Infections in clinical settings

Obese patients have increased intensive care unit (ICU) length of stay⁽Reference Dossett, Heffernan and Lightfoot¹⁴⁾ and are more likely to die⁽Reference Bochicchio, Joshi and Bochicchio^15–Reference Bercault, Boulain and Kuteifan¹⁷⁾ in the hospital. Further, several epidemiological investigations have reported that obesity increases infection susceptibility in clinical settings⁽Reference Falagas and Kompoti⁹⁾. Obese patients are prone to developing post-operative complications. In fact, numerous studies have reported obesity to be an independent risk factor for post-operative infections⁽Reference Olsen, Nepple and Riew^18–Reference Swenne, Lindholm and Borowiec²⁸⁾. In a recent secondary analysis of a large prospective observational study including critically ill and injured patients remaining in the ICU for 48 h or more, obesity was reported to be an independent risk factor for catheter and blood stream infections⁽Reference Dossett, Dageforde and Swenson¹¹⁾. A study in critically injured blunt trauma patients reported that morbid obesity (BMI≥40) was associated with increased risk of pneumonia and urinary tract infection but not with increased mortality⁽Reference Dowsey and Choong²¹⁾.

The impact of obesity on clinical outcomes in hospitalised patients is clearly multifactorial and complex. In addition to decreased immunocompetence, there are other potential factors that may contribute to increased susceptibility to infection in the hospital setting. For example, underlying disease in the obese may inhibit proper mobility in the hospital, which can increase risk for skin breakdown⁽Reference Mathison²⁹⁾. Due to inadequate equipment or improperly trained staff, obese individuals may have prolonged visits at the hospital, thus increasing risk for acquiring nosocomial infections⁽Reference Falagas and Kompoti^9, Reference Mathison²⁹⁾. Another consideration is that pharmacokinetics of antibiotics may differ in the obese, potentially affecting susceptibility to post-operative infections⁽Reference Hanley, Abernethy and Greenblatt³⁰⁾. Therefore, it is difficult to determine the direct impact of impaired immunity on severity of nosocomial infections in obese patients, but accumulating evidence suggests a significant role.

Obesity and respiratory infections

A striking number of recent studies have reported obesity to be a predictor for a worse outcome of infection with the 2009 influenza A (H1N1) pandemic strain⁽Reference Jain and Chaves³¹⁾. In fact, several countries across the world have reported data indicating that obese individuals were disproportionately represented among influenza-related hospitalisations and deaths. Obesity or morbid obesity increased risk of ICU admission and even death among those infected with the pandemic strain⁽Reference Hanslik, Boelle and Flahault^32–Reference Morgan, Bramley and Fowlkes³⁵⁾. Those admitted to ICU had a reportedly longer duration of mechanical ventilation and increased time in ICU and hospitals compared with non-obese individuals⁽Reference Díaz, Rodríguez and Martin-Loeches³⁶⁾. Before the advent of the 2009 pandemic season, there were no such reports investigating the relationship between obesity and influenza infection in human subjects. Recently, however, Kwong et al. published a study that explored the relationship between BMI and seasonal influenza infection using a series of Canada's cross-sectional population-based health surveys⁽Reference Kwong, Campitelli and Rosella³⁷⁾. The surveys covered twelve influenza seasons. Analysis of the retrospective cohort demonstrated that the obese are at greater risk for respiratory hospitalisations during the seasonal flu periods.

In contrast to the recent surge of publications highlighting a connection between influenza severity and obesity, there is very little known about obesity and other respiratory tract infections⁽Reference Mancuso³⁸⁾. A recent study by Akiyama et al. suggests that obesity may impact the response to respiratory syncytial virus infection in children⁽Reference Akiyama, Segawa and Ida³⁹⁾. A study from Poland reported that BMI was significantly related to susceptibility to respiratory infections in children⁽Reference Jedrychowski, Maugeri and Flak¹³⁾. In critically ill trauma patients, obesity or morbid obesity was associated with respiratory infections⁽Reference Dossett, Dageforde and Swenson^11, Reference Bochicchio, Joshi and Bochicchio^15, Reference Newell, Bard and Goettler²³⁾. Conversely, a few studies have also reported obese individuals are not at greater risk for respiratory infections⁽Reference Dossett, Heffernan and Lightfoot^14, Reference Brandt, Harder and Walluscheck⁴⁰⁾. Therefore, our understanding of the effect of obesity on risk for pulmonary infection remains unclear. However, it is important to consider that obesity can complicate lung mechanics, such as restricting lung volume (reviewed in⁽Reference McClean, Kee and Young⁴¹⁾), which could potentially increase risk for pneumonia or other infections. Although the mechanisms contributing to increased susceptibility may include impaired immunity, there may be non-immune factors to consider.

Vaccination and management of infection in the obese

Vaccination is universally recommended by public health officials to combat several types of infections. However, there is some evidence suggesting that obese individuals may not respond to vaccination to the same extent as healthy-weight individuals. Obesity was associated with a poor antibody response to hepatitis B vaccination⁽Reference Weber, Rutala and Samsa^42, Reference Weber, Rutala and Samsa⁴³⁾. Additionally, overweight children displayed considerably lower anti-tetanus IgG antibodies in response to vaccination compared with healthy-weight children⁽Reference Eliakim, Swindt and Zaldivar⁴⁴⁾. This reduced immunogenicity in response to vaccination could be caused by several factors. One possibility is impaired generation and/or function of the antibody-secreting plasma cells. Another factor could be reduced absorption of the vaccine at the site of injection due to excess adiposity⁽Reference Eliakim, Swindt and Zaldivar⁴⁴⁾. Interestingly, a recent study reported that using a larger vaccine needle length resulted in considerably higher antibody titres to hepatitis B surface antigen in obese adolescents⁽Reference Middleman, Anding and Tung⁴⁵⁾. As mentioned, poor response to vaccination has important public health implications. Reduced protection against viral infections, such as hepatitis B, increases individual susceptibility and may increase the likelihood of transmission to others. Therefore, more research on how obesity may negatively impact vaccination response is necessary to ensure proper protection in this at risk population.

In addition to the infections mentioned earlier, increased BMI is associated with greater risk for several other bacterial infections including periodontal infections⁽Reference Ylöstalo, Suominen-Taipale and Reunanen¹²⁾, Staphylococcus aureus nasal carriage⁽Reference Herwaldt, Cullen and French⁴⁶⁾ and gastric infection by Helicobacter pylori ⁽Reference Perdichizzi, Bottari and Pallio⁴⁷⁾. Also, a recent study reported that obesity was significantly associated with herpes simplex virus 1 infection, which was determined by seropositivity⁽Reference Karjala, Neal and Rohrer⁴⁸⁾. Despite increased risk for several types of microbial infections, there is little known about how obesity may alter the pharmacokinetics of antimicrobial drugs (reviewed in⁽Reference Hanley, Abernethy and Greenblatt³⁰⁾). Studies assessing dosing of the antibacterial drug, vancomycin, suggest that obese patients may require different dosages⁽Reference Bearden and Rodvold⁴⁹⁾ and different dosing intervals⁽Reference Bauer, Black and Lill⁵⁰⁾ compared with non-obese individuals. An additional study of the antimicrobial drug, linezolid, reported diminished serum concentrations in the obese compared with healthy-weight volunteers receiving the same dose⁽Reference Stein, Schooley and Peloquin⁵¹⁾. As pointed out previously, there is greater risk of skin breakdown in obese individuals in clinical settings due to restricted mobility, improperly sized rooms and equipment and the special challenges of caring for patients undergoing or post-bariatric surgery⁽Reference Falagas and Kompoti^9, Reference Mathison²⁹⁾. Additionally, excess adiposity may reduce tissue perfusion and affect wound healing⁽Reference Mathison²⁹⁾. Taken together, it is clear that obesity predisposes individuals to nosocomial infections, and that careful consideration of infection prevention and treatment is required.

Immunomodulatory adipokines and hormones in obesity

The primary adipose derived immunomodulatory adipokines include leptin, adiponectin and the pro-inflammatory cytokines: TNFα, IL-6 and IL-1β⁽Reference Fantuzzi^76, Reference Tilg and Moschen^77, Reference Koerner⁷⁹⁾. Adiponectin, levels of which are decreased during obesity, has been shown to alter natural killer cell cytotoxicity and cytokine production by human myeloid cells⁽Reference Wolf, Wolf and Rumpold^80, Reference Kim, Kim and Han⁸¹⁾. Conversely, there is excess production of TNFα, IL-6 and IL-1β in white adipose tissue of the obese⁽Reference Tilg and Moschen⁷⁷⁾. These cytokines can be secreted into the blood and potentially have distal effects; however, exactly how chronic production of these cytokines impacts cellular immunity remains to be elucidated. It is possible that chronic exposure to pro-inflammatory cytokines may desensitise immune cells to inflammatory responses during an actual infection⁽Reference Ziegler-Heitbrock, Wedel and Schraut⁸²⁾.

The pleiotropic effects of leptin on immune cell activity are highly diverse and complicated⁽Reference La Cava and Matarese⁵⁴⁾. Nearly all cells of the innate immune system express the isoform of the leptin receptor, obRb, required for leptin signalling⁽Reference Lord, Matarese and Howard^83–Reference Caldefie-Chezet, Poulin and Tridon⁸⁷⁾. In monocytes, leptin up-regulates pro-inflammatory cytokine production of IL-6, IL-12 and TNFα, as well as phagocytic activity⁽Reference Mancuso, Gottschalk and Phare^55, Reference Loffreda, Yang and Lin^88, Reference Gainsford, Willson and Metcalf⁸⁹⁾. In polymorphonuclear neutrophils of healthy individuals, leptin signalling induced chemotaxis, reactive oxygen species generation and influenced oxidative capacity⁽Reference Caldefie-Chezet, Poulin and Tridon^87, Reference Caldefie-Chezet, Poulin and Vasson^90, Reference Montecucco, Bianchi and Gnerre⁹¹⁾. Natural killer cells are highly influenced by leptin signalling, including aspects of differentiation, proliferation, activation and activity⁽Reference Zhao, Sun and You^85, Reference Tian, Sun and Wei⁹²⁾. Given the importance of leptin to innate immune cell function, it follows that nearly all innate immune cells are impaired in mice lacking intact leptin signalling.

The adaptive arm of the immune response is equally affected by leptin signalling⁽Reference Lord, Matarese and Howard^83, Reference Papathanassoglou, El-Haschimi and Li⁸⁶⁾. Leptin is an important source of pro-survival signals to double-positive and single-positive thymocytes during the maturation of T-cells⁽Reference Howard, Lord and Matarese⁹³⁾. Leptin has been shown to play a key role in lymphopoieses and myelopoieses given that ob/ob mice had only 60% as many nucleated cells in bone marrow as compared with wild-type controls⁽Reference Claycombe, King and Fraker⁹⁴⁾. In the presence of a polyclonal stimulator, leptin can increase T-cell proliferation and can modulate expression of activation markers on both CD4⁺ and CD8⁺ T-cells⁽Reference Martín-Romero, Santos-Alvarez and Goberna⁹⁵⁾. Leptin can also have profound effects on cellular activity by functioning as a regulator of immune cell metabolism⁽Reference De Rosa, Procaccini and Calì^62, Reference MacIver, Jacobs and Wieman⁹⁶⁾.

Although several papers have discussed how leptin may be required for or may enhance immune cell function, few have taken into consideration the fact that obese individuals are hyperleptinaemic⁽Reference Matarese, Moschos and Mantzoros⁹⁷⁾. Therefore, in obese models, we should ask what are the potential impacts of excess leptin signalling on immune cells? Indeed, studies have demonstrated that T-cells⁽Reference Papathanassoglou, El-Haschimi and Li⁸⁶⁾ and natural killer cells⁽Reference Nave, Mueller and Siegmund⁹⁸⁾ can become resistant to leptin in rodent models of obesity. Leptin signals through a JAK/STAT (Janus kinase/signal transducer and activator of transcription) signalling pathway, resulting in translocation of the transcription factor, STAT3 (signal transducer and activator of transcription 3), into the nucleus and subsequent transcription of leptin-induced genes, including suppressor of cytokine signalling-3⁽Reference Bjørbæk, El-Haschimi and Frantz⁹⁹⁾. Suppressor of cytokine signalling 3 functions as a negative feedback mediator of JAK/STAT signalling, and thus may play an important role in impairing leptin signalling and contributing to central and peripheral leptin resistance⁽Reference Bjørbæk, Elmquist and Frantz¹⁰⁰⁾. Leptin resistance could very well explain obesity-related impaired immunity, as this would mimic a state of leptin deficiency. Leptin resistance, induced by hyperleptinaemia, would obviously not occur in ob/ob mice, which is another reason these mice are not the best model for studying obesity-related immune dysfunction. Although it is widely accepted that leptin resistance occurs centrally in the hypothalamus⁽Reference Bjørbæk, Elmquist and Frantz^100–Reference Munzberg and Myers¹⁰²⁾, peripheral leptin resistance requires further investigation.

An additional and somewhat novel consideration is how hyperleptinaemia may impact the function and distribution of regulatory T-cells (Treg). An elegant study by De Rosa et al. demonstrated a role for leptin signalling in Treg proliferation and function⁽Reference De Rosa, Procaccini and Calì⁶²⁾. Abrogation of leptin signalling alters the anergic state of Treg, and allows for enhanced proliferation. Although highly proliferating Treg tend to lose some of their suppressive activity⁽Reference De Rosa, Procaccini and Calì⁶²⁾, resistance to leptin signalling might contribute to greater Treg number. Treg have the capacity to suppress nearly all aspects of the immune response⁽Reference Miyara and Sakaguchi¹⁰³⁾. Thus, this hypothesis fits with the immunosuppressive phenotype associated with infections in the obese. In fact, a recent investigation demonstrated a greater percentage of Treg in the spleen of DIO mice despite lower levels in adipose tissue⁽Reference Deiuliis, Shah and Shah¹⁰⁴⁾. It is clear that obesity can alter Treg number and function⁽Reference Deiuliis, Shah and Shah^104–Reference Ilan, Maron and Tukpah¹⁰⁶⁾, but the extent to which this population of immune cells affects infection outcomes in the obese remains unknown.

Hyperinsulinaemia and insulin resistance are common features of obesity; however, there is little known regarding the immunomodulatory effects of excess insulin or impaired insulin signalling in the context of obesity. How the effects of insulin on cellular immunity are only partially understood. Monocytes have been shown to express insulin receptors and are insulin sensitive immune cells⁽Reference Defronzo, Soman and Sherwin^107–Reference Liang, Han and Okamoto¹¹⁰⁾. Interestingly, resting T-cells are insulin insensitive in that the insulin receptor is absent from the plasma membrane. However, once T-cells are activated by a polyclonal stimulator, such as phytohaemagglutinin or by specific antigen, effector T-cells up-regulate de novo emergence of insulin receptors⁽Reference Stentz and Kitabchi^111, Reference Viardot, Grey and Mackay¹¹²⁾. Insulin signalling induces glucose uptake, amino acid transport, lipid metabolism and can modulate T-cell activation and function⁽Reference Stentz and Kitabchi^111, Reference Helderman¹¹³⁾. Furthermore, insulin promotes an anti-inflammatory T-helper type 2 cell phenotype⁽Reference Viardot, Grey and Mackay¹¹²⁾, but MacIver et al. speculate that insulin resistance in obesity may actually enhance T-helper type 1 cell development⁽Reference MacIver, Jacobs and Wieman⁹⁶⁾. It is thus clear that insulin can have potent effects on immune cell metabolism and function, but the effects of excess insulin on immunity remain relatively unresearched.

Altered immune cell metabolism in an abnormal metabolic environment

Immune cells from both innate and adaptive defences require nutrients such as glucose, amino acids and fatty acids to meet energy needs⁽Reference Delacre, Pot and Grangette¹¹⁴⁾. However, energetic demands and nutrient preference depend on cell type and cellular activity. For example, once T-cells are activated, they become highly proliferative and secretory, and thus require an abundant source of energy that will rapidly yield large quantities of ATP⁽Reference Frauwirth and Thompson¹¹⁵⁾. Conversely, macrophages and neutrophils are generally considered to be non-proliferative and thus have a different metabolic profile and nutrient requirements⁽Reference Delacre, Pot and Grangette^114, Reference Newsholme, Rosa and Newsholme¹¹⁶⁾. Although glucose and fatty acids are important sources of energy for host defence and immune function⁽Reference Calder¹¹⁷⁾, elevated levels of these nutrients, as in the obese, may have consequences for immune cell activity.

Glucose uptake by immune cells is facilitated by the family of glucose transport proteins, GLUT. A variety of GLUT are expressed on immune cells. For example, increased expression of GLUT3 and GLUT5 occurs during the differentiation of monocytes to macrophages⁽Reference Fu, Maianu and Melbert¹¹⁸⁾. GLUT1 appears to be the primary GLUT on T-cells, and functions to maintain glucose uptake for basic metabolic requirements⁽Reference MacIver, Jacobs and Wieman^96, Reference Frauwirth and Thompson¹¹⁵⁾. Upon stimulation, GLUT1 and GLUT3 levels were shown to increase on T-cells and monocytes⁽Reference Fu, Maianu and Melbert¹¹⁸⁾. Glucose is required for proper T-cell proliferation and survival⁽Reference MacIver, Jacobs and Wieman⁹⁶⁾. However, it has also been shown that exposing T-cells to high concentrations of glucose can result in reactive oxygen species generation and lipid peroxidation⁽Reference Stentz and Kitabchi¹¹⁹⁾. Although little is known of the in vivo effects of hyperglycaemia on immune cell function, Jacobs et al. demonstrated that overexpression of GLUT1 in mouse T-cells resulted in altered T-cell metabolism and cytokine production⁽Reference Jacobs, Herman and MacIver¹²⁰⁾. The mechanisms by which elevated glucose influence immune cell function are not entirely clear, but glucose plays a crucial role in activity of immune cells, and thus excessive levels are likely to have a significant impact on cellular function.

Similar to glucose, fatty acids are important in fuelling an immune response, as they are a readily available source of abundant energy. However, the impact of excess circulating NEFA, a hallmark of obesity⁽Reference Bergman and Ader¹²¹⁾, on immune cells has not been well studied. Interestingly, SFA, such as palmitate, share similarities in chemical structure to lipopolysaccharide. This observation sparked studies indicating that SFA can induce an inflammatory response by initiating Toll-like receptor (TLR) signalling pathways⁽Reference Lee, Sohn and Rhee^122–Reference Weatherill, Lee and Zhao¹²⁶⁾. TLR are critical in inducing innate immune responses as they recognise conserved molecular patterns on microbial pathogens⁽Reference Kawai and Akira¹²⁷⁾. SFA, but not unsaturated, have been shown to activate both TLR2 and TLR4 resulting in TIR (Toll/IL-1 receptor) domain-containing adaptor-inducing interferon-β-dependent and myeloid differentiation factor 88-dependent signalling pathways and a subsequent inflammatory response⁽Reference Lee, Sohn and Rhee^122–Reference Shi, Kokoeva and Inouye¹²⁸⁾. NEFA have been shown to trigger inflammatory responses in both macrophages and dendritic cells indicating that both innate and adaptive immune responses can be affected⁽Reference Weatherill, Lee and Zhao^126, Reference Nguyen, Favelyukis and Nguyen¹²⁹⁾. The effect of elevated NEFA on insulin resistance and type II diabetes has been widely examined⁽Reference Shi, Kokoeva and Inouye¹²⁸⁾. Cells of both innate and adaptive immunity express TLR2 and TLR4, and other than a small number of studies in macrophages⁽Reference Odegaard and Chawla³,129–Reference Laine, Schwartz and Wang¹³¹⁾, there is very little known of the impact of NEFA on TLR signalling in other immune cells, such as T-cells. However, a study by Stentz and Kitabchi reported that increasing concentrations of palmitate but not unsaturated fatty acids resulted in activation of T-cells and a dose-dependent increase in cytokine production as well as reactive oxygen species generation and lipid peroxidation in vitro ⁽Reference Stentz and Kitabchi¹³²⁾.

An additional consideration in which excess fatty acids, as well as glucose and metabolic hormones, may affect immunity is highlighted in several recent studies demonstrating that T-cell populations can have distinct metabolic programmes that are critical to cell fate and function⁽Reference Michalek, Gerriets and Jacobs¹³³⁾. Michalek et al. show that CD4⁺ T-cell metabolism is fundamental in regulating differentiation to an effector or regulatory subtype. The CD4⁺ T effector subset requires glycolytic metabolism, and Treg require lipolytic oxidation⁽Reference Michalek, Gerriets and Jacobs¹³³⁾. Interestingly, recent studies have shown that CD8⁺ effector T-cells displayed a glycolytic phenotype, whereas a CD8⁺ memory T-cell population was associated with a lipid oxidation metabolic profile⁽Reference Pearce, Walsh and Cejas^134–Reference Rao, Li and Odunsi¹³⁶⁾. What remains to be studied is how nutrition may alter these distinct metabolic programmes. In the context of obesity, how do elevated levels of glucose, fatty acids and metabolic hormones, such as leptin and insulin, impact the metabolic fate of immune cells during an infection?