In assessing a patient with fever the following questions must be considered.
1. What kind of fever is it?
2. When did it start?
3. Did it commence abruptly at a particular clinical event?
5. Is it constant or intermittent?
The chronology of a febrile episode is particularly important. If the patient had been in hospital for some time before admission to the ICU, temperature charts from that time period or any recent previous admissions should be reviewed and collated with the present charts. The pattern of fever may be helpful: traditionally, a high 'spiking' fever is associated with loculated pus and chronic low-grade fever may be associated with the presence of 'low-grade' pathogens such as coagulase-negative staphylococci on intravenous cannulas or prosthetic devices.
The recent in-patient history should be reviewed. Vital clues may be found in reviewing the true sequence of events, not just from clinical records but also from the patient if possible, the relatives, and other staff. This, in turn, should be related to the temperature chart. Attention should be given to the timing of transfusion with blood or blood products. Transfusion reactions are commonly associated with fever but such febrile episodes may be delayed.
Careful examination may reveal oral or nasal trauma or sepsis, lymphadenopathy, added pulmonary sounds, the appearance of a new murmur, abdominal mass or tenderness, or a swollen joint. Particular attention should be paid to the skin around line sites for signs of inflammation or pus, to pressure areas for necrosis of underlying tissues, and to limbs for evidence of deep venous thrombosis.
Fever is a common component of adverse drug reactions. It is often accompanied by a skin rash. The timing of the fever relative to changes in therapy is of particular note, and careful assessment of any changes in therapy should be related to the temperature chart.
When presented with a febrile patient it is tempting to initiate a batch of investigations as a 'fever screen'. Such an approach is not always the most appropriate. Firstly, fever is such a common phenomenon in ICU patients that it may not always represent infection. Secondly, non-targeted diagnostic tests tend to have a very low positive yield. In one study of 464 patients following major abdominal surgery, culture-proven infection was the cause of fever in 27 per cent. Of these infections, 74 per cent could be diagnosed from clinical findings and a single appropriate test. The yields from 'routine' blood, sputum, and urine cultures were particularly low, being 5 per cent, 9 per cent, and 10 per cent respectively. Thus diagnostic tests should be targeted according to clinical suspicion and therapeutic opportunity.
Most ICU-related infections are acquired from the hosts themselves. Microbiological screening is practiced in some units, and any information on colonization patterns of the individual patient should be reviewed. Infection may also be acquired from the ICU environment. Information regarding local colonization patterns and sensitivities, particularly for organisms such as Pseudomonas species, should be collected regularly and reviewed in the context of the individual patient.
Urinary tract infection accounts for about 40 per cent of nosocomial infections in ICU patients. It is easy to forget simple tests such as dipstix urinalysis. Catheter-related urinary infection is unlikely before 48 h but thereafter becomes progressively more likely.
Sputum Gram staining is rapid and may be useful, but it may be difficult to distinguish upper respiratory flora from pathogens until culture results are available. Less contaminated specimens may be achieved by bronchoscopic lavage. Colonization of the upper airways tends to occur after 48 h of intubation.
Blood cultures should be drawn during the febrile episode and wound swabs should be taken where appropriate. Central venous catheters are a potential site of infection. Although colonization is common, central venous catheters are unlikely to represent the source of fever if they have been in place for less than 48 h unless placement is known to be contaminated. Infection becomes more common with length of placement and length of stay in the ICU; in one study, 75 per cent of catheter-related infections occurred after the sixth day in the ICU. 'Through-line' blood cultures or the use of intravenous catheter microbiological 'brushes' may indicate the presence of a colonized catheter, and swabs from around an inflamed catheter entry site may reveal the pathogen. The cessation of fever after guidewire exchange of central venous catheters is perhaps the best diagnostic test and therapeutic maneuver simultaneously. Arterial cannulas appear to be less susceptible to infection. In a recent study no positive cultures were found until more than 96 h from the time of placement. Central line tips may be cultured but, if this is requested, care should be taken to ensure skin cleansing around the insertion site before the catheter is withdrawn to reduce contamination.
Candida infection of catheters is particularly hazardous because of the high mortality associated with candidal septicemia and should be considered, particularly if there is candidal colonization elsewhere such as in the oral cavity.
Chest radiography should be routine in ICU patients with fever and should normally be performed if they have been intubated.
Ultrasound examination is particularly useful for investigation of the abdomen, particularly in the right and left upper quadrants and the pelvis, and has the advantage of portability. Postsurgical collections may represent abscess or sterile fluid collections, and diagnostic guided aspiration may be performed for microbiological cultures and to determine if formal drainage is indicated. CT scanning may be able to identify collections not seen on ultrasound but is logistically more difficult.
Intracranial infection is best sought by CT scanning, and this should certainly be performed if lumbar puncture for cerebrospinal fluid examination is proposed because of the risks associated with the presence of cerebral abscess.
Scintigraphy can be used to identify localized sepsis not apparent on other investigations. Gallium citrate demonstrates collections of degranulating leukocytes by binding to released lactoferrin. Leukocytes can be labeled with indium-111 or technetium-99, and labeled antibody techniques are being developed. Scintigraphy has a low diagnostic yield when used as a 'blind' investigation but may disclose occult infection such as osteomyelitis. It is of value in determining whether structural lesions seen with other imaging represent inflammation, for example whether or not a collection seen on CT represents an abscess ( Davis..aDd.EiOkii®nD®it,.1994).
Measurement of activation of cytokines is not sufficiently specific to distinguish infective from non-infective causes of fever, but this may become a possibility in the future.
Several metabolic conditions result in fever, including Addison's disease (acute adrenal insufficiency), thyrotoxicosis, pheochromocytoma, and malignant neuroleptic syndrome. Salicylate poisoning and anticholinergic overdose are both associated with fever, as are venous thrombosis and pulmonary embolism.
Intracranial lesions such as emboli, tumors, head injury, or neurosurgery can cause 'neurogenic hyperthermia'. This occurs in the absence of elevated systemic cytokines and is thought to be due to direct effects on the hypothalamus. It tends to be intense and short lived, although a low-grade elevation of temperature may persist for several days. This may represent a true elevation of the set-point and be mediated by prostaglandin synthesis; it is suppressed by treatment with indomethacin (Mojtz_1993).
Was this article helpful?
The term vaginitis is one that is applied to any inflammation or infection of the vagina, and there are many different conditions that are categorized together under this ‘broad’ heading, including bacterial vaginosis, trichomoniasis and non-infectious vaginitis.