Bacterial Sepsis

Etiology

The inflammatory response that leads to clinical syndromes such as sepsis is triggered by conditions that threaten the functional integrity of the host, in this case, microbial invasion. Both pro-inflammatory responses and anti-inflammatory responses characterize the host response. The extent of this process depends entirely on both pathogen factors (load, virulence, and pathogen-associated molecular pattern) and host factors (environment, genetics, age, other illnesses, and medications).

At its onset, the host cell will recognize the pathogen leading to a perpetuation of inflammation starting with leukocyte activation, then complement activation and coagulation activation, ultimately resulting in necrotic cell death. Normally, leukocyte activation, primarily neutrophils, are activated early in the inflammatory response cycle and undergo a respiratory burst, consuming lots of oxygen with the goal of creating toxic metabolites of oxygen. Neutrophils also use their myeloperoxidase enzyme to convert hydrogen peroxide to hypochlorite, a powerful germicidal agent (the active ingredient in bleach). These metabolites are stored in cytoplasmic granules and are released during neutrophil degranulation. This necrotic cell death releases further toxic metabolites, which causes further inflammatory pathway activation and further collateral tissue damage. This cascade can become a self-sustaining process if left unchecked.

Pro-inflammatory mechanisms are kept in check by the immune system via the humoral, cellular, and neural mechanisms. Regulation by the humoral mechanism is characterized by an impaired function of immune cells with the expansion of regulatory T cells and myeloid suppressor cells that further reduce inflammation. Regulation by the neural mechanism involves the hypothalamic-pituitary-adrenal axis and vagus nerve stimulating the adrenal gland to release norepinephrine and acetylcholine secretion by CD4 T cells which helps to suppress the release of pro-inflammatory cytokines. On the cellular level, there is inhibition of pro-inflammatory gene transcription.

The damage inflicted by the inflammatory process is largely due to oxidation, where the production of oxidants overwhelms the body’s endogenous antioxidant defenses. This collateral damage can be the result of inadequate antioxidant protection and is the basis behind newer research geared toward replacing endogenous antioxidants.[4]

Epidemiology

Sepsis is associated with high morbidity and mortality and most recently has been identified as affecting nearly 1.7 million US adults every year. Before 2000, mortality was noted to be as high as 50% in severe sepsis and septic shock patients.[6][10] Even with new technologies and treatments, the mortality remains around 20% to 25% nationwide, with 1 in 3 patients who die in the hospital succumbing to sepsis. Given the reported increasing prevalence of sepsis, a large amount of importance has been placed on treating and recognizing this disease process early, with the most recent budget reported at around $23.7 billion dollars in 2013.[11][12][13]

Pathophysiology

Sepsis progressing to septic shock and multi-organ failure results from a worsening circulatory insufficiency characterized by hypovolemia, myocardial depression, increased metabolic demands, and vasoregulatory perfusion abnormalities. Classically, septic shock and inflammatory shock have been described as primarily systemic vasodilatation of both arteries and veins. This dilation reduces ventricular preload as well as afterload, by decreasing systemic vascular resistance. The typical hemodynamic pattern in septic shock consists of low cardiac filling pressures, low central venous pressures (CVP), and low systemic vascular resistance (SVR). With low preload and afterload, cardiac output (CO) must increase to compensate, typically with increased heart rate (CO = Stroke Volume x Heart Rate).  This is also why sepsis is thought of as a distributive shock and is also known as hyper-dynamic or warm shock.

The vascular changes resulting in systemic vasodilatation can be attributed in large part to the dysfunction of the vascular endothelium and mitochondria. Part of the inflammatory process can be attributed to the enhanced production of nitric oxide (NO), which is a free radical and is produced in vascular endothelial cells. In addition to the inflammatory processes, sepsis is also associated with microvascular thrombosis. This is caused by activation of the clotting cascade with tissue factor and exacerbated by decreased anti-coagulation and impaired fibrinolysis. This leads to increased thrombus formation and therefore worsening tissue hypoperfusion. With thrombus formation and tissue hypoperfusion, the vascular endothelial cells also experience a loss of function with cadherin, and tight junctions are disrupted, leading to further capillary leak and increased interstitial edema. To worsen this effect, red blood cells (RBC) have been shown to have decreased deformability during this inflammatory phase and will further contribute to poor tissue oxygenation. Vasodilation, disruption of the vascular endothelium, decreased RBC deformability, with the ensuing hypotension all contribute to worsening tissue oxygenation.[4][14]

Impaired tissue oxygenation will also cause mitochondrial dysfunction and lead to the shunting of pyruvate away from the Kreb cycle, which, when it accumulates, will be converted to lactate as a means of continuing metabolism in the critically ill. Hyperlactatemia in sepsis is the result of multiple metabolic adjustments, not simply tissue hypoperfusion. Conditions that are known to favor lactate production include systemic hypoperfusion necessitating anaerobic metabolism, regional hypoperfusion, and microcirculatory dysfunction. Additionally, increased aerobic glycolysis can produce an excess of pyruvate that overwhelms pyruvate dehydrogenase used to take the pyruvate into the Kreb cycle for aerobic metabolism. With the excess pyruvate present, a portion will be shunted to lactate conversion and accumulate in the cytoplasm prior to being dispersed in the bloodstream. This occurs in white blood cells, which rely on anaerobic metabolism, especially when activated during periods of infection or inflammation, and have been shown to contribute to increased blood levels of lactic acid. This increased production and decreased clearance of lactate results in hyperlactatemia.[15][16]

As this disease process progresses, further inflammatory cytokines are released, resulting in effects seen in both cardiac activity and splanchnic vasculature. Further depression of cardiac contractility, in both systolic and diastolic phases, is a result of these circulating inflammatory cytokines. Splanchnic blood flow is also affected and primarily shunted to other organs, especially once hypotension is present. This creates an increased risk for translocation of enteric pathogens and endotoxin across the bowel mucosa and into the systemic circulation, further aggravating the inciting condition.

Alternatively, this process has been described in terms of oxygen demands and consumption. Sepsis, at its earliest manifestation, creates an increased need for oxygen locally, accompanied closely by a critical decrease in systemic oxygen delivery, limiting tissue oxygenation as described via various mechanisms above. This is followed by an increase in the systemic oxygen extraction ratio and an accompanying decrease in central venous oxygen saturation (ScvO2) and mixed venous oxygen saturation (SvO2). When the limits for how much oxygen can be extracted from blood are reached, cellular metabolism switches to anaerobic metabolism, further contributing to lactate production, as noted above. The failure to increase the oxygen extraction ratio and thus increase systemic oxygen consumption is secondary to the impairment of microvascular oxygen perfusion as well as mitochondrial dysfunction. This is the basis for following increasing lactate levels or decreasing venous oxygen saturation as a marker for disease severity and prognosis.[14]

Histopathology

The cellular and tissue changes induced by a shock are essentially those also seen in hypoxic injury states. As oxygen tension within the cell decreases, there is a loss of oxidative phosphorylation and decreased generation of ATP. The sodium pump fails, and potassium is lost with a concurrent influx of sodium and water, causing the cell to swell in size. With worsening hypoxia, there is a progressive loss of glycogen and decreased protein synthesis. As hypoxic injury continues, the cytoskeleton will disperse, and “blebs” will appear. Mitochondria will also exhibit swelling at this stage while the endoplasmic reticulum (ER) remains dilated. If oxygen is restored, a lot of these processes and injuries can be reversed. If hypoxia persists, irreversible injury and cellular necrosis will ensue. Cells at this stage exhibit severe swelling of mitochondria, extensive damage to plasma membranes, and swelling of lysosomes. Cellular death is mainly by necrosis. However, apoptosis also contributes as the mitochondria are believed to leak pro-apoptotic enzymes with swelling. Apoptosis is a pathway of cellular death that is induced by a tightly regulated program that uses its own enzymes to degrade the rest of the cell in the hopes of limiting collateral tissue damage.

In sepsis and, more specifically, septic shock, these changes typically are more evident in tissues that are heavily dependent on blood flow and the elimination of cellular by-products, i.e., brain, heart, lungs, kidneys, adrenals, and gastrointestinal (GI) tract. With respect to the kidneys, biopsies from septic patients reveal a wide range of findings with “non-specific morphologic changes,” not acute tubular necrosis, being the most commonly reported finding.[17][18] The lungs are another very commonly affected organ in septic shock as it is known to be somewhat resistant to hypoxic injury. However, in septic shock, changes of diffuse alveolar damage may develop; as a result, notably, findings of interstitial and intra-alveolar edema, inflammation, and fibrin deposition. These findings are also associated with the dreaded complication of disseminated intravascular coagulation (DIC), which will show further fibrin deposition, as a part of these micro-thrombi, in nearly all the vascular territories of the organs named above. The petechial appearance on the skin during DIC is often secondary to the consumption of platelets and clotting factors that are a part of the micro-thrombi. These changes, when viewed in isolation, are reversible; however, once a patient reaches a state where multiple organ systems are involved, the patient usually dies before they have a chance to recover.

History and Physical

The history and physical exam will vary widely depending on where they fall on the continuum from SIRS to septic shock. Important considerations when taking a history in a patient with suspected sepsis are geared toward assessing those risk factors associated with increased mortality or higher incidence of sepsis as well as inquiring about possible sources of infection. Comorbidities associated with morbidity and mortality in sepsis include active cancer, diabetes, chronic lung disease, congestive heart failure, renal insufficiency, and liver disease(cirrhosis). Age, specifically older than 65 years old, has been shown to be an independent predictor of mortality in sepsis. This is because of the association of increasing age with a decrease in the adaptive immune system, with B and T cells showing impaired functionality. Alternatively, gender does not seem to play a role in sepsis mortality, as studies focusing on this subject are conflicting.[19] The presence of any of these comorbidities should heighten the suspicion of the patient developing sepsis and prompt an earlier intervention if possible.

Gathering information geared toward finding an infectious source will help tailor further intervention and management. Notably, a patient’s recent history, hospitalization, and exposure to drug-resistant organisms may help direct what treatment to initiate. Retrospectively, sources of infection have been shown to be in the order of prevalence: pneumonia, intra-abdominal infections, and urinary tract infections.[1] However, the mortality rate is not related to the site of infection or the causative organism, including multi-drug-resistant organisms.[20] Even with this caveat, finding a suspected source of infection is critical in the early stages of a patient’s presentation.

The physical portion of the exam will provide critically important vital signs, allowing for the utilization of clinical screening tools.  There has been criticism of utilizing the SIRS criteria for this purpose because its specificity in sepsis remains low. However, the basic tenets of observing vital sign abnormalities and systemic response to infection remain valid. The differences and reliability of these clinical scores are discussed in further detail below. Clinical conditions that are associated with inflammatory injury are best grouped as organ systems related to their condition in order of prevalence:

• Lungs manifesting as acute respiratory distress syndrome (ARDS),
• Cardiovascular as septic shock (hypotension)
• Kidneys as acute kidney injury (AKI)
• Brain as septic encephalopathy
• Bone marrow as anemia of critical illness
• Skeletal muscle as critical illness myopathy
• Peripheral nerves as critical illness polyneuropathy

The physical exam should be focused on quickly assessing the patient with initial vitals and, if there is suspicion for sepsis, a thorough physical exam looking for a potential infectious source, including a pelvic exam if indicated.

Evaluation

With the increased emphasis placed on earlier detection of sepsis, several clinical scores have been proposed to help further predict which patients are septic and which patients will ultimately have a worse outcome. SIRS, as defined above, has been criticized for being too general, and while being very sensitive, it is also not very specific. The SIRS criteria were developed as a way of identifying this systemic response based on vital signs and lab work, not the source of infection.  However, when comparing even newer clinical tools, such as the qSOFA score (quick sepsis-related organ failure assessment), the SIRS criteria were still more sensitive.[21][22] Ultimately, the difficulty in validating and finding a clinical tool that will detect a septic patient 100% of the time is that there is no gold standard for diagnosing sepsis. 

To diagnose sepsis in a patient, we still rely on three groups of clinical and laboratory data, namely, general systemic manifestations, manifestations of organ dysfunction/failure, and finally, microbiological documentation. The general systemic manifestations are taken into account with the SIRS criteria. Manifestations of organ failure or dysfunction can be seen in changes of platelets, bilirubin, INR, creatinine, and lactic acid, among many other surrogate markers.[21] Microbiological documentation includes commonly blood cultures; however, they can include urine, peritoneal, synovial, respiratory secretions, and potentially cerebrospinal fluid(CSF).  The downside is that in over one-third of clinical cases of sepsis and septic shock, the blood cultures are negative, as well contaminants can complicate the picture.[4]

Sepsis-2 Criteria (2001) defined parameters of both physical and laboratory findings present with organ dysfunction, heavily influencing the Surviving Sepsis Campaign and CMS Core Measure (SEP-1). Organ dysfunction, as noted in SEP-1, are as follows: SBP less than 90 mm Hg, MAP less than 65 mm Hg, Acute respiratory failure (need for invasive or non-invasive mechanical ventilation), creatinine greater than 2.0, urine output less than 0.5 ml/kg per hour (for 2 consecutive hours), total bilirubin greater than 2 mg/dl, platelet count fewer than 100,000, INR greater than 1.5, or aPTT greater than 60 seconds, or lactate greater than 2 mmol/L. Evaluation for sepsis as the underlying process should be evaluated with these markers of end-organ dysfunction in mind.

With a high index of suspicion for sepsis, investigation typically includes:

• Complete blood count (CBC)with differential
• Comprehensive metabolic profile (CMP)
• Lactate
  • Critical value of greater than 2 mmol/L
• INR/aPTT
  • Elevated acutely as a result of end-organ dysfunction
• Chest x-ray and urinalysis
  • Either can be ordered to investigate for possible sources of infection
• Blood Cultures with or without other potential source cultures
  • At least 2 blood cultures are recommended before administering antibiotics
• Optional workup of uncertain benefit
  • Procalcitonin
  • C-reactive protein
  • Lipase
  • Venous blood gas
  • DIC panel (fibrinogen, D-dimer, Fibrin degradation products)
  • TSH
  • Rapid flu, respiratory virus panel
  • CT head with or without lumbar puncture
  • CT chest/abdomen/pelvis
  • MRI cervical/thoracic/lumbar spines (epidural abscess)

Treatment / Management

The treatment and management of sepsis have evolved consistently since it has been recognized as a disease entity. These advances can be credited for the mortality improvement in septic patients that have been seen over the last 2 decades. However, with mortality from sepsis still approaching 20% to 25%, there are improvements still needed to treat and manage these complex patients. Therapies are directed at the basic elements of sepsis as a syndrome of infection, the host response, and organ dysfunction. The initial management of infection requires forming a probable diagnosis, obtaining cultures, and initiating appropriate and timely empirical anti-microbial therapy and source control. Attenuating the host response and accompanying organ dysfunction are both parts of ongoing cardiopulmonary resuscitation. 

The initial resuscitation period was previously goal-directed, bundled care, something which the Surviving Sepsis Campaign guidelines promote. Three large trials subsequently have shown no mortality benefit over usual care, which likely reflects the improvement of care, in general, with more of a trend toward early goal-directed therapies.[6][23][24][25] Despite the numerous trials and recommendations evaluating the components of this resuscitation, it remains a subject of debate and the source of ongoing clinical trials. The management, because of its complexities, has been grouped into 2 bundles of care, a 3 and 6-hour acute sepsis bundle, and a 24-hour sepsis management bundle.(A1)

Components of Acute Sepsis Bundle

Three-Hour Bundle

• Obtain appropriate cultures before administration of antibiotics
• Obtain plasma lactate level
• Administer appropriate broad-spectrum antibiotics
• Begin administration of crystalloid at 30 ml/kg for hypotension or lactate level greater than or equal to 4 mmol/L within the first 3 hours

Six-Hour Bundle

• Start vasopressors if the patient is hypotensive during or after fluid resuscitation to maintain a mean arterial pressure level greater than or equal to 65 mmHg
• In the event of persistent hypotension after initial fluid administration, or if initial lactate is greater than 4 mmol/L, reassess volume status and tissue perfusion and document findings
• Re-measure lactate if initial lactate was elevated

Components of the Sepsis Management Bundle (to be completed in the next 24 hours)

• Administer low-dose steroid if indicated: Hydrocortisone at a dose of 200 mg per day
• Maintain blood glucose levels less than 180 mg/dL
• Maintain plateau airway pressure at less than 30 cm H2O in ventilator-dependent patients: Employ a lung-protective ventilator strategy
• Reassess antibiotic therapy daily for de-escalation when appropriate
• Early Initiation of enteral feeding rather than complete fast or parenteral nutrition
• Perform source control, where indicated, within 12 hours of identification
• Administer prophylaxis for deep vein thrombosis
• Administer stress-ulcer prophylaxis to prevent upper gastrointestinal (GI) bleeding

Antimicrobial therapy is strongly recommended to be started as soon as possible after the diagnosis of severe sepsis or septic shock has been made.  Delays in initiating antibiotic therapy have been associated with increased mortality. Recommendations have been made for initiating therapy within 1 hour of diagnosis; however, there appears to be some dispute about the importance of this timing.[26] Despite the observation that mortality from sepsis is not related to the site of infection or the causative organism, including multi-drug resistant organisms, the delay in administering appropriate antibiotics does affect mortality.[27] For this reason, broad-spectrum antibiotics targeting the suspected source are recommended, not necessarily the most appropriate antimicrobial agent.(A1)

Obtaining appropriate cultures is intended to influence care after this acute resuscitation phase has already passed.  Lots of emphasis has been placed in the past on the timing and technique by which to obtain a blood culture with the greatest likelihood of returning a true positive result.  There has been no correlation between the timing of a blood culture draw and detecting significant bacteremia.  Recommendations from the Infectious Disease Society of America (IDSA) emphasize drawing them in a sterile manner from at least two different locations, with the volume of the blood cultured being emphasized rather than timing.[28]

Volume Resuscitation, in the acute phase, is recommended as being 30 ml/kg of crystalloid fluids. This fixed number was taken as the average volume of fluid given to patients in the PROCESS[23], ARISE[24], and PROMISE[25] trials. The volume of intravenous fluids to give in early resuscitation is still a point of contention and debate, as half of those patients in the study did not receive 30 ml/kg, and half of those patients required more. The ideal fluid to administer is also a contested point, with a general recommendation for "crystalloid fluids" such as normal saline, Ringer's lactate, or balanced crystalloid solutions from the most recent Surviving Sepsis Campaign guidelines.[8] Recent large trials and reviews on the subject report some benefits with buffered crystalloid solutions (Ringer’s lactate) over normal saline.[29][30] The guidelines for treating and managing sepsis do not reflect these findings.(A1)

Vasopressors are indicated in septic shock if the condition fails to respond to the intravenous fluid bolus as recommended above.  Vasopressors provide additional vascular contraction to assist in maintaining beneficial blood pressure, with a target MAP of greater than 65 mm Hg.[31] The vasopressor choice initially should be norepinephrine, with vasopressin added if norepinephrine after titration dose fails to improve blood pressure by itself. Epinephrine should be used as a second-line agent to norepinephrine.  Recommendations at this point are to avoid the use of dopamine except in carefully selected patients. If there is evidence of myocardial dysfunction, recommendations support the addition of a dobutamine infusion to ongoing vasopressor therapy. Vasopressors should be administered through a central venous catheter as soon as the placement is possible.[8](A1)

Re-evaluation is critically important to guide the management of these patients after the initial resuscitation period. This typically includes a thorough clinical examination and the concurrent variables of heart rate, blood pressure, urine output, and respiratory rate, among others. There have been multiple attempts at clearly defining these goals for acute resuscitation, and have included, at different times, measuring central venous pressure (CVP), mixed venous oxygen saturation (SvO2), trending lactic acid, stroke volume measurements, mean arterial pressure (MAP), and urine output. CVP measurements have been fraught with imprecision and not found to be useful in predicting whether or not a patient will respond to additional intravenous fluid administration. Mean arterial pressure is the force behind perfusion in the peripheral organs and tissues and should be the goal for both fluid and vasopressor administration. An elevated lactic acid, while not directly a result of tissue perfusion, especially in sepsis, has been directly related to increased mortality. Lactate clearance has also been suggested as a goal for early resuscitation, as a decrease of only 10% in the first 6 hours corresponded to an 11% decrease in the likelihood of mortality, as compared to patients who were unable to achieve this.[15][16] However, there is weak evidence of a reduction in mortality when lactate-guided resuscitation is employed over usual care.[8] For compliance with CMS core measures, as noted in the guidelines above, documented reassessment of volume status and tissue perfusion should include vital signs, cardiopulmonary, capillary refill, pulse, and skin findings.(B2)

The widely held belief that in sepsis tissue, hypoxia was the problem led to liberal transfusions of packed red blood cells to increase oxygen delivery in the early years of sepsis management. This, however, has since been disproven with the transfusion requirements in septic shock trials addressing mortality with a lower threshold transfusion of 7 g/dL versus a 9 g/dL. Targeting a lower transfusion threshold of 7 g/dL in septic patients is the current recommendation unless there are signs of critical coronary disease, myocardial ischemia, or acute hemorrhage.[8][32](A1)

Corticosteroids have 2 important implications in severe sepsis and septic shock, namely that they have anti-inflammatory activity and that they magnify the vaso-constricting response to catecholamines. The conflicting support for the routine use of steroids in severe sepsis and septic shock has led to inconsistent recommendations for its use, despite more than 50 years of trials looking at its benefit in septic shock. Recently the CORTICUS and ADRENAL trials focusing on this point have failed to show a direct mortality benefit. However, there was a decreased length of ICU stay and days on vasopressors in survivors.[33][34] The APROCCHSS trial, in contrast, looking at a fludrocortisone and hydrocortisone combination, did show a mortality benefit in septic shock patients without an increase in adverse events.[35] Currently, the recommendations support its use in refractory septic shock that has not responded to intravenous fluids and vasopressor therapy as well a low-dose prolonged course should be continued while the patient is on vasopressor therapy. Hydrocortisone is the preferred corticosteroid because of its mineralocorticoid effects.(A1)

Mechanical ventilation is an important component of management as the respiratory system, predominantly the lungs, is the most commonly affected organ in severe sepsis and septic shock. In the event, the patient experiences acute lung injury that progresses to ARDS requiring mechanical ventilation, a lung-protective strategy is needed. In the landmark ARDSNet trial, patients who were receiving mechanical ventilation and had developed ARDS had lower mortality and decreased ventilator-dependent days with a lung-protective strategy. This strategy employed a lower tidal volume strategy of 6 ml/kg of predicted body weight and kept plateau pressures under 30 cm H2O.[36](A1)

Targeted treatments at different intersections in the newer models of sepsis have been continually developed and usually have been unsuccessful at providing any mortality benefit routinely. Drotrecogin alfa, which is recombinant human-activated protein C, was approved by the FDA in 2001 and was thought to promote fibrinolysis and inhibit thrombosis, which would have a significant impact on the pro-inflammatory state and complications involved in sepsis and septic shock. This was not the case, as it was withdrawn from the market after further studies were published showing no benefit and potentially worse outcomes. Further efforts were made to look at the administration of monoclonal antibodies against TNF, blocking IL-1 activity, granulocyte colony-stimulating factor (filgrastim), NO inhibitors, anti-thrombin, N-acetylcysteine, and using antibodies to endotoxin; however, these too failed to demonstrate efficacy.  Despite these failures, further development of other immunomodulating agents and the proposed "metabolic cocktail" continue to be pursued, as mortality, even with newer therapies, remains unacceptably high.[19][37](B2)

Differential Diagnosis

Endocrine

• Adrenal insufficiency
• Thyrotoxicosis

Toxidromes

• Salicylate toxicity
• Anticholinergic toxicity
• Malignant hyperthermia
• Neuroleptic malignant syndrome
• Heat stroke

Severe Inflammatory States

• Major trauma
• Pancreatitis
• Anaphylaxis
• Burns

Shock Types

• Cardiogenic shock
• Distributive shock (i.e., sepsis)
• Obstructive shock
• Hypovolemic shock

Prognosis

Although morbidity and mortality from sepsis have declined in the last 2 decades considerably, there remains a large need for improvement in management. This decline in mortality associated with sepsis can be in part attributed to earlier recognition and intervention, as well as advances in the understanding of this disease process. Important markers for disease severity and prognostic implications, such as the SIRS or qSOFA criterion, have been able to provide some framework for treatment as well as insight into the prognosis for affected patients. Goal-directed therapies that have contributed to this decline in mortality are geared towards goals that have significant prognostic implications. For example, lactic acid accumulation in the setting of sepsis has been shown to be highly specific for predicting the acute phase of death and in-hospital mortality as it increases with levels greater than 4 mmol/L having a 28.4% likelihood of mortality. Although using lactate to guide therapy has not been shown to independently affect mortality in these patients, 2-hour lactate clearance has been shown to be an independent predictor of in-hospital mortality.[15][16] With the implementation of bundle-based care, retrospective studies have also shown greater in-hospital mortality when compared to times to the administration of broad-spectrum antibiotics and completion of a 3-hour bundle. This has some prognostic value attached to it, as completion of the intravenous fluid bolus was not shown to affect mortality.[27] Clinical scores such as the qSOFA and SIRS scores have been able to be applied in early settings, with worsening scores corresponding to greater in-hospital mortality.[21][22] For patients requiring ICU-level care, the SOFA score has been shown to be the best predictor for in-hospital mortality.[38] Using the prognostic implications from these clinical scores can help with discussing goals of care as recommended by recent Surviving Sepsis Campaign guidelines.[8]

The large majority of clinical trials and interventions involving sepsis and septic shock address relatively short-term outcomes, typically 28 day in-hospital mortality. Long-term mortality data with respect to many of the interventions already discussed above is lacking. Mortality following hospitalization from sepsis is high, with rates reported around 31% in the first year and nearing 43% by year 2.[39] Recognized long-term complications of patients with sepsis include critical illness weakness, delirium, and acute lung injury, which have a greater impact on morbidity and mortality outside of the first 28 days.[40] Other investigations into late predictors of mortality suggest a pro-inflammatory state of accelerated atherosclerosis and chronic immunosuppression that also contributes to the high mortality rate seen in these patients.[41] Despite advances in sepsis management resulting in an improvement in short-term mortality, the prognosis for patients surviving sepsis still remains poor, with overall mortality rates remaining seemingly unaffected.