Protocol for Fever Management in Hospitalized Adults
For adult hospital patients with new fever, prioritize accurate temperature measurement using central methods (bladder catheter, esophageal thermistor) or oral/rectal routes, perform targeted diagnostic workup based on clinical assessment rather than reflexive testing, and use antipyretics primarily for symptomatic relief rather than temperature reduction alone. 1, 2
Temperature Measurement
Use the most reliable method available for accurate fever detection:
- Central monitoring is preferred when devices are already in place: pulmonary artery catheter thermistors, bladder catheters, or esophageal balloon thermistors 1, 2
- Oral or rectal temperatures are recommended when central monitoring is unavailable 1, 2
- Avoid unreliable methods including axillary, tympanic membrane, temporal artery thermometers, or chemical dot thermometers for clinical decision-making 1, 2
- Fever is defined as a single temperature measurement ≥38.3°C (101°F) 2
Initial Clinical Assessment
Conduct a focused history and physical examination to identify potential infectious and non-infectious sources:
- Evaluate for infectious causes: pneumonia (most common in ICU), urinary tract infection, catheter-related bloodstream infection, surgical site infection, sinusitis 1, 2
- Consider non-infectious causes: drug fever, venous thrombosis, pulmonary embolism, acalculous cholecystitis, pancreatitis, acute myocardial infarction, stroke, malignancy, transfusion reactions 1
- Not all fevers require investigation—those with obvious non-infectious etiology (e.g., immediate postoperative period) may not need extensive workup 1
Diagnostic Workup
Tailor investigations based on clinical suspicion rather than ordering reflexive "fever workup":
Imaging Studies
- Chest radiograph is recommended for all ICU patients with new fever, as pneumonia is the most common infection causing fever in critically ill patients 3, 2
- CT imaging should be performed for post-surgical patients (thoracic, abdominal, or pelvic surgery) with persistent fever when initial workup is non-diagnostic 1, 2
- Bedside diagnostic ultrasound of the abdomen is recommended for patients with recent abdominal surgery or abdominal symptoms 1, 2
Microbiological Studies
- Blood cultures should be obtained before initiating antimicrobial therapy, particularly if the patient is seriously ill or deteriorating 2
- Targeted cultures based on suspected source (sputum, urine, wound, catheter tip) rather than reflexive pan-culturing 1
- Rapid diagnostic testing strategies are recommended when available 1
Laboratory Tests
- Biomarkers (procalcitonin, C-reactive protein) can assist in guiding discontinuation of antimicrobial therapy 1
- Consider complete blood count, metabolic panel, liver function tests, lactate dehydrogenase based on clinical context 1
Antipyretic Therapy
Antipyretics should be used primarily for symptomatic relief and patient comfort, not routinely to reduce temperature:
First-Line Treatment
- Paracetamol (acetaminophen) 1000 mg orally every 4-6 hours (maximum 4 g/day) is the first-line antipyretic for patients capable of oral intake 3, 4
- Paracetamol has superior cardiovascular and gastrointestinal safety compared to NSAIDs 3
- Paracetamol/ibuprofen combination (500 mg/150 mg) may be more effective than paracetamol alone for bacterial fever at 1 hour, though both are equally effective at 2 hours 4
Route Selection
- Oral route is preferred for all patients capable of oral intake 3
- Intravenous paracetamol is preferable when IV access exists and patients cannot take oral medications due to persistent vomiting, altered mental status, or NPO status 3
- Avoid intramuscular route due to injection site pain, tissue trauma, and risk of hematoma (especially in anticoagulated patients) 3
Dosing Considerations
- Reduce dosage in patients with hepatic insufficiency or history of alcohol abuse 3
- Contraindicated in acute liver failure 3
Evidence on Mortality
- Antipyretic therapy does not improve mortality outcomes: A meta-analysis of 13 RCTs (n=1,963) showed no improvement in 28-day mortality (RR 1.03; 95% CI 0.79-1.35), hospital mortality, or shock reversal 3, 5
- Therefore, treat for comfort, not the thermometer reading 3, 2
Non-Pharmacologic Measures
Physical cooling methods are generally not recommended:
- Avoid routine physical cooling (tepid sponging, fanning) as these cause discomfort without improving outcomes 3
- Cooling devices should only be considered for refractory fevers unresponsive to antipyretics 3
- Environmental measures are acceptable: reduce excessive stimuli, lower ambient temperature during hot months, uncover patient as tolerated 3
- Maintain head of bed elevation 15-30° to prevent aspiration 3
Special Populations and Contexts
Post-Surgical Patients
- Fever immediately postoperatively typically does not require investigation 1
- Persistent fever beyond several days without identified cause warrants CT imaging of the surgical area in collaboration with surgical service 3, 2
Critically Ill/ICU Patients
- Fever occurs in 26-88% of adult ICU patients depending on definition and cohort 1
- Early antimicrobial therapy may improve outcomes when infection is suspected—do not delay effective treatment 2
- Diagnostic studies should focus on potential sources identified by history and examination rather than reflexive testing 1
Neutropenic Patients
- Hospitalization and empiric antibacterial therapy with vancomycin plus antipseudomonal antibiotics is recommended for neutropenic fever 2
- Consider discontinuation at 72 hours in low-risk patients with negative cultures who have been afebrile ≥24 hours 2
Common Pitfalls to Avoid
- Do not treat the thermometer reading rather than the patient's symptoms and comfort 2
- Do not use unreliable temperature methods (tympanic, temporal) for critical decisions 2
- Do not delay antimicrobial therapy when infection is suspected—this increases mortality 2
- Do not order reflexive pan-cultures without clinical suspicion of specific sources 1
- Do not use antipyretics routinely just to reduce temperature in the absence of patient discomfort 3, 2