Article Text

Download PDFPDF

Stroke-induced immunosuppression and poststroke infection
  1. Kaibin Shi1,2,
  2. Kristofer Wood2,
  3. Fu-Dong Shi1,2,
  4. Xiaoying Wang3,
  5. Qiang Liu1,2
  1. 1 Departments of Neurology, Tianjin Neurological Institute, Tianjin Medical University General Hospital, Tianjin, China
  2. 2 Departments of Neurology, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona, USA
  3. 3 Departments of Neurology and Radiology, Neuroprotection Research Laboratory, Massachusetts General Hospital, Neuroscience Program, Harvard Medical School, Boston, Massachusetts, USA
  1. Correspondence to Dr Qiang Liu, Departments of Neurology Tianjin Neurological Institute, Tianjin Medical University General Hospital Tianjin China; qliu.asu{at}


Infections occur commonly after stroke and are strongly associated with an unfavourable functional outcome of these patients. Approaches for effective management of poststroke infection remain scarce, presenting an urgent need for preventive anti-infection strategies for patients who have suffered a stroke. Emerging evidence indicates that stroke impairs systemic immune responses and increases the susceptibility to infections, suggesting that the modification of impaired immune defence could be beneficial. In this review, we summarised previous attempts to prevent poststroke infections using prophylactic antibiotics and the current understanding of stroke-induced immunosuppression. Further elucidation of the immune mechanisms of stroke will pave the way to tailored design of new treatment to combat poststroke infection via modifying the immune system.

  • stroke
  • infection
  • post-stroke immunosuppression

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Infectious complications, pneumonia, urinary tract infections and infections in other organ systems, are common in patients with stroke with an incidence of ~30%.1–3 Poststroke infection is associated with about 20% of the deaths and related to considerable morbidity in stroke survivors.1 4–6 Given the well known detrimental effects of stroke-associated infection, effective management is critical. Antibiotics are the traditional approach used to manage infections, however, the recently completed clinical trials haven’t demonstrated significant benefit of prophylactic antibiotics,7–10 presenting an urgent need to better understand the pathogenesis of stroke-associated infection and identify viable approaches to combat infectious complications. The inhibition of immunity after stroke has been recognised as a key contributor to infection in patients with stroke. Our increasing knowledge on stroke-induced immunosuppression poses an opportunity to boost immune defence and restrict poststroke infection. In this review, by summarising previous studies regarding attempts to manage poststroke infections and mechanisms of stroke-induced immunosuppression, we aim to provide insight into the basis of stroke-induced immunosuppression and propose new modalities to restore host immune defence after stroke.

Prophylactic antibiotic treatment

Studies in animal models of ischaemic stroke have demonstrated that preventive treatment with antibiotics reduces the incidence of infections, and improves mortality and neurological function.9 Based on these encouraging findings, a series of clinical trials that tested the safety and efficacy of prophylactic use of antibiotics in patients with stroke have been conducted (table 1). Among these 15 clinical studies, patient inclusion, stroke types, antibiotics selection and treatment duration differ from each other. The majority of these studies treated patients with broad-spectrum antibiotics to cover the most common causative bacteria of pneumonia and urinary tract infections, except for three studies that tested the neuroprotective effects of minocycline, which has inadequate bacterial coverage for patients with stroke; the infection results were not reported in these studies as well.11–13 A meta-analysis of a portion of these studies concluded that preventive treatment with antibiotics could reduce infection rates, but failed to reduce mortality and improve functional outcomes.14 This observation resembles findings in two recently completed phase III trials.7 8 Results from the Preventive Antibiotics in Stroke Study (PASS), which included 2358 patients from 30 Dutch centres, show that preventive antibiotic treatment after stroke was able to reduce poststroke infections, but did not improve functional outcomes at 3 months.8 The other phase III trial, testing prophylactic antibiotics after acute stroke for reducing pneumonia in patients with dysphagia (STROKE-INF), enrolled 1217 patients from 48 stroke units in the UK; it reported that prophylactic antibiotics did not reduce the frequency of pneumonia within 14 days after stroke onset, either as defined by algorithm or diagnosed by a physician. The secondary end point analyses showed that prophylactic antibiotics did not improve functional recovery at 3 months or mortality.7

Table 1

Clinical studies of antibiotic treatment in patients with stroke

Although the PASS and STROKE-IFN studies are different in many aspects, such as treatment time window, antibiotics selection and patient inclusion criteria, the results of the two studies are highly consistent, that is, the treatment regimens of both studies failed to reduce the rate of pneumonia. Additionally, a recent study which tested antibiotic treatment in patients selected by the ultrasensitive procalcitonin test did not find a reduction of pneumonia in antibiotic-treated patients either.15 Therefore, since the antibiotic selection and treatment time window in these studies cannot prevent poststroke pneumonia, a possible explanation is that stroke-associated pneumonia might partly be a pneumonitis with more sterile components.16 In light of this speculation, the pathophysiological changes in the lung poststroke warrant further investigations for possible identification of new treatment strategies.

Although these clinical studies indicated that several antibiotics in different patients with stroke are safe, limitations regarding the selection of antibiotics for patients with stroke were also noted. For example, some antibiotics have neurotoxic effects and can worsen outcome. A preclinical study found that enrofloxacin (a fluoroquinolone antibiotic) treatment from day 1 to day 7 in rats undergoing middle cerebral artery occlusion, an ischaemic stroke model, worsened outcomes at 1 month.17 In this context, antibiotics that have neuroprotective properties should be considered in priority, like minocycline12 18 and ceftriaxone.19 However, the benefit of ceftriaxone was not demonstrated in PASS, which yielded to the secondary consideration of antibiotics choice, that is, whether broad-spectrum antibiotics are suitable for intervention in patients with stroke. Exposure to these broad-spectrum antibiotics may cause dysregulation of normal flora in the body, which might influence the stroke outcome, as suggested by a recent animal study in which broad-spectrum antibiotic treatment caused depletion of mice intestinal flora, worsening the outcome of stroke.20 Therefore, the selection of antibiotics for patients with stroke must comprehensively consider the benefits as well as secondary adverse effects of the treatment.

As a severe type of stroke, intracerebral haemorrhage (ICH) accounts for about 10%–15% of all stroke types and is associated with high mortality and morbidity.21 22 Infection is also a common and important complication in patients with ICH, and it can influence neurological function in the acute phase as well as long-term recovery.5 23 The reported infection rates in patients with ICH are similar to those with acute ischaemic stroke (AIS), about 30% of all patients, with higher rates in more severe patients admitted to an intensive care unit.24 25 Currently, studies that have specifically tested antibiotic treatment in patients with ICH are still lacking. However, four of the published studies mentioned above (table 1) also recruited patients diagnosed with ICH. Two studies included only 26 and 11 patients with ICH, respectively, and thus are not suitable for subgroup analysis. In PASS, 269 patients with ICH were enrolled; subgroup analysis indicated that preventive ceftriaxone does not improve outcome in these patients either, however, whether or not preventive ceftriaxone was able to reduce the rate of pneumonia post ICH was not reported. In the STROKE-IFN Study, 125 patients with ICH were included; subgroup analysis indicated that prophylactic antibiotic treatment might reduce the rate of pneumonia in patients with ICH within 14 days after stroke, a noteworthy finding that warrants further confirmation in future studies with larger sample sizes.

Overall, for patients with AIS, prophylactic antibiotic treatment can reduce the rates of both total and urinary tract infections, but no such benefit was seen for stroke-associated pneumonia. In addition, preventive antibiotic treatment does not improve the outcome in patients. Hence, these findings do not support the use of antibiotics in a preventive manner. Preclinical and clinical evidence for the use of antibiotics in patients with ICH requires further enlightenment.

Stroke-induced immunosuppression

The immune system and nervous system crosstalk with each other via multiple facets to maintain the homoeostasis of both systems under physiology conditions.26–30 Severe brain injuries, including stroke, can interrupt such balance and lead to a series of changes in both systems.31 Brain injury after an ischaemic or haemorrhagic stroke leads to the activation and infiltration of inflammatory cells into the brain.32–35 While poststroke inflammation may contribute to the clearance of tissue debris and tissue repair, most published literature indicates that inflammation in the brain during the acute phase of stroke promotes the expansion of stroke lesions and worsens neurodeficits.36–38 Further, inhibition of the immune response in the brain during the acute phase can limit the extent of stroke injury.33 34 38–42 Conversely, the injured brain can reshape peripheral immunity and transition the functional status of the peripheral immune system from competence to suppression after the acute phase of stroke, as manifested by lymphopenia, decreased levels of inflammatory cytokines, monocyte and lymphocyte dysfunction, and atrophy of secondary lymphoid organs.31 36 43 44 Brain lesion size has been considered as an independent risk factor of poststroke immunosuppression and infectious complications.45 It has been postulated that stroke-induced immunosuppression might be an adaptive response to acute brain injury, because systemic immunosuppression may limit the overwhelming inflammation in the brain or reduce the occurrence of autoimmune reactions against neuroantigens.46 47 However, a protective role of immunosuppression after stroke, if any, is uncertain and remains to be defined.

A consequence of immunosuppression has been linked to the increased risk of infection after stroke onset.47 An established concept is that brain-derived neurogenic innervations are in control of systemic immunity. In the context of stroke, brain injury-induced activation of neurogenic pathways, including the sympathetic innervation, hypothalamic-pituitary-adrenal (HPA) axis and parasympathetic innervation work together to influence the magnitude and intensity of systemic immune response.43 48 The central mediators of these pathways include norepinephrine, acetylcholine and glucocorticoid hormones.27 31 Receptors of these molecules are broadly expressed in immune cells.43 49 Activation of the sympathetic nervous system induces the release of catecholamines from sympathetic nerve terminals. Transient increase of catecholamine release often leads to increased blood immune cells by mobilising them from peripheral reservoirs such as the spleen; however, stroke induces a prolonged elevation of circulating catecholamines,44 which promotes the apoptosis of immune cells and leads to a decrease of peripheral immune cells and a bias towards T helper cell 2 (Th2) immune response.50 51 Glucocorticoids, key mediators for HPA axis function, have antiproliferative effects to induce apoptosis of immune cells.52 Additionally, glucocorticoids can promote the production of anti-inflammatory cytokines like transforming growth factor β (TGF-β),53 while inhibiting the production of proinflammatory cytokines, such as interleukin (IL) 1, IL-8 and umor necrosis factor α (TNF-α).54 The anti-inflammatory effects of cholinergic pathways mediated by acetylcholine was recently uncovered; macrophages can be quickly inhibited when exposed to acetylcholine, with decreased secretion of proinflammatory cytokines like TNF-α, IL-1β and IL-18.55 In addition to peripheral immunosuppression, stroke also suppresses immunity in the ischaemic brain. Ischaemic neurons can secrete several neurotransmitters like acetylcholine (Ach), glutamate, serotonin, and so on, which can directly interact with the receptors of peripheral infiltrated lymphocytes and modulate their functional status.43 A model of stroke-induced immunosuppression is  illustrated in figure 1 using natural killer (NK) cells as an example. The identification of the mechanisms underlying stroke-induced immunosuppression offers the possibility of the use of immune modulation as a novel approach to boost host defence and combat poststroke infections.

Figure 1

Schematic diagram of mechanisms of stroke-induced immunosuppression with NK cells as an example. In the early stages of stroke (<24 hours), ischaemic neuron-recruited NK cells are swiftly mobilised into ischaemic areas, where they promote neuronal death (not shown in figure).34 Subsequently (>48 hours?), ischaemic neuron-derived signals can turn off NK cells that express neurotransmitter receptors. At the peripheral level, ischaemic brain injury influences the sympathetic, parasympathetic (vagus nerve) and/or hypothalamic-pituitary-adrenal (HPA) axis systems that suppress NK cell-mediated immunity. Differences in the spectrum of neurogenic innervations, immune cell subsets and soluble mediators in the CNS versus the periphery may differentially affect NK cell deficiency in these two compartments. BBB, blood brain barrier; BDNF, brain-derived neurotrophic factor; CNS, central nervous system; EGF, epithelial growth factor; GABA, gamma-aminobutyric acid; NK, natural killer; UTP, uridine triphosphate.

Potential therapies to combat stroke-associated infection

Animal experiments demonstrated that pharmacological blockade of the sympathetic pathway or HPA axis by β-blockers or glucocorticoid receptor antagonists could reduce stroke-induced immune dysfunction and poststroke infection to improve animals’ functional outcome (table 2).48 56 However, the clinical relevance remains unclear (table 3). A small prospective study indicated that β-blocker use is associated with less severe stroke at presentation, with lower thrombin, haemoglobin A1C levels, and erythrocyte sedimentation rate.57 Another retrospective analysis of 841 patients with ischaemic stroke, 10.6% of whom received β-blockers during hospitalisation, also implied that β-blockers might be neuroprotective, as the use of β-blockers was associated with reduced risk of early death; however, this association no longer exists when excluding mortality from cardiovascular causes.58 Of interest, a study that included 625 patients, among which 553 patients were admitted with AIS and 72 with ICH, reported that β-blocker exposure, defined as receiving β-blockers prior to stroke and continued treatment during hospitalisation, did not reduce the risk of stroke-associated pneumonia, but reduced urinary tract infection rates. In addition, patients receiving β-blocker therapy showed a higher 30-day mortality than those without exposure to β-blockers.59 However, a recent non-randomised study of 5212 patients with ischaemic stroke suggested that, either treatment with β-blockers before stroke onset or starting treatment after stroke onset was associated with reduced frequency of pneumonia, and β-blocker exposure during stroke was associated with reduced mortality.60 In PASS, 885 patients received β-blockers, but subgroup analysis of this factor was not reported and further analysis regarding the safety and efficacy of such therapy on stroke outcome, as well as in combination with preventive antibiotic treatment, is awaited. Thus, the safety of systemic β-blockade remains controversial in the context of patients with stroke, and randomised trials may be warranted for further determination. In particular, the adverse effects of systemic inhibition of sympathetic input by β-blockers might counteract its protective effect of immune modulation, which warrants further verification. Ultimately, developing selective blockers of lymphocytic β-adrenergic receptors or downstream signalling pathways might avoid the adverse systemic impacts.

Table 2

Preclinical studies regarding neurogenic pathway modulation after stroke

Table 3

Clinical use of β-blockers in patients with stroke

Boosting peripheral immunity might serve as another viable approach to countering poststroke infections and avoiding the systemic effects of neurogenic innervations. Preclinical data have demonstrated that adoptive transfer T or NK cells, as well as injection of IFN-γ to animals during the subacute stage of ischaemic stroke (>24 hours) reduces the rate of pneumonia and improves the outcome.43 56 Additionally, treating animals subjected to brain ischaemia with α-galactosylceramide, a specific activator that can promote proinflammatory  cytokines production by the invariant natural killer T cell  (NKT) (iNKT) cell, is able to reduce stroke-associated infections.48 Currently, the safety and efficacy of this approach has not been tested in clinical trials. Nevertheless, before translation to the clinic, the contribution of different cell types and cytokines to restrict stroke-associated infections require further investigations.


Poststroke infections, especially pneumonia, still present challenges for clinical management of patients with stroke. Preventive administration of antibiotics can somewhat reduce infection rates but not pneumonia, and cannot improve the outcome and survival in patients with stroke; thus, the use of antibiotics in a prophylactic manner remains controversial. Modulation of the immune system via neurogenic pathways may serve as a potential therapy for patients with stroke. Reported clinical studies indicated a discrepancy in the responsiveness to treatment between poststroke pneumonia and other infections, such as urinary tract infection. This suggests that poststroke pneumonia could be more refractory to treatment than other types of infections. Considering the currently limited treatment options for stroke, future design of stroke treatment is imperative to mitigate immunosuppression, and thus to decrease the risk of infectious complications after stroke.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.


  • Contributors QL, F-DS and XW formulated the concept, KS reviewed the articles and drafted the manuscript, KW contributed to discussion and editing the manuscript.

  • Funding This study was supported in part by the National Basic Research Program of China, grant 2013CB966900; National Science Foundation of China, grant 81230028, 81301044, 81471535; American Heart Association, grant 16SDG27250236; National Institutes of Health, grant R01NS092713; National Multiple Sclerosis Society, grant RG-1507-05318.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; externally peer reviewed.

  • Data sharing statement No additional data are available.