This is the current release of the guideline.

This guideline updates a previous version: Practice guidelines for evaluating new fever in critically ill adult patients. Clin Infect Dis 1998 May;26:1042-59.

Practice guidelines for evaluating new fever in critically ill adult patients. Crit Care Med 1998 Feb;26(2):392-408.

Fever

Diagnosis
Evaluation
Management

Critical Care
Infectious Diseases
Internal Medicine
Neurology
Oncology
Pulmonary Medicine
Surgery

Physicians

• To update the practice parameters for the evaluation of adult patients who develop a new fever in the intensive care unit, for the purpose of guiding clinical practice.
• To continue to promote the rational consumption of resources and an efficient evaluation
• To assess how fevers should be evaluated in a prudent and cost-effective manner

Adult patients in the intensive care unit with new-onset fever

Evaluation of the Febrile Patient with Differential Diagnosis

1. Measurement of temperature
   • Accurate methods and devices
   • Triggers for clinical assessment of new-onset fever
2. Obtaining blood cultures
   • Timing
   • Patients with indwelling catheter
   • Disinfection practices
   • New-onset fever
3. Intravascular catheters
   • Clinical assessment of patient
   • Suspected infection
   • Cultures: blood, catheters, infusate
4. Pulmonary infection
   • Suspected lower respiratory tract infection
   • Computed tomography of the chest
   • Laboratory assessment of respiratory secretions and pleural fluid
5. Gastrointestinal tract
   • Suspected Clostridium difficile infection
   • Stool
   • Laboratory assessment: C. difficile, enzyme immunoassay (EIA) for toxin A and B, tissue culture assay, enteric pathogens, ova and parasites, norovirus
   • Flexible sigmoidoscopy (not routinely recommended)
   • Vancomycin
6. Urinary tract
   • Suspected urinary tract infection
   • Urine for microscopic exam, Gram stain, culture
7. Sinuses
   • Suspected sinusitis
   • Puncture and aspiration of fluid: Gram stain, cultures
8. Fever within 72 hours of surgery
   • Chest x-ray and urinalysis and culture considered but not recommended
   • Assessment of surgical wounds
   • Suspect deep venous thrombosis, superficial thrombophlebitis, pulmonary embolism
9. Surgical site infection
   • Assessment of surgical incision
   • Laboratory assessment of purulence and drainage
10. Central nervous system
    • Indications for imaging studies and lumbar puncture
    • Assessment of cerebrospinal fluid
11. Serum procalcitonin levels and endotoxin activity assay
12. Evaluation of new medications
13. Empirical antibiotic therapy

• Prevalence of new-onset fever
• Thresholds for diagnosis
• Prevalence of intensive care unit-acquired infection
• Prevalence of surgical site infection
• Prevalence of thrombotic conditions

Searches of Electronic Databases

The task force members provided personal experience and determined the published literature (MEDLINE articles, textbooks, etc.) from which consensus was obtained. Published literature was reviewed and classified into one of four categories, according to study design and scientific value.

Not stated

Weighting According to a Rating Scheme (Scheme Given)

1. Randomized, prospective, controlled investigation
2. Nonrandomized, concurrent, or historical cohort investigation
3. Peer-reviewed, state-of-the-art articles, review articles, editorials, or substantial case series
4. Non-peer-reviewed published opinions, such as textbook statements or official organizational publications

Review of Published Meta-Analyses
Systematic Review

Not stated

Expert Consensus (Nominal Group Technique)

The task force met twice in person, several times by teleconference, and held multiple e-mail discussions during a 2-year period to identify the pertinent literature and arrive at consensus recommendations. Consideration was given to the relationship between the weight of scientific evidence and the strength of the recommendation. Draft documents were composed and debated by the task force until consensus was reached by nominal group process.

Level 1: Convincingly justifiable on scientific evidence alone

Level 2: Reasonably justifiable by available scientific evidence and strongly supported by expert critical care opinion

Level 3: Adequate scientific evidence is lacking but widely supported by available data and expert critical care opinion

The guideline developers reviewed published cost analyses.

Not stated

Not stated

Levels of recommendations (Level 1-3) are defined at the end of the "Major Recommendations" field.

Recommendations for Measuring Temperature

1. Choose the most accurate and reliable method to measure temperature based on the clinical circumstances of the patient. Temperature is most accurately measured by an intravascular, esophageal, or bladder thermistor, followed by rectal, oral, and tympanic membrane measurements, in that order. Axillary measurements, temporal artery estimates, and chemical dot thermometers should not be used in the intensive care unit (ICU) (level 2). Rectal thermometers should be avoided in neutropenic patients (level 2).
2. Any device used to measure temperature must be maintained and calibrated appropriately, using the manufacturer's guidelines as a reference (level 2).
3. Any device used to measure temperature must be used in a manner that does not facilitate spread of pathogens by the instrument or the operator (level 2).
4. The site of temperature measurement should be recorded with the temperature in the chart (level 1).
5. A new onset of temperature of >38.3°C is a reasonable trigger for a clinical assessment but not necessarily a laboratory or radiologic evaluation for infection (level 3).
6. A new onset of temperature of <36.0°C in the absence of a known cause of hypothermia (e.g., hypothyroidism, cooling blanket, etc.) is a reasonable trigger for a clinical assessment but not necessarily a laboratory or radiologic evaluation for infection (level 3).
7. Critical care units could reduce the cost of fever evaluations by eliminating automatic laboratory and radiologic tests for patients with new temperature elevation (level 2). Instead, these tests should be ordered based on clinical assessment. A clinical and laboratory evaluation for infection, conversely, may be appropriate in euthermic or hypothermic patients, depending on clinical presentation.

Recommendations for Obtaining Blood Cultures

1. Obtain three to four blood cultures within the first 24 hrs of the onset of fever. Every effort must be made to draw the first cultures before the initiation of antimicrobial therapy. They can be drawn consecutively or simultaneously, unless there is suspicion of an endovascular infection, in which case separate venipunctures by timed intervals can be drawn to demonstrate continuous bacteremia (level 2).
2. Additional blood cultures should be drawn thereafter only when there is clinical suspicion of continuing or recurrent bacteremia or fungemia or for test of cure, 48–96 hrs after initiation of appropriate therapy for bacteremia/fungemia. Additional cultures should not be drawn as a single specimen but should always be paired (level 2).
3. For patients without an indwelling vascular catheter, obtain at least two blood cultures using strict aseptic technique from peripheral sites by separate venipunctures after appropriate disinfection of the skin (level 2).
4. For cutaneous disinfection, 2% chlorhexidine gluconate in 70% isopropyl alcohol is the preferred skin antiseptic, but tincture of iodine is equally effective. Both require >30 secs of drying time before proceeding with the culture procedure. Povidone iodine is an acceptable alternative, but it must be allowed to dry for >2 mins (level 1).
5. The injection port of the blood culture bottles should be wiped with 70 to 90% alcohol before injecting the blood sample into the bottle to reduce the risk of introduced contamination (level 3).
6. If the patient has an intravascular catheter, one blood culture should be drawn by venipuncture and at least one culture should be drawn through an intravascular catheter. Obtaining blood cultures exclusively through intravascular catheters yields slightly less precise information than information obtained when at least one culture is drawn by venipuncture (level 2).
7. Label the blood culture with the exact time, date, and anatomic site from which it was taken (level 2).
8. Draw 20 to 30 mL of blood per culture (level 2).
9. Paired blood cultures provide more useful information than single blood cultures. Single blood cultures are not recommended, except in neonates (level 2).
10. Once blood cultures have been obtained after the onset of new fever, additional blood cultures should be ordered based on clinical suspicion of continuous or recurrent bacteremia or fungemia (level 2).

Recommendations for Management of Intravascular Catheters

1. Examine the patient at least daily for inflammation or purulence at the exit site or along the tunnel, and assess the patient for signs of venous thrombosis or evidence of embolic phenomena (level 2).
2. Any expressed purulence from the insertion site should be Gram stained and cultured (level 2).
3. If there is evidence of a tunnel infection, embolic phenomenon, vascular compromise, or septic shock, the catheter should be removed and cultured and a new catheter inserted at a different site (level 2).
4. With short-term temporary catheters—peripheral venous catheters, noncuffed central venous catheters, or arterial catheters—if catheter-related sepsis (i.e., source of the infection is a colonized catheter) is considered likely, the suspect catheter or catheters should be removed and a catheter segment cultured. Blood cultures should be obtained as well. With all short-term catheters, a 5- to 7-cm intracutaneous segment should be cultured to document the source of bacteremia; with short peripheral venous or arterial catheters, the tip should be cultured; with longer central venous catheters, the intracutaneous segment and tip should be cultured; and with pulmonary artery catheters, the introducer and the pulmonary artery catheter should be cultured (level 1).
5. At least two blood cultures should be obtained. At least one blood culture should be obtained peripherally by venipuncture. One specimen should be obtained from the suspected catheter (level 1). If a quantitative culture system is available, it should be used to diagnose the catheter as the source of bacteremia/fungemia. Alternatively, differential time to positivity can be used if both blood cultures are positive for the same organism. The distal port is the logical port from which to draw cultures. When short-term, uncuffed central venous catheters are suspected of infection, it is usually more efficient to remove the existing catheter and replace it than to draw quantitative cultures (level 2).
6. Do not routinely culture all catheters removed from intensive care unit (ICU) patients. Culture only those catheters suspected of being the source of infection (level 2).
7. It is not necessary to routinely culture infusate specimens as part of the evaluation for catheter-related infections, unless there is clinical suspicion for infected infusate or blood products (level 2).

Recommendations for Evaluation of Pulmonary Infections

If a febrile patient is suspected of having a lower respiratory tract infection by clinical or radiographic assessment:

1. A chest imaging study should be obtained. In most cases, an upright portable anteroposterior chest radiograph is the most feasible study to obtain. Posterior-anterior chest radiographs with lateral view or computed tomography (CT) scan offer more information and should be obtained when clinically indicated, especially to rule out opportunistic infections in immunocompromised patients (level 1).
2. Obtain one sample of lower respiratory tract secretions for direct examination and culture before initiation of or change in antibiotics. Expectorated sputum, induced sputum, tracheal secretions, or bronchoscopic or nonbronchoscopic alveolar lavage material can be used effectively. If pneumonia is documented by physical examination and radiographic evaluation, a decision to employ bronchoscopy or other invasive diagnostic approaches should be considered based on an individual basis and the availability of local expertise (level 2).
3. Respiratory secretions obtained for microbiological evaluation should be transported to the laboratory and processed in <2 hrs (level 2).
4. Respiratory secretions that are judged to be appropriate samples by the laboratory should be evaluated by Gram-negative stain and cultured for routine aerobic and facultative bacteria. Additional stains, rapid tests, cultures, and other tests should be performed as epidemiologically appropriate (level 2).
5. Quantitative cultures can provide useful information in certain patient populations when assessed in experienced laboratories; however, quantitative cultures have not yet been sufficiently standardized nor have they been shown to alter outcome for this technique to be considered part of routine evaluation (level 2).
6. Pleural fluid should be obtained with ultrasound guidance for Gram-negative stain and routine culture (with other studies as clinically indicated) if there is an adjacent infiltrate or another reason to suspect infection and the fluid can be safely aspirated (level 2).

Recommendations for Evaluation of the Gastrointestinal Tract

If more than two stools per day conform to the container in which they are placed in a patient at risk for Clostridium difficile and if clinical evaluation indicates that a laboratory evaluation is necessary:

1. Send one stool sample for C. difficile common antigen, enzyme immunoassay (EIA) for toxin A and B, or tissue culture assay (level 2).
2. If the first specimen for C. difficile is negative and testing is performed by an EIA method, send an additional sample for C. difficile EIA evaluation. A second specimen is not necessary if the common antigen test was negative (level 2).
3. If severe illness is present and rapid tests for C. difficile are negative or unavailable, consider flexible sigmoidoscopy (level 3).
4. If severe illness is present, consider empirical therapy with vancomycin while awaiting diagnostic studies. Empirical therapy is not generally recommended if two stool evaluations are negative using a reliable assay. Although it may be more cost-effective than making the diagnosis, the empirical use of antibiotics, especially vancomycin, is discouraged because of the risk of producing resistant pathogens (level 2).
5. Stool cultures for other enteric pathogens are rarely indicated in a patient who did not present to the hospital with diarrhea or in patients who are not human immunodeficiency virus (HIV) infected. Send stool cultures for other enteric pathogens and examine for ova and parasites only if epidemiologically appropriate or evaluating an immunocompromised host (level 2).
6. Test stool for norovirus if the clinical and epidemiologic setting is appropriate. Testing for norovirus is usually only available in state laboratories and is usually performed in outbreak settings. Obtain consultation with infection control and public health authorities (level 3).

Recommendations for Evaluation of the Urinary Tract

1. For patients at high risk for urinary tract infection (kidney transplant patients, granulocytopenic patients, or patients with recent urologic surgery or obstruction), if clinical evaluation suggests a patient may have symptomatic urinary tract infection, a laboratory evaluation is necessary. Obtain urine for microscopic exam, Gram-negative stain, and culture (level 2).
2. Patients who have urinary catheters in place should have urine collected from the sampling port of the catheter and not from the drainage bag (level 2).
3. Urine should be transported to the laboratory and processed within 1 hr to avoid bacterial multiplication. If transport to the laboratory will be delayed for >1 hr, the specimen should be refrigerated. Alternatively, a preservative could be used but is less preferable to refrigeration (level 2).
4. Cultures from catheterized patients showing >10³ cfu/mL represent true bacteriuria or candiduria, but neither higher counts nor the presence of pyuria alone are of much value in determining if the catheter-associated bacteriuria or candiduria is the cause of a patient's fever; in most cases, it is not the cause of fever (level 1).
5. Gram stains of centrifuged urine will reliably show the infecting organisms and can aid in the selection of anti-infective therapy if catheter-associated urosepsis is suspected (level 1).
6. Rapid dipstick tests are not recommended for patients with urinary catheters in the analysis of possible catheter-associated infection (level 1).

Recommendations for Evaluation of the Sinuses

1. If clinical evaluation suggests that sinusitis may be a cause of fever, a computed tomography (CT) scan of the facial sinuses should be obtained (level 2).
2. If the patient has not responded to empirical therapy, puncture and aspiration of the involved sinuses under antiseptic conditions should be performed (level 2).
3. Aspirated fluid should be sent for Gram-negative stain and culture for aerobic and anaerobic bacteria and fungi to determine the causative pathogen and its antimicrobial susceptibility (level 1).

Recommendations for Evaluation of Fever Within 72 Hours of Surgery

1. A chest radiograph is not mandatory during the initial 72 hrs postoperatively if fever is the only indication (level 3).
2. A urinalysis and culture are not mandatory during the initial 72 hrs postoperatively if fever is the only indication. Urinalysis and culture should be performed for those febrile patients having indwelling bladder catheters for >72 hrs (level 3).
3. Surgical wounds should be examined daily for infection. They should not be cultured if there is no symptom or sign suggesting infection (level 2).
4. A high level of suspicion should be maintained for deep venous thrombosis, superficial thrombophlebitis, and pulmonary embolism, especially in patients who are sedentary, have lower limb immobility, have a malignant neoplasm, or are taking an oral contraceptive (level 2).

Recommendations for Evaluation of Surgical Site Infection

1. Examine the surgical incision at least once daily for erythema, purulence, or tenderness as part of the fever evaluation (level 2).
2. If there is suspicion of infection, the incision should be opened and cultured (level 2).
3. Gram-negative stain and cultures should be obtained from any expressed purulence obtained from levels within the incision consistent with a deep incisional or organ/space surgical site infection. Tissue biopsies or aspirates are preferable to swabs (level 3).
4. Drainage from superficial surgical site infections may not require Gram-negative stain and culture because incision, drainage, and local care may be sufficient treatment and antibiotic therapy may not be required. Superficial swab cultures are likely to be contaminated with commensal skin flora and are not recommended (level 2).
5. Standard guidelines should be used to define burn wound infection (level 3).

Recommendations for Evaluation of Central Nervous System Infections

1. If altered consciousness or focal neurologic signs are unexplained, lumbar puncture should be considered in any patient with a new fever, unless there is a contraindication to lumbar puncture (level 3).
2. For a patient with a new fever and new focal neurologic findings suggesting disease above the foramen magnum, an imaging study is usually required before lumbar puncture. If a mass is present, neurology/neurosurgery consultation is required to determine the optimal diagnostic approach (level 2).
3. In febrile patients with an intracranial device, cerebrospinal fluid (CSF) should be obtained for analysis from the CSF reservoir. If CSF flow to the subarachnoid space is obstructed, it may be prudent to also obtain CSF from the lumbar space (level 3).
4. In patients with ventriculostomies who develop stupor or signs of meningitis, the catheter should be removed and the tip cultured (level 3).
5. CSF should be evaluated by Gram-negative stain and culture, glucose, protein, and cell count with differential. Additional tests for tuberculosis, viral and fungal disease, neoplasia, etc., should be performed as dictated by the clinical situation (level 2).

Recommendation for Using Biomarkers to Determine the Cause of Fever

1. Serum procalcitonin levels and endotoxin activity assay can be employed as an adjunctive diagnostic tool for discriminating infection as the cause for fever or sepsis presentations (level 2).

Recommendations for Recognizing Noninfectious Causes of Fever

1. Consider all new medications and blood products the patient has received. Ideally, if the suspected drug can be stopped, do so. If the drug cannot be stopped, consider a comparable substitute (level 2).
2. Fever induced by drugs may take several days to resolve. Establishing a temporal relationship between fever and the offending agent may be helpful in establishing the diagnosis (level 3).

Recommendations for Empiric Therapy of Fever

1. When clinical evaluation suggests that infection is the cause of fever, consideration should be given to administering empirical antimicrobial therapy as soon as possible after cultures are obtained, especially if the patient is seriously ill or deteriorating (level 1).
2. Initial empirical antibiotic therapy should be directed against likely pathogens, as suggested by the suspected source of infection, the patient risk for infection by multidrug-resistant pathogens, and local knowledge of antimicrobial susceptibility patterns (level 1).

Definitions:

Level 1: Convincingly justifiable on scientific evidence alone

Level 2: Reasonably justifiable by available scientific evidence and strongly supported by expert critical care opinion

Level 3: Adequate scientific evidence is lacking but widely supported by available data and expert critical care opinion

None provided

The type of evidence supporting the recommendations is not specifically sated for each recommendation.

Appropriate and efficient evaluation of new fever in critically ill adult patients that promotes the rational consumption of resources

• Esophageal probes are uncomfortable in alert or spontaneously breathing patients. The theoretical risk of an esophageal probe eroding or perforating the esophagus when left in place for extended periods of time makes this probe impractical for use in the critically ill patient.
• The patient often perceives rectal temperature measurement as unpleasant and intrusive. Moreover, there is a small risk of trauma or perforation to the rectum, which is a particular problem in patients who are neutropenic, coagulopathic, or who have had recent rectal surgery. Rectal temperature measurements have also been implicated in spreading enteric pathogens such as Clostridium difficile or vancomycin-resistant enterococci via the device or the operator.
• Oral probes can damage oral mucosa, especially in patients with abnormal mucosa due to trauma, thermal injury, infection, surgery, cancer, or cytotoxic drugs.
• Direct measurement of the tympanic membrane temperature requires an electronic probe, is painful in awake patients, and risks trauma to the tympanic membrane.
• There is concern that saline used to obtained a sputum sample dilutes the specimen and could introduce pathogens present in the tube biofilm or upper airway into the lower airway.
• The empirical use of antibiotics, especially vancomycin, is discouraged because of the risk of producing resistant pathogens.
• False positive and false negative laboratory results due to contamination of diagnostic specimens

The usual contraindications to lumbar puncture detected by computed tomography scanning include lateral shift of midline structures, loss of the suprachiasmatic and basilar cisterns, obliteration of the fourth ventricle, or obliteration of the superior cerebellar and quadrigeminal plate cisterns with sparing of the ambient cisterns.

An implementation strategy was not provided.

Getting Better

Effectiveness

Not applicable: The guideline was not adapted from another source.

1998 (revised 2008 Apr)

Infectious Diseases Society of America - Medical Specialty Society
Society of Critical Care Medicine - Professional Association

Infectious Diseases Society of America
Society of Critical Care Medicine

Task Force of the Society of Critical Care Medicine and the Infectious Diseases Society of America

Task Force Members: Naomi P. O'Grady, MD; Philip S. Barie, MD, MBA, FCCM; John G. Bartlett, MD; Thomas Bleck, MD, FCCM; Karen Carroll, RN; Andre C. Kalil, MD; Peter Linden, MD; Dennis G. Maki, MD; David Nierman, MD, FCCM; William Pasculle, MD; Henry Masur, MD, FCCM

Dr. Bartlett holds consultancies with HIV-Bristol-Myers, Abbott, Merck, Johnson & Johnson, and Tibotec; and a patent with Gilead. The remaining authors have not disclosed any potential conflicts of interest.

This is the current release of the guideline.

This guideline updates a previous version: Practice guidelines for evaluating new fever in critically ill adult patients. Clin Infect Dis 1998 May;26:1042-59.

Practice guidelines for evaluating new fever in critically ill adult patients. Crit Care Med 1998 Feb;26(2):392-408.

None available

None available

This summary was completed by ECRI on January 15, 1999. The information was verified by the guideline developer as of March 22, 1999. This summary was updated by ECRI Institute on September 10, 2008. The updated information was verified by the guideline developer on October 20, 2008.

This NGC summary is based on the original guideline, which is subject to the guideline developer's copyright restrictions.