Global, regional, and national estimates of the impact of a maternal Klebsiella pneumoniae vaccine: A Bayesian modeling analysis

Chirag K Kumar; Kirsty Sands; Timothy R Walsh; Seamus O’Brien; Mike Sharland; Joseph A Lewnard; Hao Hu; Padmini Srikantiah; Ramanan Laxminarayan

doi:10.1371/journal.pmed.1004239

. 2023 May 22;20(5):e1004239. doi: 10.1371/journal.pmed.1004239

Global, regional, and national estimates of the impact of a maternal Klebsiella pneumoniae vaccine: A Bayesian modeling analysis

Chirag K Kumar ¹, Kirsty Sands ², Timothy R Walsh ², Seamus O’Brien ³, Mike Sharland ⁴, Joseph A Lewnard ⁵, Hao Hu ⁶, Padmini Srikantiah ⁶, Ramanan Laxminarayan ^1,^7,^*

PMCID: PMC10270628 PMID: 37216371

Abstract

Background

Despite significant global progress in reducing neonatal mortality, bacterial sepsis remains a major cause of neonatal deaths. Klebsiella pneumoniae (K. pneumoniae) is the leading pathogen globally underlying cases of neonatal sepsis and is frequently resistant to antibiotic treatment regimens recommended by the World Health Organization (WHO), including first-line therapy with ampicillin and gentamicin, second-line therapy with amikacin and ceftazidime, and meropenem. Maternal vaccination to prevent neonatal infection could reduce the burden of K. pneumoniae neonatal sepsis in low- and middle-income countries (LMICs), but the potential impact of vaccination remains poorly quantified. We estimated the potential impact of such vaccination on cases and deaths of K. pneumoniae neonatal sepsis and project the global effects of routine immunization of pregnant women with the K. pneumoniae vaccine as antimicrobial resistance (AMR) increases.

Methods and findings

We developed a Bayesian mixture-modeling framework to estimate the effects of a hypothetical K. pneumoniae maternal vaccine with 70% efficacy administered with coverage equivalent to that of the maternal tetanus vaccine on neonatal sepsis infections and mortality. To parameterize our model, we used data from 3 global studies of neonatal sepsis and/or mortality—with 2,330 neonates who died with sepsis surveilled from 2016 to 2020 undertaken in 18 mainly LMICs across all WHO regions (Ethiopia, Kenya, Mali, Mozambique, Nigeria, Rwanda, Sierra Leone, South Africa, Uganda, Brazil, Italy, Greece, Pakistan, Bangladesh, India, Thailand, China, and Vietnam). Within these studies, 26.95% of fatal neonatal sepsis cases were culture-positive for K. pneumoniae. We analyzed 9,070 K. pneumoniae genomes from human isolates gathered globally from 2001 to 2020 to quantify the temporal rate of acquisition of AMR genes in K. pneumoniae isolates to predict the future number of drug-resistant cases and deaths that could be averted by vaccination.

Resistance rates to carbapenems are increasing most rapidly and 22.43% [95th percentile Bayesian credible interval (CrI): 5.24 to 41.42] of neonatal sepsis deaths are caused by meropenem-resistant K. pneumoniae. Globally, we estimate that maternal vaccination could avert 80,258 [CrI: 18,084 to 189,040] neonatal deaths and 399,015 [CrI: 334,523 to 485,442] neonatal sepsis cases yearly worldwide, accounting for more than 3.40% [CrI: 0.75 to 8.01] of all neonatal deaths. The largest relative benefits are in Africa (Sierra Leone, Mali, Niger) and South-East Asia (Bangladesh) where vaccination could avert over 6% of all neonatal deaths. Nevertheless, our modeling only considers country-level trends in K. pneumoniae neonatal sepsis deaths and is unable to consider within-country variability in bacterial prevalence that may impact the projected burden of sepsis.

Conclusions

A K. pneumoniae maternal vaccine could have widespread, sustained global benefits as AMR in K. pneumoniae continues to increase.

Using a Bayesian modelling approach, Ramanan Laxminarayan and colleagues, explore the global impact of a maternal vaccine against Klebsiella pneumoniae on neonatal sepsis and associated mortality.

Author summary

Why was this study done?

Approximately 1 million newborns yearly die within the first 4 weeks of life due to bacteria infecting their bloodstream with Klebsiella pneumoniae (K. pneumoniae) as the leading cause of such infections.
There have been numerous recent advancements in developing viable K. pneumoniae vaccine and antibody-based treatments in preclinical models, with some treatments reaching Phase 1 clinical trials.
The impacts of vaccination must be quantified and may prove useful to prioritizing vaccine distribution and better understanding the burden of sepsis.

What did the researchers do and find?

Using a Bayesian mixture-model based on clinical surveillance of neonatal sepsis, we present country-specific estimates for the number of deaths and cases of antimicrobial-resistant neonatal sepsis caused by K. pneumoniae that would be averted if a vaccine with 70% efficacy were given to pregnant mothers.
We find that most cases of K. pneumoniae neonatal sepsis are resistant to first-line treatments, such as ampicillin and gentamicin.
We estimate that a vaccine with 70% efficacy could prevent 399,015 [95th percentile credible interval (CrI): 334,523 to 485,442] cases and 80,258 [CrI: 18,084 to 189,040] neonatal deaths.

What do these findings mean?

A maternal vaccine that confers newborns with protection from K. pneumoniae infection could reduce neonatal sepsis deaths in many low- and middle-income countries (LMICs) by nearly 15%.
This would help to achieve targets set by the World Health Organization (WHO) for improved child health globally and to mitigate inequities in neonatal survival in low- and middle-income settings compared to high-income settings.
Reducing cases of neonatal sepsis by vaccination could also contribute to reduced antibiotic use, subsequent improvements in antimicrobial resistance (AMR) rates, and a reduction in healthcare utilization and expenditure.

Introduction

Each year, an estimated 800,000 newborns die within the first 4 weeks of life due to sepsis [1–3]. Several studies, including most recently the Child Health and Mortality Prevention Surveillance (CHAMPS) [4] and the Burden of Antibiotic Resistance in Neonates from Developing Societies (BARNARDS) [5], indicate that Klebsiella pneumoniae is the leading cause of neonatal sepsis (Fig A in S1 Text). Increasing multidrug resistance (MDR) and lack of access to appropriate antibiotics contribute to high levels of morbidity and mortality among K. pneumoniae sepsis cases [6–8]. Though improved sanitation and infection control could limit K. pneumoniae transmission, they can be challenging to implement in many resource-poor settings [9]. A maternal vaccine against K. pneumoniae that confers protection to newborns via transplacental antibody transfer could reduce the burden of neonatal sepsis, but the extent of corresponding reductions in sepsis remains unknown.

Primary approaches to developing a K. pneumoniae vaccine have targeted the K capsular polysaccharide antigens or the O lipopolysaccharide antigens—both of which demonstrate significant diversity across K. pneumoniae strains—or an inactivated whole cell [10]. Animal data from an early polysaccharide vaccine construct demonstrated a robust immune response and subsequent neutralization of K. pneumoniae bacterial isolates in mouse models when administered intravenously [11]. A 24-valent Klebsiella capsular polysaccharide vaccine tested in humans demonstrated high anti-polysaccharide immunoglobin titers [12], but a clinical trial to validate this vaccine was inconclusive because of a limited supply of materials [13]. More recently, there is growing interest in conjugate polysaccharide vaccines targeting the K and/or O antigens [14–17] with the first such vaccine entering a clinical phase [18,19]. Combined, these data suggest that a maternal vaccine to prevent K. pneumoniae demands further attention and investigation. As with previous [20] and future [21] proposed maternal vaccination programs, this vaccine would be administered to pregnant mothers, triggering an immune response that protects the mother and, critically, transfers maternal antibodies to the developing baby, providing protection to the neonate (or young infant) through the period of greatest risk [22]. Vaccination could reduce the overall burden of sepsis and decrease antibiotic use (fewer K. pneumoniae sepsis cases) [23] and K. pneumoniae nosocomial outbreaks (vaccination generates community-level immunity) [24], which are common in LMICs [5]. Additionally, by averting infection and reducing the need for treatment, vaccination may prevent lengthy hospital stays for newborn sepsis, thereby reducing the economic burden on LMIC families where healthcare costs are often deferred to the patient [7].

In this paper, we estimated the global, regional, and national impact of a potential maternal vaccine against K. pneumoniae deployed at the current levels of coverage of the maternal tetanus vaccine. We also examined the antimicrobial resistance (AMR) of sepsis-causing isolates. Finally, we estimated the continued benefits of such a vaccine in 2030 under future scenarios of increasing AMR.

Methods

We used data from 3 global clinical data sets of neonatal sepsis and a Bayesian modeling framework to estimate the annual neonatal infections and deaths by country caused by K. pneumoniae (although the methodology is applicable to any other sepsis-causing bacterial etiology; Figs O to U in S1 Text). We used target vaccine efficacy and coverage to project the total number of infections and deaths that could be averted using a maternal vaccine. We complemented these analyses with a mathematical model that estimated the fraction of infections and deaths that are resistant to the antibiotics recommended by World Health Organization (WHO) treatment guidelines for neonatal sepsis (ampicillin and gentamicin as a first-line treatment and amikacin and ceftazidime as a second-line treatment) [25] and those most used to treat neonatal sepsis (adding meropenem to the above list) [7].

Data

Three data sets were used to quantify K. pneumoniae neonatal sepsis burden: the Child Health and Mortality Prevention Surveillance (CHAMPS) [4], the Burden of Antibiotic Resistance in Neonates from Developing Societies (BARNARDS) [5], and the Global Neonatal Sepsis Observational Study (NeoObs) [26]. CHAMPS, initiated in 2016, conducts minimally invasive tissue sampling (MITS) on deceased children under 5 to determine causal drivers of death at 8 study sites in 7 low- and middle-income countries (LMICs) across Africa and South-East Asia (Bangladesh, 2 sites in Ethiopia, Kenya, Mali, Mozambique, Sierra Leone, and South Africa). In contrast, BARNARDS and NeoObs are birth-cohort studies that tracked neonates prospectively through their first month of life, evaluating incidence of sepsis, the antibiotic susceptibility of associated isolates, and resulting mortality. BARNARDS surveilled neonates across 12 study sites in 7 countries in Africa and South-East Asia (2 sites in Bangladesh, India, 2 in Pakistan, Ethiopia, 3 in Nigeria, 2 in Rwanda, and South Africa) from 2015 to 2017. Finally, NeoObs reported neonatal sepsis cases, deaths, and antibiotic susceptibility of the isolated bacteria in 19 sites across 11 countries in Europe (Italy, Greece), Latin America (2 sites in Brazil), Africa (Kenya, Uganda, 3 sites in South Africa), South-East Asia (Bangladesh, 3 sites in India, 2 sites in Thailand), and the Western Pacific (3 sites in China, Vietnam) from 2018 to 2020. Table A in S1 Text provides the number of neonates surveilled over the study period by country. We restricted all analyses to cases within each study with culture-confirmed K. pneumoniae sepsis in neonates based on isolates obtained from blood or cerebrospinal fluids (CSF) (Fig B in S1 Text).

Modeling techniques

We used a Bayesian mixture-modeling approach to aggregate data across the 3 studies and produce consistent and robust estimates of neonatal sepsis infections, deaths, and drug resistance distributions. We estimated p_s,l, the probability that K. pneumoniae was the cause of death for a neonate who died from neonatal sepsis, in each location, l, and for each study, s. We assumed a noninformative uniform prior on p_s,l at all locations. We modeled whether K. pneumoniae was present in a clinical sample from a neonate who died of culture-confirmed sepsis as a binomial random variable and analytically solved for the posterior distribution of p_s,l (Section 1 in S2 Text):

p_{s, l} \sim β (N_{k p}^{s, l} + 1, N_{d}^{s, l} - N_{k p}^{s, l} + 1),

where N_kp^{s, l} is the number of neonates who were culture-positive for K. pneumoniae and died, N_d^{s, l} is the number of neonates who had culture-confirmed sepsis (from any etiology) and died, both quantities at each location, l, in each study, s. β indicates the standard univariate beta distribution.

In countries where multiple studies conducted surveillance (ex., both CHAMPS and BARNARDS had study sites in the country), we obtained multiple independent estimates of the distribution of p_s,l, one for each study that had a site in the country. We used a mixture-modeling approach to aggregate these distributions so that each independent distribution for p_s,l was weighted proportional to the number of neonates sampled in the study at that location:

p_{l} = \sum_{s = 1}^{3} \frac{N_{d}^{s, l}}{N_{d}^{1, l} + N_{d}^{2, l} + N_{d}^{3, l}} p_{s, l} .

We used a mixture-modeling approach rather than aggregating the samples into 1 beta distribution to account for subpopulation variability across sites within the same country from different studies. To efficiently characterize the resulting distribution, we drew 10,000 samples using Latin hypercube sampling (LHS) [27]. We chose LHS over another distribution sampling technique such as Monte Carlo because LHS converges to the true distribution faster.

We extended this analysis to estimate the absolute number of deaths attributable to K. pneumoniae using data from the United Nations Child Health Epidemiology Reference Group (CHERG) [1] and the University of Washington’s Institute for Health Metrics and Evaluation Global Burden of Disease (GBD) [28,29] on the total number of neonatal deaths and the total number of neonatal sepsis deaths. Estimates of p_l were mapped into estimates of the number of deaths and the percentage of all neonatal deaths attributable to K. pneumoniae by scaling the estimates of p_l by the estimates with uncertainty from CHERG and GBD for the total number of neonatal sepsis deaths. Unless otherwise mentioned, we report values derived using CHERG data.

We determined the number of vaccine-avertable deaths associated with the deployment of a maternal K. pneumoniae vaccine at an effective coverage level equal to that of the maternal tetanus vaccine (median: 90%; range: 38.5% to 100% of pregnant women immunized) [30]. For modeling purposes, we assumed that the efficacy of the vaccine would be 70%: This estimate is based on a conjugate vaccine candidate that targets the 15 most common K. pneumoniae capsular serotypes that cause invasive infections in neonates. Current sero-epidemiology suggests that these serotypes account for approximately 70% of neonatal sepsis cases [14]. To determine the avertable cases from avertable deaths, we estimated the case fatality ratio (CFR) for K. pneumoniae sepsis using data from BARNARDS. We used the same Bayesian beta framework with a uniform prior described above (we treated whether the neonate died from K. pneumoniae or recovered as the binomial event for which we seek to estimate the probability parameter) to estimate the CFR.

To correctly propagate uncertainty across combination products (i.e., when trying to determine the number of cases from the number of deaths using the CFR), we individually multiplied independent samples from the distributions for p_s,l and the CFR. We summarized the resulting distribution using the median and 95% credible intervals (CrIs). We use this approach throughout for propagating uncertainty.

Finally, we estimated antibiotic resistance profiles using data from BARNARDS and NeoObs at the WHO regional level (Section 2 in S2 Text). We used the Bayesian beta and mixture-modeling framework with a noninformative prior described earlier to estimate the probability that an isolate was resistant to a given antibiotic.

We extrapolated our estimates from the selected countries for which we had neonatal sepsis surveillance data global estimates for all countries of avertable neonatal sepsis deaths and cases by fitting regression equations. We related the total number of neonatal deaths due to sepsis (for which estimates are available from CHERG) to the estimated number of avertable neonatal deaths if a K. pneumoniae vaccine were distributed as described above using standard ordinary least squares. We also calculated the number of avertable antibiotic-resistant neonatal deaths by country using our estimates for antibiotic resistance at the WHO regional level.

To estimate the future benefits of such a vaccine, we analyzed the temporal trends in the presence of AMR genes in K. pneumoniae by using isolate genomes available on PathogenWatch [31] and in the BARNARDS data set. We limited our analysis to human isolates from clinical sources that were not taken as rectal or carriage samples (S1 Table and Fig C in S1 Text). We gathered metadata on the isolates, including their date, source, and country. In total, we analyzed 9,070 K. pneumoniae genomes from 68 countries from 2001 to 2020 (Figs D and E in S1 Text). To identify antimicrobial resistance genes (ARGs) in each genome, we used ABRicate (v1.1.0) with the reference database ResFinder (last updated on June 18, 2022). We fit linear probability model equations to the temporal trends in the number of aminoglycoside-resistant genes (AGRGs) and carbapenem-resistant genes (CRGs) and extrapolated those trends to 2030 (i.e., to predict the future number of AGRGs and CRGs). We used the fit linear probability model to determine the projected increase in ARGs and resistance by 2030 and used that to estimate the increase in the absolute number of aminoglycoside- and carbapenem-resistant neonatal sepsis cases and deaths.

Fig F in S1 Text shows a schematic diagram of the model, and Table B in S1 Text denotes all model parameters.

Results

Effect of a K. pneumoniae maternal vaccine on neonatal deaths

We initially focused on the 18 countries for which we had surveillance data on bacterial neonatal sepsis (Fig A in S1 Text). A potential vaccine could reduce neonatal sepsis and improve newborn survival (Fig 1). Our estimates of the proportion of neonatal deaths averted with maternal K. pneumoniae vaccination are consistent whether we use CHERG or GBD data as a reference source. Under the assumption that K. pneumoniae vaccine coverage matches current effective maternal tetanus vaccination coverage levels [30], a median estimated 57,392 [95th percentile Bayesian CrI: 22,248 to 125,803] deaths, accounting for 3.83% [CrI: 1.48 to 8.40] of newborn fatalities, could be averted each year in the initial 18 countries analyzed. We find a similar benefit in reducing neonatal sepsis cases, with an estimated 286,499 [CrI: 239,259 to 346,496] cases averted yearly.

Fig 1 — (A) Median estimated percent of neonatal deaths that are avertable with a maternal vaccine with respect to all neonatal deaths in a given country. (B) Median estimated number of neonatal deaths that are avertable with a maternal vaccine in a given country. (C) Median estimated number of neonatal cases of sepsis that are avertable with a maternal vaccine in a given country. GBD indicates data from the Global Burden of Disease study and CHERG indicates data from the United Nations Child Health Epidemiology Reference Group. Error bars indicate 95th percentile CrIs. A pseudo log transformation is done for values between 0 and 1 in panel B. CHERG, Child Health and Epidemiology Reference Group; CrI, credible interval; GBD, Global Burden of Disease.

The benefits of a maternal K. pneumoniae vaccine vary across countries. While the relative impact of vaccination on neonatal mortality ranges a few percentage points (Fig 1A), the estimates for the reduction in total neonatal sepsis deaths and cases span orders of magnitude across countries, reflecting the variability in the burden of neonatal deaths across countries. We project that vaccination will have the greatest relative effect on neonatal mortality in Ethiopia, Rwanda, and Sierra Leone and a smaller effect in Greece, China, and Pakistan (Fig 1A), countries where a lower fraction of neonatal sepsis deaths were associated with K. pneumoniae (Fig A in S1 Text). However, although the relative impact of vaccination on neonatal mortality (i.e., percent of deaths averted with respect to all neonatal deaths) in Pakistan is smaller compared to other countries considered, we estimate that Pakistan will have one of the greatest absolute reductions in neonatal mortality (i.e., a high number of absolute deaths/cases averted; Fig 1B and 1C), consistent with previous work that Pakistan experiences a significant K. pneumoniae burden from nosocomial outbreaks [5].

Across all WHO regions, K. pneumoniae isolates are highly resistant to most antibiotics (Fig 2). We focused on those antibiotics often prescribed to treat neonatal sepsis: ampicillin, gentamicin, ceftazidime, amikacin, and meropenem (Fig G in S1 Text for all antibiotics used to treat sepsis) [7,25]. Prevalence of resistance to ampicillin is highest: 89.82% [CrI: 64.97 to 99.49] of K. pneumoniae isolates are resistant to ampicillin worldwide (Fig 2B). The average rate of resistance to gentamicin, which is usually prescribed in combination with ampicillin as the first-line treatment for sepsis per current WHO guidelines [32], is 57.22% [CrI: 31.540 to 80.42] (Fig 2B). Comparatively, fewer isolates were resistant to ceftazidime and amikacin—frequently employed as a second-line combination to treat neonatal sepsis—with 44.63% [26.84 to 72.20] and 35.80% [14.69 to 58.53], respectively, isolates identified as resistant (Fig 2B). Approximately 22.43% [5.24 to 41.42] of isolates from neonates who died were resistant to meropenem.

Because ampicillin and gentamicin are prescribed together as the first-line treatment for neonatal sepsis in many countries, we considered the joint distribution of sepsis isolates that were resistant to neither, one, or both antibiotics (Fig 2C and 2D). Resistance is high against both drugs, with an average of 86.38% [CrI: 73.09 to 94.74] of isolates resistant to both drugs across all regions (Fig 2D). Notably, rates of resistance to ampicillin alone were substantially higher than to gentamicin alone (Fig 2D). In fact, less than 5% of isolates were resistant to just gentamicin (Fig 2D). Consequently, resistance to gentamicin almost guaranteed resistance to ampicillin (Fig 2D).

Different antibiotics within the same antibiotic class display varying geographic trends. For instance, resistance to ampicillin is approximately constant across all geographic regions. Comparatively, rates of resistance to ceftazidime and meropenem are variable across geographic regions: Rates of resistance to ceftazidime in Africa are high and close to those of ampicillin, approximately 60 percentage points greater than rates of resistance to meropenem. Resistance to meropenem is low (7.01% [CrI: 3.21 to 12.64]) in Africa and substantially higher in South-East Asia (66.58% [CrI: 52.68 to 78.82]), greater than resistance to ceftazidime (Fig 2B). Among aminoglycosides, resistance rates to amikacin and gentamicin are similar in all regions except Africa where gentamicin resistance is 60 percentage points greater than amikacin resistance (Fig 2B). Finally, we consider trends across antibiotics within the same treatment regimen; in particular, we compare ampicillin and gentamicin (first-line treatment) against ceftazidime and amikacin (second-line treatment). While rates of resistance to ampicillin are consistently greater than to gentamicin, rates of resistance to ceftazidime are greater than amikacin only in Africa and the Western Pacific (Fig 2B).

Across all regions, we generally did not find an increase in drug-resistant K. pneumoniae isolates from neonates who died compared with all K. pneumoniae isolates in general: The prevalence of resistance to a specific drug class is approximately equal whether the isolate was taken from a septic neonate who died or not (Fig 2A and 2B). However, there are 2 key exceptions: In South-East Asia, rates of resistance to amikacin, gentamicin, and ceftazidime are higher in neonates who died, and rates of resistance to ampicillin are lower among all neonates. Nevertheless, we do not find that isolates from neonates who succumbed to sepsis were more likely to exhibit AMR (Fig 2A and 2B), resistance to ampicillin or gentamicin (Fig 2C and 2D), or higher MDR (Fig 2E and 2F) in comparison to all isolates.

Most K. pneumoniae have become resistant to multiple antibiotics (Fig 2E and 2F). On average, more than 99% of isolates were resistant to at least 1 drug. Many isolates were resistant to multiple antibiotics, and isolates that were resistant to none were rare (0.15% [CrI: 2.35 × 10−9 to 18.17]). The highest MDR rates are observed in the Eastern Mediterranean and South-East Asia where over 98% of K. pneumoniae are resistant to either 4 or 5 different antibiotics (Fig 2F); however, in Africa, most isolates are resistant to 3 antibiotics only with fewer isolates resistant to 4 or 5 antibiotics (Fig 2F).

Projected global and future benefits of a K. pneumoniae maternal vaccine

We expanded our estimates of the effects of a maternal K. pneumoniae vaccine on childhood mortality and neonatal sepsis to all countries using regression models as described above (Table C in S1 Text, R² of 84.0%, p < 0.001). We project that 14.91% [CrI: 5.26 to 21.24] of all neonatal sepsis deaths in countries of interest (all countries except for high-income nations)—that is, 80,258 [CrI: 18,084 to 189,040] neonatal deaths or 3.40% [CrI: 0.75 to 8.01] of all neonatal mortality—and 399,015 [CrI: 334,523 to 485,442] neonatal sepsis cases could be averted yearly once vaccination coverage reaches that of the maternal tetanus vaccine (Figs 3A and 3B, and Figs H and I in S1 Text for the 95th percentile CrIs, and S2 Table for the numeric values). We project the greatest reduction in overall neonatal mortality (i.e., fraction of deaths reduced due to vaccination) to be in Africa, specifically in sub-Saharan Africa, with a significant but lesser reduction in South-East Asia, the Eastern Mediterranean, and parts of Latin America. The greatest decreases in absolute deaths from neonatal sepsis are in India followed by sub-Saharan Africa; there were approximately equal decreases in the number of deaths in the other regions.

Fig 3 — (A) Projected percent of neonatal deaths with respect to all neonatal deaths that could be averted due to a maternal K. *pneumoniae* vaccine. (B) Projected median number of neonatal sepsis deaths averted due to maternal vaccination. (C) Projected number of ceftazidime-resistant neonatal sepsis deaths averted due to maternal vaccination. (D) Projected number of meropenem-resistant neonatal sepsis deaths averted due to maternal vaccination. The maps are reprinted from pygal_maps_world under GNU GPL.

In general, we project regions with the greatest reduction in K. pneumoniae neonatal sepsis are likely to experience the greatest reduction in neonatal deaths due to antimicrobial-resistant K. pneumoniae sepsis (Figs 3C and 3D and Figs J–L in S1 Text). However, whereas overall neonatal mortality would be considerably reduced across Africa and South Asia, the greatest reduction in resistant K. pneumoniae deaths would occur in Central Africa, East Africa, and India (Fig 3). Antibiotic resistance remains a significant issue in the treatment of K. pneumoniae because the projected number of both ceftazidime-resistant deaths (Fig 3C) and meropenem-resistant deaths (Fig 3D) is only marginally less than the projected number of total sepsis deaths (Fig 3B and S2 Table). There are likely to be more ceftazidime-resistant deaths than meropenem-resistant deaths averted globally. However, the relative burden of ceftazidime resistance is greater in North Africa while the burden of meropenem resistance is greater in South Asia. Regardless, for both drugs, we estimate high resistance rates worldwide and a significant number of deaths caused by drug-resistant infections could be averted through maternal vaccination.

Finally, we project the benefits of a maternal vaccine in future scenarios of increased AMR prevalence. Available K. pneumoniae genomes indicate that the prevalence of carbapenemase ARGs (CRGs) is increasing significantly worldwide (Fig 4A) at a rate of 0.050 [95th percentile linear probability model confidence interval: 0.042 to 0.058, p < 0.001] ARGs being acquired per isolate yearly (Table D in S1 Text; R² = 91.1%). Comparatively, aminoglycoside ARGs (AGRGs) increased significantly in the early 2000s (Fig 4A), but subsequently have not increased significantly (p = 0.436) and are constant at an average of 2.69 AGRGs per isolate.

(A) Average number of ARGs that confer resistance to aminoglycosides (N = 24,359) and carbapenems (N = 5,189) per isolate by year of collection of K. *pneumoniae* isolates. (B) Projected median number of meropenem-resistant neonatal deaths averted by 2030 due to maternal vaccination based on the trends shown in Fig 4A. The maps are reprinted from pygal_maps_world under GNU GPL. AMR, antimicrobial resistance; ARG, antimicrobial resistance gene.

Based on the increasing rates of CRGs, we extrapolated the trends from historic data to estimate the number of CRGs present in 2030 and calculated the associated future burden of meropenem-resistant K. pneumoniae deaths that could be averted by a potential maternal vaccine (Figs 4B and Fig M in S1 Text for the 95th percentile CrIs). The map both illustrates neonatal survival trends similar to those described above and the new regional benefits from vaccination as AMR increases. Countries that may experience only marginal benefits from vaccine distribution—such as those in the Americas including Venezuela and Colombia, those in the Western Pacific including Indonesia, those in the Eastern Mediterranean including Afghanistan, and those in South-East Asia including Bangladesh—are likely to experience greater reductions in neonatal mortality when accounting for increases in AMR.

Discussion

K. pneumoniae is the leading cause of newborn sepsis deaths worldwide (Figs O–U in S1 Text); however, no K. pneumoniae vaccine is currently approved. We estimate the global benefits of a potential maternal K. pneumoniae vaccine in reducing neonatal mortality and AMR. A K. pneumoniae vaccine has the potential to significantly improve child wellbeing, reduce a large fraction (15%) of neonatal deaths and sepsis infections, and will benefit additional geographic areas as AMR is projected to increase. Most neonatal sepsis cases are multidrug resistant, further underscoring the benefits of vaccination. Nevertheless, our modeling suggests that the bacteria causing neonatal sepsis in neonates that eventually die do not have higher antimicrobial or MDR rates. Instead, resistance is high in all isolates and against all drugs, underscoring neonatal fragility and suggesting that other factors such as access to effective drugs and antibiotic dosing regiments or standard of care may be driving differential mortality [7]. Regardless, our estimates indicate that maternal vaccination against K. pneumoniae would have global benefits and improve efforts toward reaching child survival targets in LMICs worldwide.

Our methodology represents an advantage over using just clinical data or a single study to draw conclusions and make estimates: We applied a Bayesian framework to aggregate data from various studies and generate increased predictive power. This approach more accurately reflects the underlying uncertainty and allows meaningful inference and projections, even when there are few reported and annotated cases of culture-confirmed neonatal sepsis. However, further work is required to develop a vaccine that is approved, widely distributable, and affordable. Most K. pneumoniae neonatal sepsis cases occur in premature infants [26], but the success of a maternal vaccine will depend on robust transplacental antibody transfer from mother to a developing fetus, which peaks late in the third trimester of pregnancy. Additional evaluation of the adequacy of maternal antibody transfer for premature infants is important to inform maternal vaccine development efforts.

There are several potential caveats in our study. Our estimates of averted cases and deaths assume that vaccination rates instantly reach their maximum in all countries—that is, we do not model the reduction in cases of K. pneumoniae sepsis as vaccine rollout progresses. To address this, we have conducted sensitivity analyses where we vary the vaccine efficacy/coverage (S2 Table). Our estimates for the number of avertable deaths respond linearly to vaccine efficacy and coverage. Other studies that estimate the effect of a hypothetical intervention against a communicable disease have used a similar assumption [33,34]. As more details of the potential K. pneumoniae vaccine are determined, we can refine our model accordingly. Since little is known about the factors that drive nosocomial infections and hospital-acquired neonatal sepsis, the regression models we used to produce estimates of averted cases in all countries are simple and consider only health variables. Nevertheless, previous work that has attempted to extrapolate disease burden and rates of AMR across countries has used a similar econometric-like approach [29]. Finally, our estimates of the CFR of K. pneumoniae neonatal sepsis may be biased due to loss-to-follow-up: In particular, we are unable to account for whether there is a sampling bias of a disproportionate number of deaths being recorded. Further data is required to address this issue.

The uncertainty in our estimates largely comes from uncertainty in the underlying values for annual neonatal sepsis deaths from CHERG and GBD, which is considerable. Although estimates for the percentage of averted deaths are consistent (Fig 1A), estimates for the number of avertable deaths using data from CHERG as opposed to GBD often differ by an order of magnitude (Fig 1B). This is because of underlying differences in the number of neonatal sepsis deaths from the 2 sources. Estimates using CHERG data consistently indicate a higher number, though estimates derived from GBD data are still substantial. In many cases, the CrIs from the estimates using different data sources overlap. To address this issue, better country-level data on the burden of neonatal sepsis are needed. Moreover, the sample sizes from all 3 studies were limited and we were unable to delineate hospital outbreaks of K. pneumoniae caused by a single strain that may have differential vaccine aversion. We have aimed to address this limitation using the Bayesian approach and noninformative prior, as discussed above, to enable meaningful inference while considering latent noise in the data.

Finally, because of the limited data available on the etiologies causing neonatal sepsis, our analysis only considers country-level trends for K. pneumoniae neonatal sepsis cases and WHO regional trends for AMR. However, there is heterogeneity within a country and region; our modeling is unable to capture that. We do not know whether heterogeneity within a country is due to external factors or rather indicates study-level variability: Given our limited data and few countries for which we have multiple study sites, we do not model study-level variability, though this is an area of future work. Additional work is required to better understand within-country and within-region trends: Preliminary data indicates variability in sepsis levels within a country and across private and public hospitals [5,35,36]. Moreover, we may be subject to a sampling bias because the data from the 3 studies we used are point estimates of the impact of sepsis at specific healthcare facilities within a country rather than surveillance over the whole country. In particular, we must limit our analysis to cases of culture-confirmed K. pneumoniae, but doing so makes our estimates a probable lower bound on the actual number of averted deaths and cases: it is likely that not all cases of neonatal sepsis are identified as such and cultured. Even if they are cultured, they may not test positive due to low blood volume, a bias whose impact across study sites likely varies appreciably but is unknown [37]. Additional work is required to better understand the factors that impact neonatal sepsis rates within a region. We also lacked data from many countries. We tried to minimize the effect of this by projecting the number of averted cases by country of maternal vaccination on antimicrobial-resistant K. pneumoniae neonatal sepsis on a regional basis and limiting our extrapolations to LMICs. Moreover, we used a uniform prior over the probability of mortality from K. pneumoniae to consider the variability in K. pneumoniae mortality across all countries in a region: This ensures that the CrIs for our projections to other countries include other mortality rates that may be the true underlying mortality rate for that country. Nonetheless, additional data on K. pneumoniae neonatal sepsis across more countries is required to reduce the uncertainty in our estimates.

Future work includes a cost-effectiveness analysis to estimate the economic and social implications of maternal vaccination. A K. pneumoniae vaccine would reduce the economic burden of infection for both the patient and the hospital: quantifying these impacts is critical to understanding the greater societal impacts of this vaccine, especially as the burden of sepsis is highest in LMICs that have larger health expenditure (Fig N in S1 Text). Additionally, K. pneumoniae is known to be the leading driver of sepsis, especially in the late neonatal stage: reducing sepsis during this timeframe would be significantly advantageous to maternal admissions and mitigating healthcare burden as it would reduce the length of prolonged hospital stays. Beyond reducing cases and deaths, a K. pneumoniae vaccine would likely also reduce antibiotic usage and thus may help improve antibiotic stewardship in LMICs and reduce AMR rates. It is noteworthy that in many LMICs, antibiotic availability is challenging (particularly for the more potent drugs) and the cost is in many cases deferred to the family; therefore, a maternal vaccine will also alleviate drug demand and provide local financial benefits. Moreover, vaccination may reduce the severity of disease and have positive social ramifications by reducing stigma surrounding infection and helping mothers who would otherwise leave their jobs to care for sick newborns [38]. We have highlighted the main benefits of the rollout of a potential K. pneumoniae maternal vaccine: reducing global cases and deaths of resistance and susceptible neonatal sepsis, reducing overall neonatal mortality, and improving childhood health.

Supporting information

S1 Text. Supplementary materials.

Fig A. Raw data. Calculated percent of neonatal sepsis deaths that are associated (i.e., an isolate from the neonate who died was culture-positive) with various etiologies across each study by location. Table A. Number of neonates who died of neonatal sepsis divided by number of neonates surveilled by location. Fig B. Flow diagram summarizing cases of culture-confirmed sepsis used in the main analysis of vaccine-avertable sepsis and AMR. Fig C. Flow diagram summarizing data collection and cleaning of the Klebsiella pneumoniae genomes used in the antimicrobial resistance genes (ARG) prevalence analysis. BARNARDS refers to data gathered from the Burden of Antimicrobial Resistance in Neonates in Developing Societies study. Fig D. Distribution of available genomic data from PathogenWatch and the Burden of Antimicrobial Resistance in Neonates from Developing Societies study by year used in the antimicrobial resistance gene prevalence analysis. Fig E. Tree map of the distribution of available K. pneumoniae isolates across countries for use in the prevalence of antimicrobial resistance genes analysis. Colors have no meaning and are used to create contrast between countries. Fig F. Schematic diagram of the modeling framework. Table B. Model parameters. Note that this refers to values that are used as inputs to various modeling stages, not quantities that are predicted through the modeling process described in Fig F. Fig G. Raw resistance data by study. Table C. Regression analysis results for the model used to extrapolate the number of averted deaths from those countries for which we have data to all countries. Fig H. 2.5th percentile (i.e., lower bound) of estimates represented as maps shown in Fig 3. The maps are reprinted from pygal_maps_world under GNU GPL. Fig I. 97.5th percentile (i.e., lower bound) of estimates represented as maps shown in Fig 3. The maps are reprinted from pygal_maps_world under GNU GPL. Fig J. As Fig 3C and 3D but for ampicillin. Median estimates shown on top. 2.5th percentile shown in middle. 97.5th percentile shown on bottom. The maps are reprinted from pygal_maps_world under GNU GPL. Fig K. As Fig 3C and 3D but for gentamicin. Median estimates shown on top. 2.5th percentile shown in middle. 97.5th percentile shown on bottom. The maps are reprinted from pygal_maps_world under GNU GPL. Fig L. As Fig 3C and 3D but for amikacin. Median estimates shown on top. 2.5th percentile shown in middle. 97.5th percentile shown on bottom. The maps are reprinted from pygal_maps_world under GNU GPL. Table D. Regression analysis results for the model used to estimate the yearly rate of increase in antimicrobial resistance genes. Fig M. Credible interval of map shown in Fig 4B. 2.5th percentile shown on top and 97.5th percentile shown on bottom. The maps are reprinted from pygal_maps_world under GNU GPL. Fig N. Health expenditure as a fraction of the country’s GDP. Data courtesy of the WHO, Global Health Observatory (2022). Map reprinted from OurWorldInData under a CC-BY license. Original: https://ourworldindata.org/grapher/total-healthcare-expenditure-gdp. Fig O. As Fig 1A but for other etiologies of interest. Median estimated fraction of neonatal deaths averted given maternal vaccination against a specific pathogen at 70% efficacy and coverage equivalent to that of the maternal tetanus vaccine. Median shown; error bars indicate 95th percentile Bayesian credible intervals. GBD refers to data from the Global Burden of Disease study, and CHERG refers to data from the Child Health and Epidemiology Reference Group. Fig P. As Fig 1B but for other etiologies of interest. Median estimated number of avertable neonatal sepsis deaths given maternal vaccination against a specific pathogen at 70% efficacy and coverage equivalent to that of the maternal tetanus vaccine. Median shown; error bars indicate 95th percentile Bayesian credible intervals. A pseudo log transform is done for values between 0 and 1. GBD refers to data from the Global Burden of Disease study, and CHERG refers to data from the Child Health and Epidemiology Reference Group. Fig Q. As Fig 1C but for other etiologies of interest. Median estimated number of avertable neonatal sepsis cases given maternal vaccination against a specific pathogen at 70% efficacy and coverage equivalent to that of the maternal tetanus vaccine. Median shown; error bars indicate 95th percentile Bayesian credible intervals. GBD refers to data from the Global Burden of Disease study, and CHERG refers to data from the Child Health and Epidemiology Reference Group. Fig R. As Fig 2A but for other etiologies and relevant antibiotics of interest. Estimated median fraction of isolates from neonates who died with culture-confirmed sepsis that are resistant to various drugs across WHO regions. Median shown; error bars indicate 95th percentile Bayesian credible intervals. Fig S. As Fig 2B but for other etiologies and relevant antibiotics of interest. Estimated median fraction of isolates from neonates with culture-confirmed sepsis that are resistant to various drugs across WHO regions. Median shown; error bars indicate 95th percentile Bayesian credible intervals. Fig T. As Fig 2E but for other etiologies of interest. Antibiotics considered are shown in Figs R and S. Median shown; error bars indicate 95th percentile Bayesian credible intervals. Fig U. As Fig 2F but for other etiologies of interest. Antibiotics considered are shown in Figs R and S. Median shown; error bars indicate 95th percentile Bayesian credible intervals.

(PDF)

Click here for additional data file.^{(4.4MB, pdf)}

S2 Text. Extended methods.

Derivation of the mixture model posterior distribution (Section 1) and details of antimicrobial susceptibility testing done (Section 2).

(PDF)

Click here for additional data file.^{(273.2KB, pdf)}

S1 Table. Details of the analyzed genomes used in this study including FASTA file ID, year, source, country of origin, and number of identified antimicrobial resistance genes.

(XLSX)

Click here for additional data file.^{(797.3KB, xlsx)}

S2 Table. Impact of maternal vaccination by country.

Country-specific estimates of overall averted cases, deaths, and result on neonatal mortality as well as antibiotic resistant deaths.

(XLSX)

Click here for additional data file.^{(148.9KB, xlsx)}

Acknowledgments

We thank the BARNARDS and NeoObs groups for sharing their data. In particular, TRW would like to personally thank Kathryn Thomson and Rebecca Milton, both part of the BARNARDS group. We thank Dianna Blau from the CHAMPS group for valuable feedback. We would also like to thank each CHAMPS site team. We thank Nicole Benson, at the Bill & Melinda Gates Foundation, Mateusz Hasso-Agopsowicz and Isabel Frost, both at the World Health Organization, and Sonya Davey, at Brigham and Women’s Hospital for valuable feedback. We thank Sally Atwater for valuable editorial suggestions on drafts.

Abbreviations

AGRG: aminoglycoside-resistant gene
AMR: antimicrobial resistance
ARG: antimicrobial resistance gene
BARNARDS: Burden of Antibiotic Resistance in Neonates from Developing Societies
CFR: case fatality ratio
CHAMPS: Child Health and Mortality Prevention Surveillance
CHERG: Child Health Epidemiology Reference Group
CRG: carbapenem-resistant gene
CrI: credible interval
CSF: cerebrospinal fluid
GBD: Global Burden of Disease
LHS: Latin hypercube sampling
LMIC: low- and middle-income country
MDR: multidrug resistance
MITS: minimally invasive tissue sampling
WHO: World Health Organization

Data Availability

All underlying model code and freely and publicly available data (i.e., CHERG and GBD data, maternal tetanus vaccination rates) that were used in this analysis are available from https://github.com/ChiragKumar9/KpVaccine. The CHAMPS data are freely available online upon request. In this analysis, we used CHAMPS Level 2: De-Identified Data, which are all available at the following link after signing a data transfer agreement: https://champshealth.org/data/. BARNARDS data are owned by the Ineos Oxford Institute for Antimicrobial Resistance. They are freely available upon request from the principal investigators of the study. For details, please reach out to Kathryn Thomson at kathryn.thomson@zoo.ox.ac.uk. NeoObs data are owned by the Global Antibiotic Research and Development Partnership and are freely available upon request. For details, please contact Sally Ellis at sellis@gardp.org. The genomes analyzed in this paper are all available through the European Nucleotide Archive under project number PRJEB33565. Individual accession numbers for all genomes analyzed are also available as part of S1 Table.

Funding Statement

This work was supported by the Bill & Melinda Gates Foundation (grant OPP1190803; to RL), the Smith-Newton Environmental Fellowship provided through the High Meadows Environmental Institute at Princeton University (no grant number; to CKK), the US Centers for Disease Control and Prevention (grant 21IPA2113462; to RL), and the US National Science Foundation (grant CCF1918628; to RL). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Liu L, Oza S, Hogan D, Perin J, Rudan I, Lawn JE, et al. Global, regional, and national causes of child mortality in 2000–13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet. 2015;385(9966):430–440. doi: 10.1016/S0140-6736(14)61698-6 [DOI] [PubMed] [Google Scholar]
2.Okomo U, Akpalu EN, Le Doare K, Roca A, Cousens S, Jarde A, et al. Aetiology of invasive bacterial infection and antimicrobial resistance in neonates in sub-Saharan Africa: a systematic review and meta-analysis in line with the STROBE-NI reporting guidelines. Lancet Infect Dis. 2019;19(11):1219–1234. doi: 10.1016/S1473-3099(19)30414-1 [DOI] [PubMed] [Google Scholar]
3.Seale AC, Blencowe H, Manu AA, Nair H, Bahl R, Qazi SA, et al. Estimates of possible severe bacterial infection in neonates in sub-Saharan Africa, south Asia, and Latin America for 2012: a systematic review and meta-analysis. Lancet Infect Dis. 2014;14(8):731–741. doi: 10.1016/S1473-3099(14)70804-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Taylor AW, Blau DM, Bassat Q, Onyango D, Kotloff KL, El Arifeen S, et al. Initial findings from a novel population-based child mortality surveillance approach: a descriptive study. Lancet Glob Health. 2020;8(7):e909–e919. doi: 10.1016/S2214-109X(20)30205-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Group BARNARDS, Sands K Carvalho MJ, Portal E, Thomson K, Dyer C, et al. Characterization of antimicrobial-resistant Gram-negative bacteria that cause neonatal sepsis in seven low- and middle-income countries. Nat Microbiol. 2021. Apr;6(4):512–23. Available from: http://www.nature.com/articles/s41564-021-00870-7. doi: 10.1038/s41564-021-00870-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Chaurasia S, Sivanandan S, Agarwal R, Ellis S, Sharland M, Sankar MJ. Neonatal sepsis in South Asia: huge burden and spiralling antimicrobial resistance. BMJ. 2019. Jan 22;364:k5314. Available from: http://www.bmj.com/content/364/bmj.k5314. doi: 10.1136/bmj.k5314 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Thomson KM, Dyer C, Liu F, Sands K, Portal E, Carvalho MJ, et al. Effects of antibiotic resistance, drug target attainment, bacterial pathogenicity and virulence, and antibiotic access and affordability on outcomes in neonatal sepsis: an international microbiology and drug evaluation prospective substudy (BARNARDS). Lancet Infect Dis. 2021;21(12):1677–1688. doi: 10.1016/S1473-3099(21)00050-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.World Health Organization. Antimicrobial resistance and primary health care. World Health Organization; 2018. [Google Scholar]
9.Bebell LM, Muiru AN. Antibiotic use and emerging resistance: how can resource-limited countries turn the tide? Glob Heart. 2014;9(3):347–358. doi: 10.1016/j.gheart.2014.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Choi M, Tennant SM, Simon R, Cross AS. Progress towards the development of Klebsiella vaccines. Expert Rev Vaccines. 2019;18(7):681–691. doi: 10.1080/14760584.2019.1635460 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Feldman MF, Mayer Bridwell AE, Scott NE, Vinogradov E, McKee SR, Chavez SM, et al. A promising bioconjugate vaccine against hypervirulent Klebsiella pneumoniae. Proc Natl Acad Sci U S A. 2019;116(37):18655–18663. doi: 10.1073/pnas.1907833116 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Campbell WN, Hendrix E, Cryz S Jr, Cross AS. Immunogenicity of a 24-valent Klebsiella capsular polysaccharide vaccine and an eight-valent Pseudomonas O-polysaccharide conjugate vaccine administered to victims of acute trauma. Clin Infect Dis. 1996;23(1):179–181. doi: 10.1093/clinids/23.1.179 [DOI] [PubMed] [Google Scholar]
13.Edelman R, Talor DN, Wasserman SS, McClain JB, Cross AS, Sadoff JC, et al. Phase 1 trial of a 24-valent Klebsiella capsular polysaccharide vaccine and an eight-valent Pseudomonas O-polysaccharide conjugate vaccine administered simultaneously. Vaccine. 1994;12(14):1288–1294. doi: 10.1016/s0264-410x(94)80054-4 [DOI] [PubMed] [Google Scholar]
14.Lam MM, Wick RR, Judd LM, Holt KE, Wyres KL. Kaptive 2.0: updated capsule and lipopolysaccharide locus typing for the Klebsiella pneumoniae species complex. Microb Genomics. 2022;8(3). doi: 10.1099/mgen.0.000800 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Rollenske T, Szijarto V, Lukasiewicz J, Guachalla LM, Stojkovic K, Hartl K, et al. Cross-specificity of protective human antibodies against Klebsiella pneumoniae LPS O-antigen. Nat Immunol. 2018;19(6):617–624. doi: 10.1038/s41590-018-0106-2 [DOI] [PubMed] [Google Scholar]
16.Malachowa N, Kobayashi SD, Porter AR, Freedman B, Hanley PW, Lovaglio J, et al. Vaccine protection against multidrug-resistant Klebsiella pneumoniae in a nonhuman primate model of severe lower respiratory tract infection. MBio. 2019;10(6):e02994–e02919. doi: 10.1128/mBio.02994-19 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Pennini ME, De Marco A, Pelletier M, Bonnell J, Cvitkovic R, Beltramello M, et al. Immune stealth-driven O2 serotype prevalence and potential for therapeutic antibodies against multidrug resistant Klebsiella pneumoniae. Nat Commun. 2017;8(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Ramírez Sevilla C, Gómez Lanza E, Manzanera JL, Martín JAR, Sanz MÁB. Active immunoprophyilaxis with uromune decreases the recurrence of urinary tract infections at three and six months after treatment without relevant secondary effects. BMC Infect Dis. 2019. Oct 28;19(1):901. doi: 10.1186/s12879-019-4541-y [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Szijártó V, Guachalla LM, Hartl K, Varga C, Badarau A, Mirkina I, et al. Endotoxin neutralization by an O-antigen specific monoclonal antibody: a potential novel therapeutic approach against Klebsiella pneumoniae ST258. Virulence. 2017;8(7):1203–1215. doi: 10.1080/21505594.2017.1279778 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Roper MH, Vandelaer JH, Gasse FL. Maternal and neonatal tetanus. Lancet. 2007;370(9603):1947–1959. doi: 10.1016/S0140-6736(07)61261-6 [DOI] [PubMed] [Google Scholar]
21.Lawn JE, Bianchi-Jassir F, Russell NJ, Kohli-Lynch M, Tann CJ, Hall J, et al. Group B Streptococcal Disease Worldwide for Pregnant Women, Stillbirths, and Children: Why, What, and How to Undertake Estimates? Clin Infect Dis. 2017. Nov 6;65(suppl_2):S89–99. doi: 10.1093/cid/cix653 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Zinkernagel RM. Maternal antibodies, childhood infections, and autoimmune diseases. N Engl J Med. 2001;345(18):1331–1335. doi: 10.1056/NEJMra012493 [DOI] [PubMed] [Google Scholar]
23.Lewnard JA, Lo NC, Arinaminpathy N, Frost I, Laxminarayan R. Childhood vaccines and antibiotic use in low- and middle-income countries. Nature. 2020. May 7;581(7806):94–99. Available from: http://www.nature.com/articles/s41586-020-2238-4. doi: 10.1038/s41586-020-2238-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Anderson RM. The concept of herd immunity and the design of community-based immunization programmes. Vaccine. 1992;10(13):928–935. doi: 10.1016/0264-410x(92)90327-g [DOI] [PubMed] [Google Scholar]
25.Obiero CW, Seale AC, Berkley JA. Empiric treatment of neonatal sepsis in developing countries. Pediatr Infect Dis J. 2015;34(6):659. doi: 10.1097/INF.0000000000000692 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Russell NJ, Stohr W, Plakkal N, Cook A, Berkley JA, Adhisivam B, et al. Patterns of antibiotic use, pathogens and clinical outcomes in hospitalised neonates and young infants with sepsis in the NeoOBS global neonatal sepsis observational cohort study. medRxiv. 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Stein M. Large Sample Properties of Simulations Using Latin Hypercube Sampling. Technometrics. 1987. May 1;29(2):143–151. Available from: https://www.tandfonline.com/doi/abs/10.1080/00401706.1987.10488205 [Google Scholar]
28.Murray CJL, Aravkin AY, Zheng P, Abbafati C, Abbas KM, Abbasi-Kangevari M, et al. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020. Oct 17;396(10258):1223–1249. Available from: https://www.sciencedirect.com/science/article/pii/S0140673620307522. doi: 10.1016/S0140-6736(20)30752-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Murray CJ, Ikuta KS, Sharara F, Swetschinski L, Aguilar GR, Gray A, et al. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022. Feb 12;399(10325):629–655. Available from: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(21)02724-0/fulltext. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Vandelaer J, Birmingham M, Gasse F, Kurian M, Shaw C, Garnier S. Tetanus in developing countries: an update on the Maternal and Neonatal Tetanus Elimination Initiative. Vaccine. 2003;21(24):3442–3445. doi: 10.1016/s0264-410x(03)00347-5 [DOI] [PubMed] [Google Scholar]
31.Argimón S, David S, Underwood A, Abrudan M, Wheeler NE, Kekre M, et al. Rapid genomic characterization and global surveillance of Klebsiella using pathogenwatch. Clin Infect Dis. 2021;73(Supplement_4):S325–S335. doi: 10.1093/cid/ciab784 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Wen SCH, Ezure Y, Rolley L, Spurling G, Lau CL, Riaz S, et al. Gram-negative neonatal sepsis in low- and lower-middle-income countries and WHO empirical antibiotic recommendations: A systematic review and meta-analysis. PLoS Med. 2021. Sep 28;18(9):e1003787. Available from: https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1003787 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Fu H, Lewnard JA, Frost I, Laxminarayan R, Arinaminpathy N. Modelling the global burden of drug-resistant tuberculosis avertable by a post-exposure vaccine. Nat Commun. 2021. Jan 18;12(1):424. Available from: http://www.nature.com/articles/s41467-020-20731-x. doi: 10.1038/s41467-020-20731-x [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Birger R, Antillón M, Bilcke J, Dolecek C, Dougan G, Pollard AJ, et al. Estimating the effect of vaccination on antimicrobial-resistant typhoid fever in 73 countries supported by Gavi: a mathematical modelling study. Lancet Infect Dis. 2022. Feb 3. Available from: https://www.sciencedirect.com/science/article/pii/S1473309921006277. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Mashau RC, Meiring ST, Dramowski A, Magobo RE, Quan VC, Perovic O, et al. Culture-confirmed neonatal bloodstream infections and meningitis in South Africa, 2014–19: a cross-sectional study. Lancet Glob Health. 2022. Aug;10(8):e1170–e1178. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2214109X22002467. doi: 10.1016/S2214-109X(22)00246-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Chaurasia S, Sankar M, Agarwal R, Yadav C, Arya S. Investigators of the Delhi Neonatal Infection Study (DeNIS) collaboration. Characterization and antimicrobial resistance of sepsis pathogens in neonates born in tertiary care centres in Delhi, India: a cohort study. Lancet Glob Health. 2016;4(10):e752–e760. [DOI] [PubMed] [Google Scholar]
37.Zea-Vera A, Ochoa TJ. Challenges in the diagnosis and management of neonatal sepsis. J Trop Pediatr. 2015. Feb 1;61(1):1–13. Available from: https://academic.oup.com/tropej/article-lookup/doi/10.1093/tropej/fmu079 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Morgan R, Ayiasi RM, Barman D, Buzuzi S, Ssemugabo C, Ezumah N, et al. Gendered health systems: evidence from low-and middle-income countries. Health Res Policy Syst. 2018;16(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS Med. doi: 10.1371/journal.pmed.1004239.r001

Decision Letter 0

Philippa Dodd

14 Oct 2022

Dear Dr Laxminarayan,

Thank you for submitting your manuscript entitled "Global, regional, and national estimates of the impact of a maternal Klebsiella pneumoniae vaccine: A Bayesian modeling analysis" for consideration by PLOS Medicine.

Your manuscript has now been evaluated by the PLOS Medicine editorial staff as well as by an academic editor with relevant expertise and I am writing to let you know that we would like to send your submission out for external peer review.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.

Please re-submit your manuscript within two working days, i.e. by Oct 18 2022 11:59PM.

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed all checks it will be sent out for review.

Feel free to email us at plosmedicine@plos.org if you have any queries relating to your submission.

Kind regards,

Philippa Dodd, MBBS MRCP PhD

Editor

PLOS Medicine

PLoS Med. doi: 10.1371/journal.pmed.1004239.r002

Decision Letter 1

Philippa Dodd

26 Jan 2023

Dear Dr. Laxminarayan,

Thank you very much for submitting your manuscript "Global, regional, and national estimates of the impact of a maternal Klebsiella pneumoniae vaccine: A Bayesian modeling analysis" (PMEDICINE-D-22-03384R1) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Feb 16 2023 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Philippa Dodd, MBBS MRCP PhD

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

GENERAL

Please respond to all editor and reviewer requests detailed below, in full.

Please include line numbers starting at 1 on the title page and in continuous sequence thereafter

COMPETING INTERESTS STATEMENT

“The sponsor of the study provided input in the writing of the report, acting in a purely scientific capacity.” Please elaborate further on the precise role that the funder played in writing the manuscript. What does “purely scientific capacity” encompass? Please include details of each individual affiliated to the funding body that contributed to manuscript and provide details of each individual’s specific role in the preparation of the manuscript.

DATA AVAILABILITY STATEMENT

Thank you for agreeing to make your data available upon request. The Data Availability Statement (DAS) requires revision. For each data source used in your study:

a) If the data are freely or publicly available, note this and state the location of the data: within the paper, in Supporting Information files, or in a public repository (include the DOI or accession number).

b) If the data are owned by a third party but freely available upon request, please note this and state the owner of the data set and contact information for data requests (web or email address). Note that a study author cannot be the contact person for the data.

c) If the data are not freely available, please describe briefly the ethical, legal, or contractual restriction that prevents you from sharing it. Please also include an appropriate contact (web or email address) for inquiries (again, this cannot be a study author).

RESEARCH IN CONTEXT

Please remove this section and all sub-sections from the manuscript

ABSTRACT

Please structure your abstract using the PLOS Medicine headings (Background, Methods and Findings, Conclusions).

Please combine the Methods and Findings sections into one section, “Methods and findings”.

Abstract Background:

Please ensure that the final sentence clearly states the study question.

“…resistant to antibiotic treatment regimens…” suggest introducing carbapenem resistance here at the outset, rather than later

Please define “K. pneumoniae” after the full name in the lines above, prior to first use here

Abstract Methods and Findings:

“We used data from three global studies…” It would be helpful to include further details here - which countries did those studies include, how many participants, over what time frame (s) how many had K. pneumoniae and so on

“We quantified the rate of acquisition of antimicrobial resistance genes in K. pneumoniae isolates…” it would be helpful to include details of how you did this

“…predict the future burden…” how do you define “burden”?

Please quantify the main results with p values, as well as 95% CrIs. When reporting p values please report as p<0.001 and where higher the exact p-value as p=0.002, for example. If not including p-values, not for the purpose of transparent data reporting please provide reasons as to why not.

Please ensure that all numbers presented in the abstract are present and identical to numbers presented in the main manuscript text.

Please provide the actual numbers of events for the outcomes, not just summary statistics/percentages

Did you adjust your analyses for any variable factors? If so, please also detail these

In the last sentence of the Abstract Methods and Findings section, please describe the main limitation(s) of the study's methodology.

AUTHOR SUMMARY

At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

METHODS and RESULTS

We ask the following of all authors of modelling studies. Please see below and in your rebuttal, please sign post to where in the manuscript the relevant information can be located:

• Please provide a diagram that shows the model structure, including how the disease natural history is represented, the process and determinants of disease acquisition, and how the putative intervention could affect the system.

• Please provide a complete list of model parameters, including clear and precise descriptions of [the meaning of each parameter, together with the values or ranges for each, with justification or the primary source cited, and important caveats about the use of these values noted].

• Please provide a clear statement about how the model was fitted to the data [including goodness-of-fit measure, the numerical algorithm used, which parameter varied, constraints imposed on parameter values, and starting conditions].

• For uncertainty analyses, please state the sources of uncertainties quantified and not quantified [can include parameter, data, and model structure].

• Please provide sensitivity analyses to identify which parameter values are most important in the model. Uncertainty estimates seek to derive a range of credible results on the basis of an exploration of the range of reasonable parameter values. The choice of method should be presented and justified.

• Please discuss the scientific rationale for this choice of model structure and identify points where this choice could influence conclusions drawn. Please also describe the strength of the scientific basis underlying the key model assumptions.

We agree with the reviewer (please see below) that additional information regarding the datasets used would be helpful. Please revise accordingly, as for the abstract.

Please remove role of the funding source from the end of the methods section and include only in the manuscript submission form

Please ensure that results are quantified with p-values as well as 95% CrIs. When reporting p-values please report as p<0.001 and where higher, the exact p-value as p=0.002, for example. If not including p-values, not for the purpose of transparent data reporting please provide reasons as to why.

Please refrain from discussing your results in the results section (see page 11, second paragraph) and report only your findings. Please move any discussion of your results to the appropriate section of your discussion

Please ensure all abbreviations are clearly defined at first use – ARGs and CRGs, for example on page 13 – I couldn’t see a previous definition but my apologies if I missed it

FIGURES

All the figures are very small and therefore not easily interpretable by the reader. Please see below for specific comments. Suggest considering placing a title above each image in the figures to help the reader but we leave this to your discretion.

Please ensure that each figure is affiliated to an appropriate caption which clearly describes its content without the need to refer to the text. Please ensure that any and all abbreviations are defined within the appropriate legend/caption

Please consider avoiding the use of green or red to make figures more accessible to those with colour blindness

Figure 1: we agree with the reviewers that the figure is not very accessible, text is too small and the presentation of different axes is a bit confusing. Please show the axis beginning at zero. If this is not possible, please show a break in the axis.

In the figure caption please clearly state the meaning of the bars and whiskers and please define GBD and CHERG

Figure 2: all the text in the figures in too small for the reader, including the axis labels. The error bars are very distracting and too large, they should not meet each other. The disparity in bar sizes makes it very difficult to read the data presented here

Figure 3: As above – these are also very small and as such it is difficult to appreciate the different colour densities which we like. It may also help the reader to title each of the individual maps within the figure. Please revise accordingly. Please confirm that the appropriate usage rights apply to the use of the maps. Please see our guidelines for map images: https://journals.plos.org/plosmedicine/s/figures#loc-maps

Figure 4: as above, please also define abbreviations - AMR, ARGs, CRGs. The legend is hard to read and could be overlayed on the graph and enlarges, for example.

SUPPORTING FILES

Please ensure that all figures/tables are affiliated to an appropriate caption which clearly describes their contents. Suggest considering applying some universal formatting to the figures throughout the main manuscript and the supporting files. A number of bar graphs are presented but each are very different, please revise accordingly

Kp Appendix – please ensure abbreviations are defined throughout

Fig S1 – please ensure colours are accessible to those with colour blindness

Fig S3 – please define abbreviations

Fig S5 – is difficult to interpret without a legend for the colours, please include

Fig S7 onwards – captions for the maps insufficiently describes what they show. What is meant by lower and upper bound, for example – you can work it out, but it’s not obvious to the reader. Do these bounds need to presented as maps? Perhaps a table would be preferable here?

What are the items listed as 1-9 above where you refer to a CSV file for the “complete table”. Please clarify

Fig S13-S16 - please define the meaning of the bars and whiskers and define the abbreviations GBD and CHERG

Fig S17-20 – see earlier re: size of error bars, colours, uniformity, definitions of all abbreviations

** We note that there are a lot of figures presented here and (as above) we suggest that much of this data would be more accessible if presented as tables **

DISCUSSION

Please remove the declaration of interests and data sharing statements form the end of the discussion – these should be included in the relevant parts of the manuscripts only – they will be complied as meta-data

REFERENCES

Please ensure that your referencing follows PLOS Medicine’s style. Further details can be found here: https://journals.plos.org/plosmedicine/s/submission-guidelines#loc-references

In the bibliography please ensure that you list up to more but no more than 6 author names followed by et al where more than 6 authors contribute. Please ensure that journal name abbreviations are those found in the National Center for Biotechnology Information (NCBI) databases.

Comments from the reviewers:

Reviewer #1: Overall comment: In general, this is a useful manuscript that brings into focus an important issue, ie how impactful a vaccine against K pneumoniae might be for preventing neonatal sepsis. One major issue I have is that the data sources and assumptions and initial values that go into the model are not described well, which gives the impression that the results are generated from a "black box." I suggest adding a "Table 1" that clearly spells out model parameters. Less importantly, there are some style issue with the manuscript that I suggest cleaning up, in particular the inclusion of methods and discussion points in the results section; including discussion points in the results section is problematic, as it blurs the line between what the model is directly showing and authors' opinions or conclusions.

Abstract: Fix the introduction paragraph to clarify that the focus is on neonatal sepsis, not sepsis generally, and that data sources were primarily from LMIC.

Page 6, data section: CHAMPS is Child Health and Mortality Prevention Surveillance (ie, not Childhood…). Also, the authors should include a sentence or two about each data source so that readers do not need to hunt to find basic information (e.g., countries involved, who was included, method of enrollment, years, etc).

Methods, general comment: One tricky issue with maternal immunization as a method to prevent neonatal infections is that babies born preterm are often more susceptible to sepsis and may need to stay in hospital for medical support until they are large enough to go home, which results in exposures to hospital-associated pathogens, including K pneumoniae. Conversely, optimal transfer of antibodies from mom to baby across the placenta occurs late in pregnancy (3rd trimester). Did you account for this issue in the model? I was curious why you selected 70% efficacy, and whether this figure accounted for less benefit in preterm births or imperfect coverage of serotypes or just assumed that efficacy of the vaccine against sepsis caused by vaccine serotypes would be about 70% in term infants. I would include this information in the methods section.

Results, bottom of page 11. Here the authors have included text that should be in the discussion section, as it seems to include conclusions drawn from the data. For example, this is the wording:

" Nevertheless, we do not find overall trends in isolates from neonates who succumbed to sepsis

having higher rates of AMR (Fig. 2A, 2B), resistance to ampicillin or gentamicin (Fig. 2C, 2D)

or higher multi-drug resistance (MDR) (Fig. 2E, 2F), indicating that antibiotic resistance of the

disease-causing K. pneumoniae alone may not significantly impact neonatal mortality. Instead,

resistance is high in all isolates, against all drugs, suggesting that other factors such as access to

effective drugs and antibiotic dosing regiments may be driving differential mortality."

I'm trying to understand how the authors came to this conclusion, as the isolates that were collected from living babies with sepsis may have indeed caused their death (unclear if the study followed them until the end of their illness) or perhaps there was some other issue (e.g. contamination). Also, since the isolates from deceased children and living children came from two different studies with different patient populations and treatment settings, I would encourage caution in this interpretation. Also, what antibiotics each child received is likely unknown.

Page 12, 5th line under "Projected Global…" heading: typo with the comma placement in the last number in this line.

Page 12, 3rd line from the bottom: How confident do you feel about extrapolating to North America and Europe from the data sets that included little to no data from high income settings? Should those settings be excluded, or more data sources added to better support numbers from high income settings?

Page 16, discussion: In the limitations paragraph, the authors mention that the data included did reflect entire countries, but the data sources only included data from a limited number of countries overall. Isn't the complete lack of data from many countries also an issue that should be mentioned? For example, were Europe and North America extrapolated from Greece and Brazil, respectively?

Figures, general comment: The pictures are quite small relative to the text, so I find myself having to zoom in to interpret the figure then zoom out to read the information about the figure. Perhaps the journal has a way to fix this mismatch, or maybe the labeling on the figures could be larger.

Figure 1. Is there something problematic with the way the model is handling number of deaths for Greece? The B figure suggests the vaccine is causing deaths, although with the log scale and no zero it's difficult to tell, and the figure with percentages suggests a small percent of deaths are prevented.

Reviewer #2: SUMMARY

This manuscript reports estimates of the global number of neonatal deaths due to Klebsiella pneumoniae, and the number of these deaths that could be averted by a vaccine with 70% efficacy. In general the approach seems reasonable, though more detail is required on some sections, particularly on how the location-specific estimates were extrapolated to generate regional and global estimates. Specific comments are provided below.

MAJOR ISSUES

* Pages 6-7: the analytic approach appears reasonable if one assumes the fraction of sepsis deaths is homogeneous within the sampled locations, or if the data represent a representative sample of sepsis deaths within each location. It would be useful to examine these assumptions - in particular, where a particular location has several studies, how consistent are these results? If inconsistent, might one explicitly model study-level variation (would lead to greater uncertainty in estimated results, which would be appropriate if there were substantial study-level variation). I think ideally study-level variation would be examined and modelled (if there are enough locations with multiple studies), otherwise this could also be discussed in the limitations.

* Page 7: for locations with several studies, it is not clear to me why the data were not pooled in a single beta distribution. The approaches will give different answers, though perhaps trivially so, as the current approach adds a value of 1.0 to both beta parameters for each study, via the beta(1,1) prior. I can see the benefit of estimating studies results separately if there was a desire to model study-level variation (I think a beta-binomial model could do this?), yet this is not done (or at least is not described), per comment above.

* Page 8: the regression approach used to extrapolate to other countries is not described in sufficient detail. Given how important this step is to the analysis (going from 18 countries to global), it would be useful to provide additional information on covariates, functional form, evidence of model fit, etc.

MINOR ISSUES

* Page 2, abstract: If the word count allows, it would be useful to provide more specific details on the scenarios that produced the vaccine impact estimates (e.g., is this vaccination of all pregnant women globally, or some subset?)

* Page 2, abstract: is it possible to provide some measure of uncertainty around the results presented?

* Page 2, abstract: small point, but it is unclear to me what "The most significant benefits" means. Is this to be read as "largest relative benefit"? Perhaps more specific language would be useful.

* Page 6: it would be useful to expand the section describing the data, in particular to include the number of samples and distribution across regions.

* Page 6 (data): please provide some comment on possible bias that comes with restricting analyses to culture-confirmed cases (and what fraction of cases this generally represents). I realize that using culture-confirmed cases may be the only feasible approach, but still useful to know if there are any biases could result, and how they are dealt with.

* Page 7: around "effective coverage level equal to that of the maternal tetanus vaccine", it would be useful to give the median and IQR (or something similar) to give the reader some idea of the average level and variation in vaccine coverage assumed for the analysis.

* Page 7: I might describe what you used the LHS sample for, for readers who are unfamiliar.

* Page 10: "The average rate of resistance to gentamicin, which is usually prescribed in combination with ampicillin as the first- line treatment for sepsis per current WHO guidelines,32 is 57.22% [CrI: 31.540-80.42]" - I am a bit surprised by how broad the intervals are. Is this because of a small number of observations (i.e. individuals), or the extrapolation process, or something else? On a similar subject, there is substantial uncertainty around Panel A in Figure 1, but not Panels B and C. I realize that B and C are log-scale, but it appears Panel A still has more relative uncertainty (take Thailand for example where in Panel A the interval ranges from ~0 to >3 times the point estimate). Is everything here correct?

Reviewer #3: According to the authors, there has been no study projecting the effects of K. pneumoniae vaccine distribution. Given that there are numerous efforts in developing a viable K. pneumoniae vaccine, the authors tried to estimate the global, regional, and national impact of a hypothetical vaccine with 70% efficacy on neonatal sepsis infections and mortality using a Bayesian mixture-modeling framework. The estimated impact was presented in terms of number of neonatal sepsis deaths and neonatal sepsis cases averted due to the hypothetical vaccine. In addition, the authors also examined the AMR of sepsis-causing isolates and tried to project the future benefits of maternal K. pneumoniae vaccination. Overall, this is a well-designed and conducted study and the results of the study are of relevance, particularly for LMICs in Africa. Below are my specific comments.

1. Please add line numbers for easier reference.

2. The sponsor of the study provided input in the writing of the report, which might result in conflict of interest.

3. Data used in the study are not fully available.

4. Page 7, method: "We extended this analysis to estimate the absolute number.. we report values derived using CHERG data." It seems that sometimes you used CHERG data and sometimes you used GBD data to derives the parameter values. However, are data from CHERG and from GBD comparable to each other? Did you do any adjustment to the GBD data to make them comparable to the CHERG data? I did not see you mention this in the method section.

5. Page 7, method: "We attempted to minimize the impact of potential.. to estimate the CFR". I don't understand how Bayesian method can help minimize the bias here. Loss to follow-ups are usually informative. I am not sure whether Bayesian method can help reduce the bias due to informative LTFUs.

6. Method: To estimate the global, regional, and national number of cases and deaths, why did you not use some hierarchical modeling approach? Bayesian framework should be perfect to incorporate hierarchical data structure, which can make the estimates between different levels internally consistent. However, I did not see the hierarchical structure in the method section.

Any attachments provided with reviews can be seen via the following link:

[LINK]

PLoS Med. 2023 May 22;20(5):e1004239. doi: 10.1371/journal.pmed.1004239.r003

Author response to Decision Letter 1

18 Feb 2023

Attachment

Submitted filename: ResponseToReviewers-R1-02152023.docx

Click here for additional data file.^{(39.9KB, docx)}

PLoS Med. doi: 10.1371/journal.pmed.1004239.r004

Decision Letter 2

Philippa Dodd

17 Mar 2023

Dear Dr. Laxminarayan,

Thank you very much for re-submitting your manuscript "Global, regional, and national estimates of the impact of a maternal Klebsiella pneumoniae vaccine: A Bayesian modeling analysis" (PMEDICINE-D-22-03384R2) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by 3 reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosmedicine@plos.org.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Mar 24 2023 11:59PM.

Sincerely,

Philippa Dodd, MBBS MRCP PhD

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

GENERAL

Thank you for your very detailed and considered responses to previous editor and reviewer requests. Please see below for further comments/suggestions that we request you address in full.

ABSTRACT

Line 17 – please replace the sub-heading “summary” with “Abstract”

AUTHOR SUMMARY

Thank you for including an author summary some suggestions are detailed below

Line 69: suggest “pre-clinical model” perhaps instead of “animal”

Line 74: To improve accessibility to the reader, suggest the following - “Using a Bayesian mixture-model based on clinical surveillance of neonatal sepsis, we present country-specific estimates for the number of deaths and cases of antimicrobial resistant neonatal sepsis, caused by K. pneumoniae, that would be averted if a vaccine with 70% efficacy was to be given to pregnant mothers.”

Line 82: suggest moving this point to precede that at line 79

Line 85: suggest rewording this statement for improved clarity – “…countries of interest…” is rather vague, perhaps “especially in LMICs” instead, perhaps? And, “…due to all bacteria…” this is a little confusing/vague – do you mean bacteria other than K pneumoniae or all strains of K pneumoniae?

Line 88: Suggest “This would help to achieve targets set by […] for improved global child health and to mitigate against inequity in neonatal survival in low- and middle-income settings compared to high-income settings. Suggest that the blank space in brackets is completed (WHO targets?)

Line 90: Suggest “Reducing cases of neonatal sepsis by vaccination could also contribute to reduced antibiotic use, subsequent improvements in antimicrobial resistance rates and a reduction in healthcare utilization and expenditure.” Or something similar

METHODS and RESULTS

Please see statistical reviewer (reviewer #3) comments below

COMPETING INTERESTS STATEMENT

Line 492: Thank you for updating your statement. Please remove from the end of the discussion in the main manuscript and include only in the manuscript submission form when you re-submit your manuscript.

FIGURES

Figure 4: in the caption you refer to ARGs (and define them) as well as CRGs (and define these) but CRGs are not detailed anywhere in the legend of figure 4 that is available to me. Please revise as necessary.

SOCIAL MEDIA

To help us extend the reach of your research, if not already done so, please provide any Twitter handle(s) that would be appropriate to tag, including your own, your coauthors’, your institution, funder, or lab. Please detail any handles you wish to be included when we tweet this paper, in the manuscript submission form when you re-submit the manuscript.

Comments from Reviewers:

Reviewer #1: The authors appear to have addressed my concerns. I would ask for one simple correction -- CHAMPS does not do autopsies but rather Minimally Invasive Tissue Sampling (MITS).

Reviewer #2: Thank you, these revisions resolve my concerns.

Reviewer #3: I appreciate the authors' responses to my comments. I am happy with most of the responses but I cannot agree with their response to my comment #5. I don't think using a uniform prior can help reduce bias due to lost-to-follow up or any bias.. A uniform prior just means that the analyst knows little/nothing about the quantity of interest and rely almost entirely on the data to inform the estimate. It dose not help with reducing bias at all, and increasing the uncertainty of the estimate does not mean reducing bias either. It just means that you are more uncertain about the estimate.

Any attachments provided with reviews can be seen via the following link:

[LINK]

PLoS Med. 2023 May 22;20(5):e1004239. doi: 10.1371/journal.pmed.1004239.r005

Author response to Decision Letter 2

20 Mar 2023

Attachment

Submitted filename: ResponseToReviewers-Final-03202023.docx

Click here for additional data file.^{(8.7KB, docx)}

PLoS Med. doi: 10.1371/journal.pmed.1004239.r006

Decision Letter 3

Philippa Dodd

26 Apr 2023

Dear Dr Laxminarayan,

On behalf of my colleagues and the Academic Editor, Dr. Rebecca Freeman-Grais, I am pleased to inform you that we have agreed to publish your manuscript "Global, regional, and national estimates of the impact of a maternal Klebsiella pneumoniae vaccine: A Bayesian modeling analysis" (PMEDICINE-D-22-03384R3) in PLOS Medicine.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Once you have received these formatting requests, please note that your manuscript will not be scheduled for publication until you have made the required changes.

In the meantime, please log into Editorial Manager at http://www.editorialmanager.com/pmedicine/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production process.

PRESS

We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with medicinepress@plos.org. If you have not yet opted out of the early version process, we ask that you notify us immediately of any press plans so that we may do so on your behalf.

We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit http://www.plos.org/about/media-inquiries/embargo-policy/.

Thank you again for submitting to PLOS Medicine. We look forward to publishing your paper.

Best wishes,

Pippa

Philippa Dodd, MBBS MRCP PhD

PLOS Medicine

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Text. Supplementary materials.

(PDF)

Click here for additional data file.^{(4.4MB, pdf)}

S2 Text. Extended methods.

Derivation of the mixture model posterior distribution (Section 1) and details of antimicrobial susceptibility testing done (Section 2).

(PDF)

Click here for additional data file.^{(273.2KB, pdf)}

S1 Table. Details of the analyzed genomes used in this study including FASTA file ID, year, source, country of origin, and number of identified antimicrobial resistance genes.

(XLSX)

Click here for additional data file.^{(797.3KB, xlsx)}

S2 Table. Impact of maternal vaccination by country.

Country-specific estimates of overall averted cases, deaths, and result on neonatal mortality as well as antibiotic resistant deaths.

(XLSX)

Click here for additional data file.^{(148.9KB, xlsx)}

Attachment

Submitted filename: ResponseToReviewers-R1-02152023.docx

Click here for additional data file.^{(39.9KB, docx)}

Attachment

Submitted filename: ResponseToReviewers-Final-03202023.docx

Click here for additional data file.^{(8.7KB, docx)}

Data Availability Statement

[pmed.1004239.ref001] 1.Liu L, Oza S, Hogan D, Perin J, Rudan I, Lawn JE, et al. Global, regional, and national causes of child mortality in 2000–13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet. 2015;385(9966):430–440. doi: 10.1016/S0140-6736(14)61698-6 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref002] 2.Okomo U, Akpalu EN, Le Doare K, Roca A, Cousens S, Jarde A, et al. Aetiology of invasive bacterial infection and antimicrobial resistance in neonates in sub-Saharan Africa: a systematic review and meta-analysis in line with the STROBE-NI reporting guidelines. Lancet Infect Dis. 2019;19(11):1219–1234. doi: 10.1016/S1473-3099(19)30414-1 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref003] 3.Seale AC, Blencowe H, Manu AA, Nair H, Bahl R, Qazi SA, et al. Estimates of possible severe bacterial infection in neonates in sub-Saharan Africa, south Asia, and Latin America for 2012: a systematic review and meta-analysis. Lancet Infect Dis. 2014;14(8):731–741. doi: 10.1016/S1473-3099(14)70804-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref004] 4.Taylor AW, Blau DM, Bassat Q, Onyango D, Kotloff KL, El Arifeen S, et al. Initial findings from a novel population-based child mortality surveillance approach: a descriptive study. Lancet Glob Health. 2020;8(7):e909–e919. doi: 10.1016/S2214-109X(20)30205-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref005] 5.Group BARNARDS, Sands K Carvalho MJ, Portal E, Thomson K, Dyer C, et al. Characterization of antimicrobial-resistant Gram-negative bacteria that cause neonatal sepsis in seven low- and middle-income countries. Nat Microbiol. 2021. Apr;6(4):512–23. Available from: http://www.nature.com/articles/s41564-021-00870-7. doi: 10.1038/s41564-021-00870-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref006] 6.Chaurasia S, Sivanandan S, Agarwal R, Ellis S, Sharland M, Sankar MJ. Neonatal sepsis in South Asia: huge burden and spiralling antimicrobial resistance. BMJ. 2019. Jan 22;364:k5314. Available from: http://www.bmj.com/content/364/bmj.k5314. doi: 10.1136/bmj.k5314 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref007] 7.Thomson KM, Dyer C, Liu F, Sands K, Portal E, Carvalho MJ, et al. Effects of antibiotic resistance, drug target attainment, bacterial pathogenicity and virulence, and antibiotic access and affordability on outcomes in neonatal sepsis: an international microbiology and drug evaluation prospective substudy (BARNARDS). Lancet Infect Dis. 2021;21(12):1677–1688. doi: 10.1016/S1473-3099(21)00050-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref008] 8.World Health Organization. Antimicrobial resistance and primary health care. World Health Organization; 2018. [Google Scholar]

[pmed.1004239.ref009] 9.Bebell LM, Muiru AN. Antibiotic use and emerging resistance: how can resource-limited countries turn the tide? Glob Heart. 2014;9(3):347–358. doi: 10.1016/j.gheart.2014.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref010] 10.Choi M, Tennant SM, Simon R, Cross AS. Progress towards the development of Klebsiella vaccines. Expert Rev Vaccines. 2019;18(7):681–691. doi: 10.1080/14760584.2019.1635460 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref011] 11.Feldman MF, Mayer Bridwell AE, Scott NE, Vinogradov E, McKee SR, Chavez SM, et al. A promising bioconjugate vaccine against hypervirulent Klebsiella pneumoniae. Proc Natl Acad Sci U S A. 2019;116(37):18655–18663. doi: 10.1073/pnas.1907833116 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref012] 12.Campbell WN, Hendrix E, Cryz S Jr, Cross AS. Immunogenicity of a 24-valent Klebsiella capsular polysaccharide vaccine and an eight-valent Pseudomonas O-polysaccharide conjugate vaccine administered to victims of acute trauma. Clin Infect Dis. 1996;23(1):179–181. doi: 10.1093/clinids/23.1.179 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref013] 13.Edelman R, Talor DN, Wasserman SS, McClain JB, Cross AS, Sadoff JC, et al. Phase 1 trial of a 24-valent Klebsiella capsular polysaccharide vaccine and an eight-valent Pseudomonas O-polysaccharide conjugate vaccine administered simultaneously. Vaccine. 1994;12(14):1288–1294. doi: 10.1016/s0264-410x(94)80054-4 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref014] 14.Lam MM, Wick RR, Judd LM, Holt KE, Wyres KL. Kaptive 2.0: updated capsule and lipopolysaccharide locus typing for the Klebsiella pneumoniae species complex. Microb Genomics. 2022;8(3). doi: 10.1099/mgen.0.000800 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref015] 15.Rollenske T, Szijarto V, Lukasiewicz J, Guachalla LM, Stojkovic K, Hartl K, et al. Cross-specificity of protective human antibodies against Klebsiella pneumoniae LPS O-antigen. Nat Immunol. 2018;19(6):617–624. doi: 10.1038/s41590-018-0106-2 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref016] 16.Malachowa N, Kobayashi SD, Porter AR, Freedman B, Hanley PW, Lovaglio J, et al. Vaccine protection against multidrug-resistant Klebsiella pneumoniae in a nonhuman primate model of severe lower respiratory tract infection. MBio. 2019;10(6):e02994–e02919. doi: 10.1128/mBio.02994-19 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref017] 17.Pennini ME, De Marco A, Pelletier M, Bonnell J, Cvitkovic R, Beltramello M, et al. Immune stealth-driven O2 serotype prevalence and potential for therapeutic antibodies against multidrug resistant Klebsiella pneumoniae. Nat Commun. 2017;8(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref018] 18.Ramírez Sevilla C, Gómez Lanza E, Manzanera JL, Martín JAR, Sanz MÁB. Active immunoprophyilaxis with uromune decreases the recurrence of urinary tract infections at three and six months after treatment without relevant secondary effects. BMC Infect Dis. 2019. Oct 28;19(1):901. doi: 10.1186/s12879-019-4541-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref019] 19.Szijártó V, Guachalla LM, Hartl K, Varga C, Badarau A, Mirkina I, et al. Endotoxin neutralization by an O-antigen specific monoclonal antibody: a potential novel therapeutic approach against Klebsiella pneumoniae ST258. Virulence. 2017;8(7):1203–1215. doi: 10.1080/21505594.2017.1279778 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref020] 20.Roper MH, Vandelaer JH, Gasse FL. Maternal and neonatal tetanus. Lancet. 2007;370(9603):1947–1959. doi: 10.1016/S0140-6736(07)61261-6 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref021] 21.Lawn JE, Bianchi-Jassir F, Russell NJ, Kohli-Lynch M, Tann CJ, Hall J, et al. Group B Streptococcal Disease Worldwide for Pregnant Women, Stillbirths, and Children: Why, What, and How to Undertake Estimates? Clin Infect Dis. 2017. Nov 6;65(suppl_2):S89–99. doi: 10.1093/cid/cix653 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref022] 22.Zinkernagel RM. Maternal antibodies, childhood infections, and autoimmune diseases. N Engl J Med. 2001;345(18):1331–1335. doi: 10.1056/NEJMra012493 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref023] 23.Lewnard JA, Lo NC, Arinaminpathy N, Frost I, Laxminarayan R. Childhood vaccines and antibiotic use in low- and middle-income countries. Nature. 2020. May 7;581(7806):94–99. Available from: http://www.nature.com/articles/s41586-020-2238-4. doi: 10.1038/s41586-020-2238-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref024] 24.Anderson RM. The concept of herd immunity and the design of community-based immunization programmes. Vaccine. 1992;10(13):928–935. doi: 10.1016/0264-410x(92)90327-g [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref025] 25.Obiero CW, Seale AC, Berkley JA. Empiric treatment of neonatal sepsis in developing countries. Pediatr Infect Dis J. 2015;34(6):659. doi: 10.1097/INF.0000000000000692 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref026] 26.Russell NJ, Stohr W, Plakkal N, Cook A, Berkley JA, Adhisivam B, et al. Patterns of antibiotic use, pathogens and clinical outcomes in hospitalised neonates and young infants with sepsis in the NeoOBS global neonatal sepsis observational cohort study. medRxiv. 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref027] 27.Stein M. Large Sample Properties of Simulations Using Latin Hypercube Sampling. Technometrics. 1987. May 1;29(2):143–151. Available from: https://www.tandfonline.com/doi/abs/10.1080/00401706.1987.10488205 [Google Scholar]

[pmed.1004239.ref028] 28.Murray CJL, Aravkin AY, Zheng P, Abbafati C, Abbas KM, Abbasi-Kangevari M, et al. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020. Oct 17;396(10258):1223–1249. Available from: https://www.sciencedirect.com/science/article/pii/S0140673620307522. doi: 10.1016/S0140-6736(20)30752-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref029] 29.Murray CJ, Ikuta KS, Sharara F, Swetschinski L, Aguilar GR, Gray A, et al. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022. Feb 12;399(10325):629–655. Available from: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(21)02724-0/fulltext. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref030] 30.Vandelaer J, Birmingham M, Gasse F, Kurian M, Shaw C, Garnier S. Tetanus in developing countries: an update on the Maternal and Neonatal Tetanus Elimination Initiative. Vaccine. 2003;21(24):3442–3445. doi: 10.1016/s0264-410x(03)00347-5 [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref031] 31.Argimón S, David S, Underwood A, Abrudan M, Wheeler NE, Kekre M, et al. Rapid genomic characterization and global surveillance of Klebsiella using pathogenwatch. Clin Infect Dis. 2021;73(Supplement_4):S325–S335. doi: 10.1093/cid/ciab784 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref032] 32.Wen SCH, Ezure Y, Rolley L, Spurling G, Lau CL, Riaz S, et al. Gram-negative neonatal sepsis in low- and lower-middle-income countries and WHO empirical antibiotic recommendations: A systematic review and meta-analysis. PLoS Med. 2021. Sep 28;18(9):e1003787. Available from: https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1003787 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref033] 33.Fu H, Lewnard JA, Frost I, Laxminarayan R, Arinaminpathy N. Modelling the global burden of drug-resistant tuberculosis avertable by a post-exposure vaccine. Nat Commun. 2021. Jan 18;12(1):424. Available from: http://www.nature.com/articles/s41467-020-20731-x. doi: 10.1038/s41467-020-20731-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref034] 34.Birger R, Antillón M, Bilcke J, Dolecek C, Dougan G, Pollard AJ, et al. Estimating the effect of vaccination on antimicrobial-resistant typhoid fever in 73 countries supported by Gavi: a mathematical modelling study. Lancet Infect Dis. 2022. Feb 3. Available from: https://www.sciencedirect.com/science/article/pii/S1473309921006277. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref035] 35.Mashau RC, Meiring ST, Dramowski A, Magobo RE, Quan VC, Perovic O, et al. Culture-confirmed neonatal bloodstream infections and meningitis in South Africa, 2014–19: a cross-sectional study. Lancet Glob Health. 2022. Aug;10(8):e1170–e1178. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2214109X22002467. doi: 10.1016/S2214-109X(22)00246-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref036] 36.Chaurasia S, Sankar M, Agarwal R, Yadav C, Arya S. Investigators of the Delhi Neonatal Infection Study (DeNIS) collaboration. Characterization and antimicrobial resistance of sepsis pathogens in neonates born in tertiary care centres in Delhi, India: a cohort study. Lancet Glob Health. 2016;4(10):e752–e760. [DOI] [PubMed] [Google Scholar]

[pmed.1004239.ref037] 37.Zea-Vera A, Ochoa TJ. Challenges in the diagnosis and management of neonatal sepsis. J Trop Pediatr. 2015. Feb 1;61(1):1–13. Available from: https://academic.oup.com/tropej/article-lookup/doi/10.1093/tropej/fmu079 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pmed.1004239.ref038] 38.Morgan R, Ayiasi RM, Barman D, Buzuzi S, Ssemugabo C, Ezumah N, et al. Gendered health systems: evidence from low-and middle-income countries. Health Res Policy Syst. 2018;16(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Global, regional, and national estimates of the impact of a maternal Klebsiella pneumoniae vaccine: A Bayesian modeling analysis

Chirag K Kumar

Kirsty Sands

Timothy R Walsh

Seamus O’Brien

Mike Sharland

Joseph A Lewnard

Hao Hu

Padmini Srikantiah

Ramanan Laxminarayan

Roles

Abstract

Background

Methods and findings

Conclusions

Author summary

Why was this study done?

What did the researchers do and find?

What do these findings mean?

Introduction

Methods

Data

Modeling techniques

Results

Effect of a K. pneumoniae maternal vaccine on neonatal deaths

Fig 1. A maternal vaccine reduces global K. pneumoniae burden.

Fig 2. AMR distribution of vaccine-avertable deaths and cases.

Projected global and future benefits of a K. pneumoniae maternal vaccine

Fig 3. Projected global impact of a maternal K. pneumoniae vaccine.

Fig 4. Projected benefits of a maternal vaccine under future AMR scenarios.

Discussion

Supporting information

Acknowledgments

Abbreviations

Data Availability

Funding Statement

References

Decision Letter 0

Philippa Dodd

Roles

Decision Letter 1

Philippa Dodd

Roles

Author response to Decision Letter 1

Decision Letter 2

Philippa Dodd

Roles

Author response to Decision Letter 2

Decision Letter 3

Philippa Dodd

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases