© Springer Nature Singapore Pte Ltd. 2019
Hemanshu Prabhakar and Zulfiqar Ali (eds.)Textbook of Neuroanesthesia and Neurocritical Carehttps://doi.org/10.1007/978-981-13-3390-3_29

29. Sepsis

Swagata Tripathy1  
(1)
Department of Anesthesia and Intensive Care, All India Institute of Medical Sciences, Bhubaneswar, India
 
 
Swagata Tripathy
29.1 Definition
29.2 Epidemiology of Sepsis
29.3 Pathophysiology of Sepsis
29.4 Immune Dysregulation Following Neurotrauma
29.5 Infections in the Neurocritical Care Unit
29.5.1 Post-Craniotomy Infections
29.5.2 Device Related Infections
29.6 Systemic Infections
29.7 Prevention of Sepsis in the Neurointensive Care Unit
29.8 Management of Sepsis
29.8.1 Management of Infection
29.8.2 Fluid Therapy
29.8.3 Vasopressors
29.8.4 Steroids
29.8.5 Respiratory Failure
29.8.6 Kidney Dysfunction
29.8.7 Transfusion
29.8.8 Nutrition
29.9 Conclusion
References

Keywords

Dysregulated immune responseSepsisSeptic shockIntracranial infectionEVD-related infectionHealthcare-associated ventriculitis and meningitisEarly treatment

29.1 Definition

When normally sterile areas of the body are invaded by pathogenic microbes, the resulting manifestation of systemic disease that ensues is commonly referred to as sepsis. The signs and symptoms of this entity may be difficult to separate from other non-infectious conditions such as acute pancreatitis. The principal reason for this is that diverse stimuli (infectious or not) may result in activation of the immune mechanism of the body releasing endogenous peptides and cytokines.

In recent history, the definitions of sepsis and the sepsis syndrome complex have been revised three times—in 1990 and 2001, sepsis was defined as the association of inflammatory responses with proof, or suspicion, of an infectious source. In the presence of at least single organ dysfunction, sepsis progressed to “severe sepsis” and when inflammation caused by infection resulted in hypotension requiring vasopressors, it was defined as “septic shock” [1, 2]. In the recent (third) consensus conference however, the term severe sepsis was removed from the triad, with sepsis being defined as life-threatening organ dysfunction caused by a dysregulated host response to infection. Septic shock is a subset thereof in which profound abnormalities (cellular, circulatory, and metabolic) are observed, resulting in a greater risk of mortality [3].

In a clinical context, the presence of inflammatory response may be recognized by various clinical and laboratory parameters (like fever, tachycardia, tachypnoea, increased white cell counts, positive microbiological cultures, or serology). An increase in SOFA (sequential organ failure assessment) score by 2 or more points enables early recognition of organ dysfunction in patients with sepsis as per recent guidelines. New onset organ failure has been associated with an in-hospital mortality of 10% or more. Septic shock may be identified clinically when serum lactate level is >2 mmol/L or there is a need to supplement vasopressors in a patient who is adequately volume resuscitated, to maintain a mean arterial pressure >65 mmHg. A combination of sepsis and septic shock is associated with mortality rate >40%. The quick SOFA (qSOFA) has been suggested as a new bed side clinical score (at least 2 of—respiratory rate of 22/min or greater, altered mentation, or systolic blood pressure of 100 mmHg or less) which may enable rapid identification of patients likely to have sepsis in out of ICU settings like the wards, emergency areas, etc.

29.2 Epidemiology of Sepsis

There has been a progressive rise in the incidence and prevalence of sepsis globally. Sepsis has been identified as the most common cause of readmission to the ICU. It has been referred to as the final common pathway from infection to death [410]. The reported incidence and prevalence of sepsis and septic shock varies according to the definitions used, patient population studied, and geographic location. The concern over the changing definitions of sepsis affecting the epidemiology of sepsis has been recently addressed by Shankar-Hari et al. [11]. The authors compared the incidence, outcomes, trend in outcomes, and predictive validity of patients classified according to the Sepsis 2 [2] versus the Sepsis 3 criteria [3]. They found an incidence of 101.8 and 19.3 per 100,000-person years for sepsis and septic shock, respectively, in 2015. Sepsis 2 severe sepsis and Sepsis 3 sepsis have similar incidence and mortality. The Sepsis 3 criteria identifies a similar sepsis population as the Sepsis 2 criteria (overlap of 92%); the septic shock population identified is smaller and sicker, with a higher predictive validity [11].

Sepsis after neurosurgery and in the neurointensive care unit occurs in up to 36% of patients admitted for more than 48 h [12, 13]. The most common infections are pneumonia, urinary tract infections (UTI), blood stream infections, and intracranial infections such as ventriculitis and meningitis.

29.3 Pathophysiology of Sepsis

A breach in the integrity of physical barriers of the human body such as skin, mucous membrane, gastrointestinal tract (GIT), and conjunctiva by microorganisms begins a process of localized inflammation. This localized inflammation may then spill over to generate systemic manifestations such as fever, tachycardia, tachypnea, and altered white cell counts. It was previously believed that a hyperactive host response to infection affects organ function due to widespread activation of inflammatory cascade via indigenous peptides such as cytokines, interleukins, and tumor necrosis factors. However, various studies have failed to show a consistent benefit of using antagonists of TNF or IL-1 (usually increased in serum of sepsis patients and believed to be the culprits) [1416]. A meta-analysis of these pooled studies shows an overall improvement, however [17].

Another school of thought argued that septic patients may have a hypoactive response to infection; corroborating research shows that ICU patients have decreased expression of IL6 and TNF in response to endotoxin stimulation [18, 19]. Using granulocyte colony-stimulating factor to treat >700 patients with pneumonia and severe sepsis did not result in any improvement in survival [20].

The current thinking now is that the inflammatory response in patients with sepsis is complex and cannot be easily classified as enhanced or blunted. In some patients, blunting of the response may improve outcome, whereas in others a suppression may serve better. Bones has suggested that a treatment tailored to the individual patient with greater emphasis on identifying the cause than looking for a “magic bullet” may serve the “sepsis cause” better [21]. The coagulation pathway is also affected, with increased coagulability and decrease in fibrinolysis (Fig. 29.1). Alterations at the cellular level include leukocyte apoptosis, neutrophil hyperactivation, and endothelial cell failure. At a metabolic level, insulin resistance may result in hyperglycemia; a subset of patients with adrenal failure show better outcomes with steroid supplementation. High blood glucose levels have been shown to suppress polymorphonuclear neutrophils and decrease bactericidal activity; strict glycemic control may provide protection to endothelial cells [2224].
../images/455944_1_En_29_Chapter/455944_1_En_29_Fig1_HTML.png
Fig. 29.1

Pathophysiology of sepsis

29.4 Immune Dysregulation Following Neurotrauma

Evidence from animal and human studies shows that injury and ischemia in the brain tissue results in a dysregulated immune response consequential to a “brain injury induced immunosuppression syndrome.” The systemic immunosuppression has been noted after stroke, trauma, surgery, spinal cord injury, or subarachnoid hemorrhage. An intense activation of the hypothalamo-pituitary axis and the sympathetic system resulting in an early (within 12 h) increased immune response followed by a phase of delayed suppression after 24 h (lasting up to several weeks) has been seen. Mediated by b2 receptors, this phenomenon has been associated with increased susceptibility to infections in patients with neurologic insult [2528].

Studies have found that the size and site of the injury (most evidence coming from post-stroke patients) correlates with the magnitude of immunosuppression. Lesions in the orbitofrontal cortex, insular cortex, and putamen areas have a higher risk of pneumonia. Although no laterality predilection was observed, an overlap has been found in the correlates of dysphagia and pneumonia among these patients [29, 30]. Patients with intraventricular hemorrhage (IVH) show decreased baroreceptor sensitivity; this is an independent risk factor for increased infection and morbidity after IVH [31, 32].

29.5 Infections in the Neurocritical Care Unit

National Healthcare Safety Network (NHSN), Division of Healthcare Quality Promotion of the Centers for Disease Control and Prevention (CDC) for infections, includes three types of CNS infections in the acute care setting: intracranial such as brain abscess, subdural, or epidural infections and encephalitis; ventriculitis or meningitis and spinal abscess without meningitis [33].

29.5.1 Post-Craniotomy Infections

Four out of ten patients develop an infection after craniotomy [34]. Patients have a higher risk of developing pneumonia, ventilator associated infection, and urinary and intracranial infections [3537]. The demarcation of infectious from inflammatory meningitis is difficult because the clinical signs and symptoms overlap. Intracerebral hemorrhage, inflammatory reactions to interventions, and immunosuppression can significantly change CSF profiles. Nosocomial meningitis may result in severe morbidity, need for repeated surgeries, prolonged LOS, and higher hospital costs. A recent study has proposed a prediction model for risk of developing nosocomial meningitis after neurosurgery. In it SAH, CRP ≥6 mg/dL, and CSF/serum glucose ratio ≤0.4 mmol/L get 1 point, whereas CSF leak and CSF PMN neutrophils ≥50% get 1.5 points and CSF lactate ≥4 mmol/L gets 4 points. The model has shown good calibration (Hosmer–Lemeshow goodness of fit = 0.71) and discrimination (area under the receiver operating characteristic curve = 0.94). A score ≥6 points suggests a high probability of neuroinfection, for which antibiotic treatment should be considered [38].

29.5.2 Device Related Infections

Devices in the central nervous system such as external ventricular drain (EVD), lumbar drain (LD), and intracranial pressure (ICP) monitors increase the risk of infections in neuropatients. Risk factors include duration of catheterization, frequency of EVD manipulation for CSF sampling or irrigation, presence of IVH, and insertion technique [39]. The signs and symptoms, diagnosis, and treatment of hospital acquired meningitis and ventriculitis have been clearly laid down by the infectious Diseases Society of America recently [40]. Tables 29.1 and 29.2 mention the clinical features and lab parameters for detecting or suspecting meningitis and/or ventriculitis of nosocomial origin—due to devices, neurosurgery, or trauma. Advanced tests for diagnosis as suggested by the society are—an increased CSF procalcitonin or lactate to differentiate between bacterial infection or inflammatory CSF changes, detection of β–D-glucan and galactomannan in CSF for the diagnosis of fungal infection, and using nucleic acid amplification for early detection of infections. Magnetic resonance imaging with gadolinium enhancement and diffusion-weighted imaging for suspected meningitis/ventriculitis and CT scan or ultrasound abdomen for patients with infected VP shunt and abdominal symptoms is recommended for detecting abnormalities in patients with healthcare-associated ventriculitis and meningitis [40].
Table 29.1

Signs and symptoms of healthcare-associated meningitis and ventriculitis

Condition

Signs and symptoms (grade of recommendation) with healthcare-associated ventriculitis and meningitis

CSF shunt or drain

 • New onset headache, nausea, lethargy, and/or change in mental status suggest CSF shunt infection (strong, moderate)

 • Erythema and tenderness over the subcutaneous shunt tubing suggest CSF shunt infection (strong, moderate)

 • In the absence of another clear source of infection, fever could suggest CSF shunt infection (weak, low)

 • Features of peritonitis or abdominal tenderness in patients with ventriculoperitoneal shunts, in the absence of another clear etiology, indicate CSF shunt infection (strong, moderate)

 • Features of pleuritis in patients with ventriculopleural shunts, in the absence of another clear etiology, indicate CSF shunt infection (strong, moderate)

 • Demonstration of bacteremia in a patient with a ventriculoatrial shunt, in the absence of another clear source of bacteremia, is evidence of CSF shunt infection (strong, moderate)

 • Demonstration of glomerulonephritis in a patient with a ventriculoatrial shunt suggests CSF shunt infection (weak, low)

 • New or worsening altered mental status in patients with external ventricular drains suggests infection (weak, low)

 • New fever and increased CSF white blood cell count in patients with external ventricular drains could suggest infection (weak, low)

Neurosurgery or traumatic brain injury

 • New headache, fever, meningeal irritation, seizures, and/or worsening mental status suggest ventriculitis or meningitis (strong, moderate)

 • Fever, in the absence of another clear source of infection, suggests CNS infection (weak, low)

Table 29.2

Laboratory diagnosis of healthcare-associated ventriculitis and meningitis

Variable

Healthcare-associated ventriculitis and meningitis—CSF findings (grade of recommendation)

Lab reports—cell count, glucose, and protein

 • Abnormalities of CSF cell count, glucose, and/or protein may not be reliable indicators for the presence of infection in patients with healthcare-associated ventriculitis and meningitis (weak, moderate)

 • Normal CSF cell count, glucose, and protein may not reliably exclude infection in patients with healthcare-associated ventriculitis and meningitis (weak, moderate)

 • A negative CSF Gram stain does not exclude the presence of infection, especially in patients who have received previous antimicrobial therapy (strong, moderate)

Culture

 • CSF cultures are the most important test to establish the diagnosis of healthcare-associated ventriculitis and meningitis (strong, high)

 • If initial CSF cultures are negative in patients with CSF shunts or drains with suspected infection, it is recommended that cultures be held for at least 10 days in an attempt to identify organisms such as Propionibacterium acnes (strong, high)

 • If a CSF shunt or drain is removed in patients suspected of having infection, cultures of shunt and drain components are recommended (strong, moderate)

 • If a CSF shunt or drain is removed for indications other than infection, cultures of shunt or drain components are not recommended (strong, moderate)

 • Blood cultures are recommended in patients with suspected ventriculoatrial shunt infections (strong, high)

 • Blood cultures may be considered in patients with ventriculoperitoneal and ventriculopleural shunts (weak, low)

 • Single or multiple positive CSF cultures in patients with CSF pleocytosis and/or hypoglycorrhachia, or an increasing cell count, and clinical symptoms suspicious for ventriculitis or meningitis, are indicative of CSF drain infection (strong, high)

 • CSF and blood cultures in selected patients should be obtained before the administration of antimicrobial therapy; a negative CSF culture in the setting of previous antimicrobial therapy does not exclude healthcare-associated ventriculitis and meningitis (strong, moderate)

The recommendations for treatment of healthcare-associated meningitis or ventriculitis include single shot antibiotic prophylaxis with protocolized drain insertion (hair clipping and sterile field) in cases of indwelling catheters, prompt removal of catheters which are infected, and treatment according to sensitivity pattern (if available). Prolonged administration of antibiotics in patients with indwelling catheters has been seen to increase the risk of infections with multi drug resistant organisms; stopping long term prophylaxis may reduce other hospital associated infections [41, 42].

29.6 Systemic Infections

The incidence of nosocomial infections is very high in patients with neurologic injury. It varies from 20 to 40% in various conditions like TBI, ischemic stroke, intracranial hemorrhage, SAH, and status epilepticus [4345].

The most common of these infections is ventilator associated pneumonia (VAP). This is associated with poor outcomes and longer periods of mechanical ventilation and hospital stay [46]. Patients at a particularly higher risk of nosocomial pneumonia are younger males with prolonged mechanical ventilation. Intubation at the scene or in the emergency department, lower Glasgow Coma Scale score, and higher injury scores, particularly thoracic injuries, also increase the risk of pneumonia [47].VAP may occur early (within 5 days) or late (after the fifth day) of intubation or mechanical ventilation. The causative organisms associated with the two are different, with more virulent, multi drug resistant organisms isolated in late onset VAP. Methicillin-sensitive S. aureus, Haemophilus influenzae, Streptococcus pneumonia, and Acinetobacter species are common pathogens in patients with VAP [48].

Urinary tract infections (UTI) occur commonly in the neuro-ICU. Rates range from 10 to 24%. Long duration of catheterization, frequent patient transport, and underlying immunosuppression may all contribute to the high rates of UTI. Independent risk factors for developing a urinary tract infection in a cohort of patients with SAH included older age, female sex, non-infectious complications, intracerebral hemorrhage, and diabetes [49].

Catheter related blood stream infections increase morbidity and length of stay in the neuro-ICU. Incidence may be as high as 30%, with gram positive organisms isolated more often. Prolonged duration, site of central line insertion, and nursing practices may influence the rates of infection [50, 51].

Surgical site infection rates vary from 1 to 15%. Independent predictive risk factors for infection are cerebrospinal fluid leakage, external shunt, longer operation time, craniotomy, dural substitute, staples in wound closure, and further neurosurgery [52, 53]. The rate of infection after spine surgery is as high as 18%. Diabetes, prolonged operative times (>3 h), obesity, posterior approach, and number of intervertebral levels (≥7) are associated with an increased risk of SSI after spinal surgery [54, 55].

Clostridium difficile infection is seen increasingly in ICUs. In neuro-ICUs the prevalence of 0.4–0.5% has been reported with an infection rate of 8.3 per 10,000 patient days [56, 57]. Prolonged hospitalization, use of antibiotics, and advanced age are the major risk factors associated with developing C diff. infections.

29.7 Prevention of Sepsis in the Neurointensive Care Unit

Measures to reduce infections and sepsis in the critical care unit by prior interventions have been studied. Antibiotic prophylaxis in post-stroke patients and oral decontamination with povidone iodine in patients with TBI have not proven to be of any benefit—the rates of VAP have actually been seen to go up after using mouth care with povidone iodine [58, 59]. On the other hand, protocols to improve screening for swallowing difficulties have shown a reduction in the incidence of aspiration pneumonia [60]. The use of noninvasive ventilation where possible, early tracheostomy and protocolized early extubation, may reduce the days on mechanical ventilation and the rates of ventilator associated pneumonia [61, 62].

29.8 Management of Sepsis

The surviving Sepsis Campaign lays down the guidelines for quick recognition and management of sepsis. Details are out of the scope of the chapter, but the important points are highlighted.

29.8.1 Management of Infection

The SSC 2016 [63] has recommended administration of broad-spectrum antimicrobials within 1 h and rapid source control. Proper dosing of antibiotics according to the pharmacokinetic and pharmacodynamic profiles is needed. Recent studies indicate that continuous infusion of b-lactam antibiotics may result in better cure rates than intermittent bolus dosing [64, 65]. Procalcitonin levels may help to guide shorter antibiotic course and encourage the search for other causes of inflammatory response. A recent study has shown improved mortality in the group managed by procalcitonin levels [66].

29.8.2 Fluid Therapy

The guidelines recommend the administration of 30 mL/kg of intravenous crystalloid within the first 3 h. Further administration is to be guided by frequent assessment and application of fluid challenge techniques. With additional fluid based on frequent reassessment, dynamic indices of hemodynamic assessment are recommended; the application of a fluid challenge technique in which fluids should be continued as hemodynamic factors continues to improve with the use of crystalloids (balanced crystalloids or saline). Conservative resuscitation strategy may result in better pulmonary outcomes, as shown in a recent study [67]. The guidelines suggest that albumin may be considered as an alternate fluid if large volumes of crystalloids are to be infused. Although many studies do not show a mortality benefit of albumin over crystalloids, the ALBIOS trial has shown better mean arterial pressures at 6 h of resuscitation and lower total fluid balance in patients of septic shock resuscitated with albumin along with crystalloids [68].

29.8.3 Vasopressors

Guidelines recommend norepinephrine as the first-line vasopressor, being associated with lower adverse events rates such as arrhythmias than dopamine. Epinephrine and vasopressin may also be added to norepinephrine to reduce the requirement of norepinephrine—“relative vasopressin deficiency” has been described in patients with septic shock [63]. A recent meta-analysis however does not show any difference between noradrenaline and dopamine when administered to patients with hypotensive shock [69].

29.8.4 Steroids

A dose of up to 200 mg/day of intravenous hydrocortisone is recommended in patients in septic shock refractory to fluid and vasopressor resuscitation. A more recent study has not demonstrated any benefits (mortality or recovery from shock) of steroid administration in refractory septic shock [70].

29.8.5 Respiratory Failure

Respiratory failure is common in patients with sepsis, with ARDS being underrecognized in a majority of cases. Lung protective ventilation with early proning for a minimum of 12 h is recommended in the guidelines [63]. Recent studies show beneficial role of statins in ARDS, use of high flow oxygen to reduce reintubation (and for early extubation), and that conservative oxygen therapy may improve outcomes as compared to conventional doses [7174].

29.8.6 Kidney Dysfunction

The SSC guidelines recommend either continuous or intermittent RRT. CRRT may be preferred in hemodynamically unstable septic patients. The best time of initiation is still unclear as the two recent trials—AKIKI and ELAIN—have shown differing results: with AKIKI showing no difference in mortality and ELAIN showing improved mortality and renal function recovery with early onset on RRT [75, 76].

29.8.7 Transfusion

Restrictive practices are recommended, as per the SSC guidelines. These are based on the threshold for transfusion in a hemoglobin level <7 g/dL. The age of the RBC units to be transfused does not affect patient outcome [77, 78].

29.8.8 Nutrition

Adequate nutrition has an important role in patients in the neurocritical care unit. Current guidelines recommend early enteral nutrition over parenteral [63].

29.9 Conclusion

Infections and sepsis in the neurointensive care unit (neuro-ICU) are common. Different disease pathologies have different types of infections: pneumonia is most common. Craniotomy and devices place these patients at an increased risk for meningitis and ventriculitis. These patients are inherently more susceptible to infections due to dysregulated immune responses after acute brain injury. Preventive measures, early recognition, and prompt management may improve patient outcomes.

Key Points

  • The signs and symptoms of sepsis may be difficult to separate from other non-infectious conditions.

  • Sepsis after neurosurgery and in the neurointensive care unit occurs in up to 36% of patients admitted for more than 48 h.

  • The most common infections are pneumonia, urinary tract infections (UTI), blood stream infections, and intracranial infections such as ventriculitis and meningitis.

  • Craniotomy and devices place these patients at an increased risk for meningitis and ventriculitis.