Critical Care Medicine: Common ICU Conditions — Shock and Sepsis

Shock: Inadequate Tissue Perfusion

Shock exists whenever systemic tissue perfusion is insufficient to meet cellular oxygen needs. In its early phase, cells switch from aerobic to anaerobic metabolism and the resulting tissue injury is still reversible; once compensatory mechanisms are exhausted, sustained hypoperfusion progresses to cellular dysfunction and death. Four primary mechanisms produce shock — reduced circulating volume (hypovolemic), pump failure (cardiogenic), circulatory obstruction (obstructive), and inappropriate vasodilation (distributive) — and they frequently overlap in the same patient.

Hypotension alone does not diagnose shock. Blood pressure must be interpreted together with examination findings that suggest end-organ hypoperfusion (Table 1), and the trend relative to the patient’s own baseline is more informative than any single absolute value: a chronically hypertensive patient can be in shock at pressures that appear acceptable, while others tolerate long-standing low pressures without organ injury. Compensatory tachycardia may be absent in patients with conduction disease or on nodal-blocking drugs, and tachypnea is often the earliest vital-sign clue as the body compensates for a developing metabolic acidosis.

Initial Evaluation of the Patient in Shock

A structured bedside evaluation proceeds through four steps:

• Look for clinical signs of organ hypoperfusion — altered mentation, cool extremities or skin mottling, prolonged capillary refill, decreased urine output, absent bowel sounds.
• Assess hemodynamics — SBP <90 mm Hg or MAP <70 mm Hg, tachycardia.
• Check laboratory markers of tissue hypoperfusion — lactate, serum creatinine, serum pH, serum troponin.
• Perform initial ancillary testing — blood, sputum, wound, or urine cultures; complete blood count; electrocardiogram; point-of-care ultrasound (echocardiography, lung, vascular); directed radiography.

A reasonable laboratory panel for undifferentiated shock also includes renal and liver function tests, cardiac enzymes, coagulation studies, and a blood gas (venous sampling suffices for assessing acidosis when hypoxemia is absent). Serial lactate measurements both risk-stratify patients and track the response to therapy. The ECG can reveal the culprit immediately — a brady- or tachyarrhythmia, ST-elevation infarction, the S1Q3T3 pattern of pulmonary embolism, or low voltage with electrical alternans suggesting tamponade. Chest radiography separates infectious infiltrate from pulmonary edema and confirms or excludes pneumothorax.

Point-of-care ultrasound (POCUS) is increasingly used to evaluate shock. Structured protocols (RUSH, ACES, SESAME) rapidly survey cardiac function, volume status (a collapsible inferior vena cava in a spontaneously breathing patient suggests reduced intravascular volume), pleural and abdominal collections, and deep-vein thrombosis. Interpretation is operator-dependent, bedside scanning does not replace formal echocardiography, and a randomized trial of protocolized POCUS in undifferentiated shock showed no survival benefit — it is most valuable when the result will immediately change management. Echocardiography also remains useful for following cardiac response to treatment.

Classification and Causes of Shock

Each shock type has a characteristic hemodynamic profile, and matching the history, examination, and testing to that profile is what directs therapy (Table 2).

Distributive Shock

Blood volume is generally normal, but vasodilation and microvascular dysfunction produce a state of “relative” hypovolemia; cardiac output is often elevated yet cannot sustain an adequate perfusion pressure. Sepsis is the most common cause, but anaphylaxis, vasodilator drugs, and high spinal injury present similarly. Initial treatment is fluid resuscitation, with vasopressors added if fluids alone cannot support perfusion; the agent is chosen according to the underlying cause and the patient’s comorbidities.

Hypovolemic Shock

Decreased intravascular volume lowers preload and ventricular filling, diminishing stroke volume; tachycardia and peripheral vasoconstriction initially preserve perfusion of vital organs but fail as hypovolemia worsens. Treatment is aggressive volume or blood-product replacement together with control of any bleeding source. Crystalloid restores hemodynamic stability initially, but erythrocyte transfusion may ultimately be needed to prevent tissue ischemia. In stable ICU patients a hemoglobin threshold above 7 g/dL (70 g/L) is appropriate, though actively bleeding patients in shock and those with underlying cardiovascular disease may warrant higher thresholds. In severe trauma with massive transfusion requirements and coagulopathy, early resuscitation with erythrocytes, platelets, and fresh frozen plasma in a 1:1:1 ratio is supported by evidence.

Cardiogenic Shock

Myocardial infarction is the most common cause. The compensatory tachycardia and vasoconstriction of shock raise afterload and myocardial oxygen demand, which can further depress cardiac output — a vicious cycle. Initial evaluation includes a focused examination, laboratory testing for ischemia and heart failure, chest radiograph, ECG, and transthoracic echocardiogram. Bradyarrhythmias may require external or transvenous pacing, and ST-elevation infarction with laboratory evidence of coronary ischemia calls for emergent percutaneous intervention or fibrinolysis. Although not firmly evidence-based, mechanical circulatory support (ECMO, intra-aortic balloon pump, ventricular assist devices) can serve as a bridge to definitive therapy, and occasionally as destination therapy.

Obstructive Shock

Extracardiac forces impair cardiac output despite normal cardiac function, and hemodynamics often mimic cardiogenic shock. Clinical clues vary by cause: pulmonary embolism may show raised right atrial pressure (jugular venous distention, lower-extremity edema) with a clear lung exam; tamponade produces jugular venous distention, pulsus paradoxus, and distant heart sounds; tension pneumothorax produces absent breath sounds on the affected side. Lung ultrasound rapidly screens for pneumothorax and echocardiography identifies features of embolism or pericardial effusion; for constrictive or restrictive pericarditis and pulmonary arterial hypertension, invasive hemodynamic measurement may be needed. Most causes are rapidly reversible once identified.

Sepsis: Definition and Recognition

The Sepsis-3 Definition

The Third International Definitions for Sepsis and Septic Shock (Sepsis-3) define sepsis as life-threatening organ dysfunction caused by a dysregulated host response to infection. Any pathogen in any organ can trigger it, and the infection need not be disseminated. The definition marks a deliberate shift away from the older SIRS criteria — which were sensitive but poorly specific and also captured countless noninfectious inflammatory states — and it eliminates the former “severe sepsis” category. Pathophysiologically, the syndrome reflects loss of balance between proinflammatory and anti-inflammatory mediators, unregulated coagulation in the microvasculature, and dysfunction ranging from subcellular mitochondrial injury to whole organ-system failure.

Operationally, sepsis is identified when infection is known or suspected and organ-dysfunction criteria are met. The recommended instrument is the Sequential Organ Failure Assessment (SOFA) score, which grades six organ systems (respiratory, coagulation, hepatic, cardiovascular, neurologic, renal) from 0 to 4. A SOFA score of 2 or higher — or a rise of 2 or more points — marks acute organ dysfunction and predicts hospital mortality above 10%; SOFA is intended for use in ICU patients.

The qSOFA Screening Score

For rapid bedside screening outside the ICU, Sepsis-3 proposed the quick SOFA (qSOFA) score, which uses just three clinical criteria (Table 3). A qSOFA score of 2 or more in the setting of known or suspected infection predicts increased mortality and should prompt evaluation for resuscitation and consideration of ICU admission (Table 4).

Epidemiology

Estimating the burden of sepsis is difficult because definitions have evolved, recognition is inconsistent, and reporting is not standardized, but the numbers are large. Globally there are roughly 49 million cases and 11 million deaths each year — about one in five of all deaths worldwide, the vast majority in low- and middle-income countries. In the United States, sepsis complicates about 6% of adult hospitalizations (~1.7 million cases per year) with overall mortality near 15%, rising to 40% in septic shock. Incidence is not evenly distributed: sepsis is more common among Black men than other racial groups or women, and it becomes steadily more common with age, with incidence rising each year after 65. A septic patient’s mortality is at least four times that of the same underlying condition without sepsis, and each additional sepsis-related organ failure adds roughly 15% to mortality. Most cases are community-onset, the urinary and respiratory tracts are the dominant sources, and common pathogens include Staphylococcus aureus, streptococci, and enterococci among gram-positive and Escherichia coli, Klebsiella, and Pseudomonas among gram-negative organisms.

Management of Sepsis

The Hour-1 Bundle

Sepsis guidelines have long emphasized that timely care saves lives. The original 3-hour and 6-hour bundles were consolidated in 2018 into a single 1-hour sepsis bundle: obtain blood cultures, measure serum lactate, administer broad-spectrum antibiotics, and begin fluid or vasopressor resuscitation in patients with hypotension or signs of hypoperfusion. Some elements rest on weak evidence and remain controversial, but the bundle’s intent captures the two pillars of sepsis management — supporting organ perfusion and function, and controlling the infection.

Initial Resuscitation

Sepsis depresses preload (capillary leak), contractility, and vascular tone; patients may present with profound hypotension requiring large-volume resuscitation. Current guidance recommends an initial bolus of 30 mL/kg of crystalloid within the first hour when hypotension or a lactate of at least 4 mmol/L is present; lactate levels above 2 mmol/L should be rechecked within 2 to 4 hours to confirm improvement. The adequacy of resuscitation and the need for further fluid or vasopressors then require clinical judgment based on comorbidities and the available data — persistent hypotension or hypoperfusion calls for additional fluids or vasopressors.

Vasopressors

When vasopressor support is required, norepinephrine is the first-line agent because it raises systemic vascular resistance without compromising myocardial performance (stroke volume and cardiac output). If norepinephrine alone is insufficient, epinephrine, vasopressin, or dobutamine may be added. If possible, all patients on vasopressors should have an arterial catheter for continuous blood-pressure monitoring.

Antibiotic Therapy

Early antibiotics are crucial: broad-spectrum therapy should be given within the first hour of suspected sepsis, and every hour of delay (or an inappropriate regimen) increases mortality — in bacterial septic shock, each hour’s delay is associated with roughly a 7–8% rise in mortality. Empiric coverage must reach the most likely pathogens — gram-negative, gram-positive, and mixed bacterial causes — without excluding viral or fungal therapy when warranted. Choice of regimen should account for the likely source, local resistance patterns and facility antibiograms, and the patient’s history (indwelling catheters, neutropenia, immunosuppression, recent antibiotics or health-care exposure, which all raise resistance risk). Guidelines stop short of naming specific drugs, but a carbapenem or an extended-range penicillin or β-lactamase inhibitor is a safe starting point for most septic patients.

In septic shock (but not sepsis without shock), combination therapy with at least two antibiotics from different classes should target the most likely bacterial pathogen; specific multidrug-resistance risks justify empiric combination gram-negative coverage or targeted MRSA therapy. Once a pathogen is identified, therapy is narrowed to the most effective targeted regimen; when cultures are negative, de-escalation is generally recommended within the first few days if the patient is improving clinically. Typical duration is 7 to 10 days, with continual reassessment for opportunities to shorten or stop. In neutropenic patients, antibiotics continue until the planned course is complete or the neutropenia resolves (neutrophils >500/μL [0.5 × 10⁹/L]), whichever is later.

Source control is equally essential: potential sources of infection must be identified and controlled promptly — drainage of abscesses, removal of intravenous catheters suspected of being infected, and urgent procedural or surgical control of amenable foci. One exception is necrotizing pancreatitis, where definitive resection or drainage should be delayed until the extent of necrosis is clear.

Procalcitonin, a serum marker of bacterial infection, may support shortening antibiotic duration in sepsis or stopping therapy in patients initially thought to have sepsis who turn out not to. Meta-analyses show procalcitonin-guided algorithms reduce overall antibiotic use, though mortality benefit remains unproven and practice varies.

Monitoring

Continuous arterial pressure monitoring via arterial catheter is useful for any patient receiving vasopressors. By contrast, several large trials and meta-analyses show no benefit to routine pulmonary artery catheter use in septic shock — catheterized patients in many studies had higher resource use, more complications, and in some cases higher mortality.

Glucocorticoids

Glucocorticoids are suggested when hemodynamic stability cannot be restored with intravenous fluids and vasopressors. Although a clear mortality benefit has not been shown, recent trials associate glucocorticoids with faster shock reversal, fewer days of organ failure, and fewer days of mechanical ventilation. The Surviving Sepsis Campaign recommends hydrocortisone 200 mg/day for refractory shock, and SCCM/ESICM (2017) set a maximum of 400 mg hydrocortisone daily. An ACTH stimulation test is not recommended, and glucocorticoids have no role in sepsis without shock.

Nutrition and Glucose Control

Patients with sepsis or septic shock should receive early oral feeding when possible and are considered at risk for malnutrition once the ICU stay exceeds 48 hours. If oral intake is not feasible, early enteral nutrition is indicated to limit the infectious complications of delayed enteral feeding. When neither oral nor enteral nutrition is possible, the optimal timing of parenteral nutrition is uncertain; guidelines suggest ramping calories and protein up to goal gradually over 3 to 7 days to avoid refeeding syndrome and overfeeding.

Glucose control matters, but the target is not tight normalization: in critically ill medical ICU patients, tight control aiming for 80–110 mg/dL produced higher rates of severe hypoglycemia (<40 mg/dL) — sometimes fatal — than more liberal targets of 140–180 or 180–200 mg/dL. The Surviving Sepsis Campaign recommends an upper glucose limit below 180 mg/dL.

Sepsis: Final Key Points