Rudolph's Pediatrics, 22nd Ed.

CHAPTER 504. Global Burden of Respiratory Infections and Mucosal Immunology of the Respiratory Tract

Jay K. Kolls and Joseph P. Mizgerd


The World Health Organization (WHO) has been tracking the global burden of infectious diseases for several decades. During that time, mortality resulting from acute respiratory infections has been increasing, surpassing diarrheal disease in the late 1990s to become the number one killer of children worldwide. Since 1990, the WHO has used the Burden of Disease Project to assess disease-related morbidity and mortality, including the summary measure of disability-adjusted life years (DALYs) lost, which takes into account the degree and duration of morbidity in addition to mortality. When data are assessed by the DALYs measure, lung infections are again remarkably prominent. The attributable morbidity and mortality of acute respiratory infections exceeds HIV/AIDS, cancer, and heart disease. Of particular note is that these data do not include morbidity and mortality resulting from tuberculosis or to AIDS-related pneumonias, thereby underestimating the true impact of respiratory infections (Fig. 504-1). When stratified by age, there is a large and disproportionate burden of morbidity and mortality attributable to lower respiratory tract infections (LRIs) in the 0-to-4 and 5-to-14 age groups. Across the decades, even for the youngest patients, the burden of disease from LRIs exceeds that from malaria or diarrheal disease (Fig. 504-2). Although there is a strong association of LRIs with poverty, LRIs account for remarkable morbidity and mortality in wealthy countries as well.1 In the wealthiest countries, LRIs cause a greater burden than any other infectious disease. Although effective vaccines are available for specific pathogens such as Hemophilus influenza b and some serotypes of Streptococcus pneumoniae, the number of organisms capable of causing lung infections is too numerous to feasibly vaccinate against each. An approach that bolsters lung immunity per se may have the greatest impact in reducing morbidity and mortality against LRIs. It is imperative to advance our understanding of both innate and adaptive immunity to pathogens in the lung.

FIGURE 504-1. Global burden of disease as measured by the disability-adjusted life years (DALYs). Acute respiratory infections cause more morbidity than HIV/AIDS, cancer, diarrhoeal disease, malaria, or tuberculosis (TB).

FIGURE 504-2. Disproportionate burden of disease in children ages 0 to 4 and 5 to 14. Acute respiratory infection causes the greatest combined morbidity and mortality in children.


With the exceptions of neonatal pneumonias in which the lungs become seeded by hematogenous dissemination of group B streptococcus or Escherichia coli, most pneumonias result from aspiration of organisms colonizing the upper airways.2 If this inoculum is small or of lower virulence, innate immunity consisting of cough, mucociliary clearance, antimicrobial peptides, and/or resident airway or alveolar macrophages can eradicate the infection. However, if the inoculum is large (as can occur with oral anerobes in children with poor dentition) or if more virulent pathogens (such as encapsulated bacteria) are aspirated, then the end result can be pneumonia. The host inflammatory response, or recruitment of innate immunity reserves, is critical to the development and outcome of clinical pneumonia. For example, inflammation mediated by early cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) can result in fever and malaise, but these cytokines are also essential for host defenses against some pathogens in the lungs.2 Local production of granulocyte-colony stimulating factor (G-CSF) is essential to eradicate many pathogens,3,4 in part by regulating granulopoiesis and neutro-phil apoptosis. In fact, the local G-CSF response may explain why many patients with bacterial pneumonia present with elevated white blood counts in peripheral blood.3 These inflammatory responses can also regulate the severity of illness by altering ventilation-perfusion relationships leading to hypoxemia as well as regulating lung compliance, in part by regulating the amount of exudate in the alveolar spaces in the lung.


Normally, neutrophils constitute less than 1% of the cells in the alveolar air spaces. However there is a large number of marginated neutrophils in the extensive network of pulmonary capillaries, and these can be recruited to the alveolar air space very rapidly. Although some strains of Staphylococcus aureus are cleared by macrophages in the absence of recruited neutrophils, neutrophil emigration into the lung is critical for most bacteria.2 In addition, neutrophils have been shown to be essential for control of viral and fungal pneumonias.5,6 Neutrophils kill extracellular pathogens through their ability to phagocytize organisms and to generate reactive oxygen species (ROS). Neutrophils also express antimicrobial proteins in their granules including defensins, which have direct microbicidal activity,7 and lipocalin-2, which binds certain bacterial siderophores and chelates iron necessary for bacterial growth.8 In addition to phagocytosis, neutrophils can form neutrophil extracellular traps (NETs), which consist of DNA and proteins that can immobilize and kill bacteria.9


There has been an explosion in knowledge over the last decade of how the host recognizes invading pathogens. It was already known by the mid-1990s that lipopolysaccharide or endotoxin, a major component of the gram-negative cell wall, was critical in eliciting host immunity to gram-negative pathogens. In fact, mice with a naturally occurring mutation in the lps gene locus were highly attenuated for the inflammatory response to purified LPS, as measured by the ability to produce TNF-α or IL-1β. However, these mice were also highly susceptible to gram-negative bacterial pneumonia, demonstrating the critical nature of this innate recognition pathway.10 Using positional cloning, Poltarek and colleagues showed that the product of the lps gene was a member of the Toll-like receptor family of proteins, TLR4, which is required for the recognition of LPS.11 Whereas wild-type mice induce over 200 genes in the lung four hours after instillation of gram-negative bacteria, mice with a mutation in tlr4 only induce 20 to 30 genes (Fig. 504-3).10 Thus, signaling through this single receptor accounts for well over 70% of the gene expression during experimentally induced gram-negative bacterial pneumonia. Humans with mutations in Tlr4 show reduced airways inflammation due to inhaled endotoxin,12 increased mortality due to sepsis,13 and increased risks of severe respiratory syncytial virus infection.14

FIGURE 504-3. Venn diagram displaying genes upregulated at 4 hours postchallenge with the gram-negative bacteria Klebsiella pneumoniae in the Tlr4 sufficient strains: C57BL/6N, C3H/HeN, and the Tlr4 insufficient strain C3H/HeJ. Data were analyzed by using Affymetrix software MAS version 5.0 and DMT version 3.0. Genes that were included had to be significantly upregulated at 4 hours versus the zero-hour time point and had to have a fold change between these two conditions of two or higher. Lists were imported into Genespring software version 6.0 (Silicon Genetics) for graphic illustration.

TLR4 can signal through two pathways. Signaling through the MyD88 adapter protein is shared by 9 other TLR family members, including TLR2 (which recognizes lipopetides from bacteria) and TLR9 (which recognizes unmethylated CpG DNA motifs common to bacterial chromosomes). Myd88 is also downstream from TLR5, which recognizes flagellin on organisms such as Pseudomonas aeruginosa. MyD88-deficient mice challenged with P. aeruginosa mice fail to make key proinflammatory cytokines (discussed above), such as IL-1β and TNF-α.15 Interestingly, these mutations prevent the development of clinical signs of infection in mice such as tachypnea or piloerection, but the mutant mice succumb rapidly due to widespread bacterial dissemination from the lung. These data illustrate the importance of the TLR-MyD88 pathway to alert the host of ongoing infection. Other recognition pathways such as nucleotide oligomerization domain (NOD)-like receptors and retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) are also capable of signaling from conserved microbial structures. During some lung infections (such as influenza pneumonia) recognition can be accomplished by the TLR pathway or the RLR pathway, each essential only when the other is absent.16


The inflammatory pathways described above are part of the innate immune response, immunity that is nonspecific and triggered by diverse inflammatory stimuli. The innate immune system contributes directly to host defense. Also, it is important for tailoring the adaptive immune responses characterized by recombination of antibody genes in B cells or recombination of T-cell receptor genes for antigen-specific T-cell responses.

Toll-like receptors help to link the innate and adaptive immune systems. Activation of these receptors leads to “maturation” of a subset of professional antigen presenting cells called dendritic cells, named for their long finger-like processes. Signaling from TLRs results in increased expression of molecules such as major histocompatibility complex class II (MHC-II), CD80, and CD86, altogether enhancing antigen presentation to T-cells. The presentation of antigen stimulates the CD4+ T-cell to differentiate toward distinct types of cytokine-secreting T helper (Th) cells, depending on cytokines and molecules expressed on dendritic cells. Of the major lineages described to date (which remains an incomplete characterization), those demonstrated relevant to acute lung infections include Th1 and Th17 cells. Th1 cells produce interferon-γ, which stimulates phagocytes to eliminate internalized microbes, and studies in mice have shown that this class of T-cells is critical for the control of intracellular pathogens such as viruses and Mycobacterium tuberculosis.17 Humans with mutations in interferon-γ or the interferon-γ receptors are highly susceptible to tuberculosis. Th17 cells are a more recently described lineage that expresses IL-17A, IL-17F, and IL-22.18,19 These cytokines serve several functions. IL-17A can act on epithelial cells as well as to stromal cells in the bone marrow to regulate granulopoiesis20,21 as well as neutro-phil recruitment to the site of infection.22 IL-22, on the other hand, regulates epithelial cell repair and antimicrobial peptide production at mucosal sites including the lungs, skin, and gastrointestinal tract.19,20,23The importance of the Th17 lineage in humans has recently been suggested by patients with Job (or Hyper IgE) syndrome, which results from mutations in the STAT3 transcription factor. These patients have susceptibility to pulmonary infections by Staphylococcus aureus and other microbes, as well as cutaneous infections by Candida albi-cans. Recently, it was discovered that these patients have defects in antigen-specific Th17 responses.25 In further support of the notion that Th17 cells are critical for mucosal immunity is the fact that mice deficient in IL-17 receptor signaling are susceptible to C albicansinfection and also spontaneously develop cutaneous S aureus infections. Both IL-17 receptor deficient mice and mice treated with anti-IL-22 antibodies succumb to a pulmonary challenge with the gram-negative pathogen Klebsiella pneumoniae.22,24


The global burden of respiratory disease is a reflection of our intimate coevolution with bacteria that colonize mucosal sites such as the upper airway, combined with the unique physiologic aspects of the lung for ventilation and gas exchange. Host factors influenced by socioeconomic status are key to disease prevention. The successful reduction in the global burden of pediatric lung infections will only be accomplished through better hygiene, maternal–child nutrition, vaccine delivery, and elucidation of mucosal immunity in the lung. The latter is required to unravel yet-to-be-determined immunodeficiencies and to develop rational approaches to improving host defense in the lungs.