Severe sepsis is defined as the state that reflects sepsis-induced tissue hypoperfusion or multi-organ dysfunction. Usually related with hypotension, it can be occurred in normal blood pressure state in the case of partial hypoperfusion to a specific organ or toxin-induced encephalopathy. Septic shock is defined as the persistent state that systolic blood pressure<90 mm Hg (or more decreased 40 mm Hg than initial blood pressure) and mean arterial pressure (MAP)<70 mm Hg despite of appropriate volume replacement. The appropriate volume replacement is a fluid therapy with 2 L of crystalloids in healthy patients or 10 L in severe dehydrated patients. Besides 2 pair of blood culture studies before antimicrobial therapy, 1,3-β-D-glucagon assay (2B) or mannan and antimannan antibody could be helpful to diagnose invasive candidiasis assay (2C).

The effective intravenous antimicrobials, penetrate in adequate concentration into the tissue presume to be the source of sepsis, should be administrated within the first hour of recognition of septic shock. The antimicrobial regimen should be reassessed daily about effectiveness for consideration to change, discontinuation, and de-escalation (1B). Procalcitonin can be helpful to consider discontinuation of empirical antimicrobials, but only low risk of infection, the low procalcitonin level is associated with low risk of bacterial infection (2C).

Active volume resuscitation is most important role that elevates the survival rate of severe sepsis and septic shock patients as the initial management2. On this account, volume resuscitation must be done immediately and should be achieved the goals during the first 6 hours of resuscitation (1C): MAP≥65 mm Hg, central venous pressure 8-12 mm Hg (recommend 12-15 mm Hg, in the case of ventilator care or decrease ventricular compliance), hourly urine output (U/O)≥0.5 mL/kg/hr, central venous (superior vena cava) oxygen saturation (ScvO₂)≥70%, or mixed venous oxygen saturation (SvO₂)≥65%. Crystalloids are recommended as the initial volume resuscitation (1B). Hydroxyethy starch (hetastarch) is not recommended (1B). Although it is appropriate the volume of 30 mL/kg (1.5-3 L) crystalloids, active volume replacement is needed until last hemodynamic improvement (1C).

In spite of volume resuscitation, vasopressor therapy should be begun within initial 6 hours to maintain MAP (1C). Vasopressor should be administrated via central vein as possible, norepinephrine (NE) is recommended as the 1st choice drug (1C). Epinephrine can be replaced or added above NE (2B), vasopressin can be used to elevate MAP or to decrease use of NE by means of 0.03 units/min addition to NE. Dopamine is not recommended except low risk of tachyarrhythmia or bradycardia (2C).

Corticosteroids is suggested intravenous hydrocortisone alone at a dose of 200 mg/day if hemodynamic stability is not achievable despite of adequate fluid resuscitation and vasopressor therapy (2C).

Lactic acid level should be checked and normalized (2C) during initial resuscitation since some patients who elevated lactic acid level had been high mortality despite of normalized SvO₂3. Therefore, it should be considered that improve tissue perfusion by using red blood cell transfusion or dobutamin infusion when lactic acid level is over 4 mmol/L although normalized blood pressure after adequate volume resuscitation (2C).

A study reported that 77% of patients admitted in intensive care unit (ICU) experienced over moderate to severe pain14. Especially, if the pain persistent for long time, it associates with occurrence of severe degree agitation and delirium, and leads to tachyarrhythmia, hypertension, extubation failure, pneumonia, and increase mortality rate15. In alert patients, pain assessment is conducted by direct explanation of patients themselves16, but Behavioral Pain Scale (BPS)17 or Critical Care Pain Observation Tool (CPOT)18 were used in the others.

In an observational study that daily pain assessment and management effect on mechanical ventilation period, inadequate pain management cases needed more sedative drug use and had higher mortality and prolonged in-hospital stay and ventilator period. On the other hand, adequate pain management cases were associated with a shorter duration of mechanical ventilation and a reduced duration of stay in the ICU19. Thus, it is necessary to assess the pain in early period and to manage pain effectively that reduce sedative drug use and minimize delirium occurrence.

This guideline recommends the use of analgesics in an interventional treatment such as chest tube removal (+1C), and prior use of analgesics in an invasive or a pain arouse procedure (+2C). Opioids are the primary medications for managing nonneurogenic pain (+1C) and analgesics such as gabapentin and carbamazenpine can be used in neurogenic pain (+1A). Non-opiate analgesics can be used if opioids dosage is reduced (+2C).

It is necessary at times of sedation to maintain the proper level of sedation. If sedative level is too low, as anxiety and pain occurs, there is a possibility to remove such as a patient's own ventilator circuit or blood vessel equipment. On the other hand, in excessive sedative state, delirium, ventilator associated occurrence of pneumonia, extension of the mechanical ventilation time and ICU stay can be occurred.

In the guideline, pain control is recommended prior to sedation before. In an effort to adjust depending on the condition of the patient, high-frequency oscillation ventilation (HFOV) and the prone, deep sedation and neuromuscular blocking agents can be used, and in low tidal volume ventilation, can be avoid a relatively deep sedation, and after replacing endotracheal tube to tracheostomy, the agents such as analgesics and sedative drug can be reduced20.

As results of the study of mechanical ventilation withdrawal, daily interruption of sedation group was shown a significant reduction in mortality and mechanical ventilation period compare to classical management group21. This trend was more effective to be parallel with spontaneous breathing trial (SBT)22. In the PAD guideline, shallow sedation level is recommended since it improves the clinical prognosis (+1B).

As an evaluation measure of sedation, the Richmond Agitation Sedation Scale (RASS)23, Sedation Agitation Scale (SAS)24 are recommended. Although there are methods such as, Auditory Evoked potentials (AEPS), Bispectral Index (BIS), Narcotrend Index (NI), Patient State Index (PSI), and State Entropy (SE) as an measures of neurologic evaluation, they are not recommended as the primary evaluation method unless patients has been administered nerve root blocker with the exception of the case of coma, or if there is no paralysis. Sleep electroencephalogram can be used in patient with nonconvulsive seizure and is recommended If it means that electrosuppressive medication administration when elevate intracranial pressure. Non-benzodiazepines are preferred sedation agent of patients with mechanical ventilation because of low incidence of delirium to benzodiazepines (+2B).

Because the delirium in ICU is associated with increased mortality in addition to extension of admission period, and the occurrence of cognitive impairment in the future25, it is necessary to daily assessment and management (+1B). Delirium is characterized by the acute onset of cerebral dysfunction with a change or fluctuation in baseline mental status, inattention, and either disorganized thinking or an altered level of consciousness. The major risk factors of delirium are previous history of dementia, hypertension, history of alcoholism, and a severe disease on admission26. Benzodiazepine-based medicine in the ICU may be increased the risk of incidence of delirium (B) and for propofol, there are still lack of results (C). When delirium occurs, it is necessary to consider minimizing the amount of analgesics or sedative agent in use and to change. In addition, as delirium is associated to pain, it has to consider together whether pain control is successful or not. The use of dexmedetomidine can reduce the risk of delirium in mechanical ventilation patient rather than the use of benzodiazepines (B).

For evaluation of delirium, Confusion Assessment Method for the ICU (CAM-ICU)27 is mainly used. The THINK is toxic situations, hypoxemia, infection/sepsis, immobilization, nonopharmacologic interventions, and K⁺/electrolyte imbalance, they are major factors associated with delirium. As the non-medical management for delirium, early mobilization is necessary in addition to assessment and management of delirium, and visual and auditory assistance with objective indicator (+1B). Meanwhile, the prevention of delirium with drugs is not recommended yet (C). Although there is no exact evidence that haloperidol can be reduce a period of delirium, atypical antipsychotics like quetiapine can be reduce the period (C). The delirium without association to withdraw alcohol or benzodiazepines can be reduced the period by dexmedetomidine rather than benzodiazepine (+2B).

In the study, ARDS network investigators had showed significant lower mortality rate in a group with low tidal volume (VT 6 mL/kg of predicted body weight [PBW], and plateau pressure, Ppl<30 cm H₂O) than in the other group (VT 12 mL/kg of PBW, and Ppl<50 cm H₂O)30. Nowadays, it became a major option in mechanical ventilation treatment to use low tidal volume for lung-protective ventilation31.

Low positive end-expiratory pressure (PEEP) state is insufficient to keep of alveolar opening in severe respiratory distress state with pulmonary edema and end-expiratory alveolar collapse. In these situations, a low PEEP may be insufficient to stabilize alveoli and keep them open, thereby increasing the likelihood of ventilator-induced lung injury from atelectrauma. alveolar recruitment maneuver (ARM) can suppress to develop VILI theoretically32, there are debates about effect of ARM with hemodynamic unstable state, pneumothorax, and survival rate33. In the results of recent research, it is major factor of lung-protective ventilation protocol that uses low tidal volume with high PEEP34. Especially, in a study, they showed survival improvement as ARDS patients had been apply relatively higher PEEP than acute lung injury35.

As the mechanical ventilation itself can take more serious in pulmonary damage through VILI, acute lung injury (ALI)/ARDS can be occurred by mechanical ventilation of non ALI/ARDS state patients. As shown in a study that compared two groups which TV 10 mL/kg of PBW used and TV 6 mL/kg of PBW used36, and in another study that concerned about organ transplantation, compared two groups about lung harvest degree which 'TV 10-12 mL/kg of PBW, PEEP 3-5 cm H₂O' used and 'TV 6-8 mL/kg of PBW, PEEP 8-10 cm H₂O' used37, the group of low tidal volume and relative higher PEEP group had low occurrence of ALI and higher organ harvest rate available. Therefore, ALI/ARDS patients should be apply higher PEEP and non ALI/ARDS patients also should be apply about 8 cm H₂O initially and adjusted PEEP according to required FiO₂ and hemodynamic status.

Theoretically, HFOV is possible to minimize VILI38, and in a study of patients with ARDS results of the meta-analysis, it showed significant decrease in mortality. However, as the result of recent meta-analysis study, there are no significant advantages, therefore, it is not recommended as a primary treatment39.

In animal studies, prone position induces to increase homogenous ventilation and minimize VILI40. And prone position can result in improved oxygen saturation to 70% of ARDS patients41. As the result of a meta-analysis study, a prone position leads to decrease mortality rate of approximately 10% in severe ARDS (PaO₂:FiO₂ ratio<100 mm Hg) patients42.

To prevent VILI, mechanical ventilator supports can be minimized by use of extra-corporeal membrane oxygenation (ECMO) and CO₂ can be eliminated by use of extracorporeal circulation. As compared with total ECMO, partial combine mechanical ventilation with partial extracorporeal support assistance can reduce degree of lung injury because it has less complication and can make use of lower tidal volume43.

Neuromuscular blocking agents can make improvement of patient-ventilator synchrony and can be helpful management of tidal volume and airway pressure. In a study, patient with PaO₂:FiO₂ ratio<150 mm Hg who were applied neuromuscular blocking agent during 48 hours had significantly increasing survival rate in 90 days44. This effect had been significant in 16 days later and it could be interpreted that the group of neuromuscular blocking agent administration was showed low concentration of serum cytokines, that is the occurrence of less multi-organ failure to be associated with biological injury, biotrauma45.