knoll MD, Park DE, et al. Systematic Pneumonia Vaccine Trial Investigators’ Group. Geneva: Department of Vaccines
review of the effect of pneumococcal conjugate vaccine dosing schedules on and Biologicals, WHO; 2001 Available at http://apps.who.int/iris/bitstream/
vaccine-type invasive pneumococcal disease among young children. Pediatr 10665/66956/1/WHO_V_and_B_01.35.pdf.
Infect Dis J 2014;33(Suppl. 2):S109–18, doi:http://dx.doi.org/10.1097/ Zar HJ, Madhi SA, Aston SJ, Gordon SB. Pneumonia in low and middle income
INF.0000000000000078. countries: progress and challenges. Thorax 2013; thoraxjnl-2013-204247.
Domingues CMAS, Verani JR, Renoiner EIM, Brandileone MCC, Flannery B, Oliveria
LH, et al. Effectiveness of ten-valent pneumococcal conjugate vaccine against