Pyrexia - PACT - ESICM

AN ESICM MULTIDISCIPLINARY DISTANCE LEARNING PROGRAMME 

FOR INTENSIVE CARE TRAINING 

Pyrexia 

Clinical problems 

Update 2011 (pdf) 

Module Authors (Update 2011) 

George DIMOPOULOS 

AB Johan GROENEVELD 

University Hospital Attikon Medical School, University 

of Athens, Athens, Greece 

Dept of Intensive Care, Free University Hospital, 

Amsterdam, the Netherlands 

Module Authors (first edition) 

Roberto FUMAGALLI 

AB Johan GROENEVELD 

Dept of Anaesthesiology & ICU, Azienda Osp, Riuniti di 

Bergamo, Bergamo, Italy 

Dept of Intensive Care, Free University Hospital, 

Amsterdam, the Netherlands 

Module Reviewers 

Section Editor 

Pedro Póvoa and Janice Zimmerman 

AB Johan Groeneveld

Pyrexia 

Update 2011 (pdf) 

Editor-in-Chief 

Deputy Editor-in-Chief 

Medical Copy-editor 

Self-assessment Author 

Editorial Manager 

Business Manager 

Chair of Education and Training 

Committee 

Dermot Phelan, Intensive Care Dept, 

Mater Hospital/University College Dublin, Ireland 

Francesca Rubulotta, Imperial College, Charing 

Cross Hospital, London, UK 

Charles Hinds, Barts and The London School of 

Medicine and Dentistry 

Hans Flaatten, Bergen, Norway 

Kathleen Brown, Triwords Limited, Tayport, UK 

Estelle Flament, ESICM, Brussels, Belgium 

Marco Maggiorini, Zurich, Switzerland 

PACT Editorial Board 

Editor-in-Chief 

Deputy Editor-in-Chief 

Respiratory failure 

Cardiovascular critical care 

Neuro-critical care and Emergency 

medicine 

Critical Care informatics, management 

and outcome 

Environmental hazards and 

Obstetric critical care 

Infection/inflammation and Sepsis 

Kidney Injury and Metabolism. 

Abdomen and nutrition 

Peri-operative ICM/surgery and 

imaging 

Professional development and Ethics 

Education and assessment 

Consultant to the PACT Board 

Dermot Phelan 

Francesca Rubulotta 

Anders Larsson 

Jan Poelaert/Marco Maggiorini 

Mauro Oddo 

Carl Waldmann 

Janice Zimmerman 

Johan Groeneveld 

Charles Hinds 

Torsten Schröder 

Gavin Lavery 

Lia Fluit 

Graham Ramsay 

Copyright© 2011. European Society of Intensive Care Medicine. All rights reserved.

LEARNING OBJECTIVES 

After studying this module on Pyrexia, you should be able to: 

1. Assess fever in the ICU and initiate an appropriate evaluation 

2. Determine common causes of fever in the critically ill patient 

3. Manage special forms of fever 

4. Decide how and when to treat fever 

FACULTY DISCLOSURES 

The authors of this module have not reported any disclosures. 

DURATION 

7 hours 

Copyright©2011. European Society of Intensive Care Medicine. All rights reserved.

INTRODUCTION ............................................................................................................................................. 1 

1/ ASSESSING AND MEASURING FEVER IN ICU ............................................................................................... 3 

Assessment of fever of recent onset .................................................................................................................. 3 

Clinical appraisal ................................................................................................................................................. 4 

Fever – notable features and measurement ...................................................................................................... 6 

Laboratory appraisal .......................................................................................................................................... 8 

Imaging ............................................................................................................................................................... 9 

Culture techniques ............................................................................................................................................. 9 

Microbiology .................................................................................................................................................... 10 

Systemic inflammatory response syndrome (SIRS) .......................................................................................... 12 

2/ DETERMINING THE CAUSE OF FEVER IN THE CRITICALLY ILL PATIENT ....................................................... 14 

Infective causes ................................................................................................................................................ 15 

Ventilator‐associated pneumonia ............................................................................................................... 15 

Central venous catheter‐related infections ................................................................................................. 18 

Sinusitis ........................................................................................................................................................ 23 

Urinary tract infections ................................................................................................................................ 25 

Acute acalculous cholecystitis ..................................................................................................................... 26 

Other causes ................................................................................................................................................ 26 

Non‐infective causes ........................................................................................................................................ 27 

3/ FEVER IN SPECIFIC CATEGORIES OF CRITICALLY ILL PATIENT ..................................................................... 30 

The surgical critical care patient – determining the cause of fever ................................................................. 30 

Wound infection .......................................................................................................................................... 31 

The abdomen ............................................................................................................................................... 32 

Fever in immunocompromised patients .......................................................................................................... 33 

Fever in neurological disease ........................................................................................................................... 35 

Identifying special forms of fever ..................................................................................................................... 36 

4/ UNDERSTANDING AND TREATING FEVER ................................................................................................. 39 

Pathogenesis and pathophysiology .................................................................................................................. 39 

Treating fever ................................................................................................................................................... 42 

Malignant hyperthermia, neuroleptic malignant syndrome and lethal catatonia ........................................... 43 

Cooling techniques ...................................................................................................................................... 43 

CONCLUSION ............................................................................................................................................... 45 

SELF‐ASSESSMENT ....................................................................................................................................... 46 

PATIENT CHALLENGES ................................................................................................................................. 50

Introduction 

INTRODUCTION 

Thirty per cent of patients will become febrile, while up to 90% 

of patients with sepsis will experience fever, during a stay in the 

intensive care unit (ICU). Fever in critically ill patients may be 

of infective, non-infective, or mixed origin. The confirmation of 

the source of fever is often difficult which leads to a diagnostic 

dilemma and a difficult decision (to treat or not to treat) often 

resulting in a variability of treatment response from the medical 

and nursing staff. 

Fever in the ICU is an 

alarm signal most 

frequently indicating 

an activated host 

defence 

The Society of Critical Care Medicine practice parameters define fever in the 

ICU as a (core) temperature above 38.3 °C. The condition is caused by an 

imbalance between heat production and heat loss. In the clinical context, 

excessive heat generation is much more common than defective heat loss. The 

resulting disturbance may be transient and/or trivial or it may portend serious 

illness. 

This module focuses on the differential diagnosis of fever rather than on the 

antimicrobial treatment of infection. 

For current information on fever: 

Website of the Centers for Disease Control and Prevention (CDC) where current 

information on infection statistics and other relevant information is given. 

http://www.cdc.gov/ 

Website of the journal Emerging Infectious Diseases, published by the CDC 

http://www.cdc.gov/eid 

Website of the Infectious Diseases Society of America, and the Emerging 

Infections Network 

http://www.idsociety.org/ 

Website of the European Society of Clinical Microbiology and Infectious 

Diseases 

http://www.escmid.org 

Sepsis Resource Center and critical care pages 

http://www.medscape.com 

Niven DJ, Leger C, Stelfox HT, Laupland KB. Fever in the Critically Ill: A review 

of Epidemiology, Immunology, and management. J Intensive Care Med 

2011. [Epub ahead of print] PMID 21441283 

Laupland KB. Fever in the critically ill medical patient. Crit Care Med 2009; 37(7 

Suppl): S273–278. PMID 19535958 

Cunha BA, Shea KW. Fever in the intensive care unit. Infect Dis Clin North Am 

1996; 10(1): 185–209. PMID 8698990 

Marik PE. Fever in the ICU. Chest 2000; 117(3): 855–869. PMID 10713016 

[1]

Circiumaru B, Baldock G, Cohen J. A prospective study of fever in the intensive 

care unit. Intensive Care Med 1999; 25(7): 668–673. PMID 10470569 

Ryan M, Levy MM. Clinical review: fever in intensive care unit patients. Crit Care 

2003; 7(3): 221–225. PMID 12793871 

Dimopoulos G. Approach to the Febrile Patient in the Intensive Care Unit. In: 

Rello J, Kollef M, Diaz E, et al., editors. Infectious diseases in critical care. 

2nd ed. Berlin: Heidelberg; 2007. ISBN 9783540344056. pp. 1–9 

O’Grady NP, Barie PS, Bartlett JG, Bleck T, Carroll K, Kalil AC, et al. Guidelines 

for evaluation of new fever in critically ill adult patients: 2008 update 

from the American College of Critical Care Medicine and the Infectious 

Diseases Society of America. Crit Care Med 2008; 36(4): 1330–1349. 

PMID 18379262 

Dimopoulos G, Falagas ME. Approach to the febrile patient in the ICU. Infect Dis 

Clin North Am 2009; 23(3): 471–484. PMID 19665078 

Introduction 

[2]

Task 1. Assessing and measuring fever in ICU 

1/ ASSESSING AND MEASURING FEVER IN ICU 

Fever in an ICU patient is always a concern. The first and immediate priority is 

to determine its clinical significance. 

Assessment of fever of recent onset 

Fever has many causes depending on age, underlying illness, and the 

environment of the patient. Fever in a healthy adult commonly is considered as 

a result of viral infections such as influenza but in the hospital environment is 

considered of non-viral origin. In the critically ill, mechanically ventilated 

patient, for instance, the most common causes are a bacterial or fungal 

infection, unless proven otherwise. Non-infective causes of fever include 

thromboembolism, trauma, and others. The distinction between these various 

causes is important because of the difference in treatment and prognosis. In 

both medical and surgical critically ill patients, fever is caused by infective and 

non-infective conditions in roughly equal proportions. The latter tend to be 

confirmed once infective causes are ruled out; non-infective causes may include 

cerebral conditions affecting thermoregulation. Fever above 38.9 °C is more 

likely to be due to infective than non-infective causes, and vice versa. The higher 

the fever, the more likely it is to be of infective origin, but a temperature above 

41.1 °C can be of neurological origin. 

The presence of risk factors for nosocomial microbial infection 

in the critically ill patient render non-infective causes less likely. 

In fact, nosocomial infection complicates the hospital course of 

approximately 30% of critically ill patients, and fever of recent 

onset in the ICU is caused by nosocomial infection in more than 

half of cases. 

Ventilator-associated 

pneumonia, catheterrelated 

sepsis and 

sinusitis are the three 

major contributors to 

ICU fever of recent 

onset 

Risk factors for microbial infection include: 

Advanced age 

Severe underlying disease 

Neutropenia 

Immunosuppression 

Intravascular catheters 

Intubation and mechanical ventilation 

Prolonged ICU stay 

Prostheses 

Foreign bodies 

Prior surgery 

Bladder catheters and wound drains 

Nasogastric tubes 

Neurological disease with impaired consciousness. 

Stress ulcer prophylaxis is considered a risk factor for nosocomial infections 

associated with gastric colonisation by enteric organisms. In a large, hospitalbased 

pharmaco-epidemiologic cohort, acid-suppressive medication use was 

associated with 30% increased odds of hospital-acquired pneumonia while in 

subset analyses, statistically significant risk was demonstrated only for protonpump 

inhibitor use. 

[3]


More importantly, the presence of invasive devices predispose to infection. 

Intravascular catheters are associated with catheter-related blood stream 

infections. Endotracheal intubation and mechanical ventilation are risk factors 

for ventilator-associated pneumonia and the presence of a nasogastric or 

nasotracheal tube is a risk factor for sinusitis. 

Yeast and fungal infections are common in patients with severe underlying 

disease, in neutropenia, diabetes mellitus, renal failure, diabetes and after 

multiple courses of antibiotics. Furthermore, gastrointestinal surgery, open 

wounds, and a prolonged ICU stay, are risk factors for deep fungal infections. 

Risk factors for nosocomial infections in the critically ill are studied in: 

Girou E, Stephan F, Novara A, Safar M, Fagon JY. Risk factors and outcome of 

nosocomial infections: results of a matched case-control study of ICU 

patients. Am J Respir Crit Care Med 1998; 157(4 Pt 1): 1151–1158. PMID 

9563733 

Herzig SJ, Howell MD, Ngo LH, Marcantonio ER. Acid-suppressive medication 

use and the risk for hospital-acquired pneumonia. JAMA 2009; 301(20): 

2120–2128. PMID 19470989 

See ESICM Flash Conference: Ludwig Kramer. Epidemiology of nosocomial 

infections in ICU, Berlin, 2007. 

Appropriate investigations of a patient with fever should not involve an 

undirected battery of imaging, laboratory and microbiological tests but should 

be selected on the basis of a thorough clinical evaluation and targeted toward 

suspected sources of infection. Expeditious diagnosis is key to early effective 

therapy. In the diagnostic investigation of fever, the following sequence 

represents reasonable practice: 

See ESICM Flash Conference: Jean Carlet. Rapid identification and treatment of 

infection, Barcelona, 2006. 

Clinical appraisal 

Clinical assessment starts with a full history and complete physical examination. 

Assessment of fever of recent onset in the critically ill raises a number of 

questions: 

[4]


 

 

 

 

 

 

When did the fever start and did it relate to any clinical events e.g. to 

drainage of an infected collection or after removal of a central venous 

catheter (CVC), when catheter-related infection is suspected 

Is there a clinically recognisable focus of infection 

What are the likely micro-organisms involved 

How high is the temperature 

Are there risk factors for microbial infection 

Are there possible non-infective causes 

Q. In the context of fever of recent onset, what are the important 

items in the clinical history and why 

A. In the case of nosocomial infection, important history might include prior 

haematological disease e.g. acquired immunodeficiency syndrome (AIDS), since 

chronic infective disease may flare up in the presence of a decreased 

immunocompetence. Other items from the history are the duration of tracheal or nasal 

intubation, mechanical ventilation and indwelling central venous catheters. You will 

want to know how long these different foreign bodies have been in place as a pointer to 

the likelihood of infection. 

The physical signs of nosocomial infections can be subtle particularly in the 

patient with neutropenia or other causes of immune-suppression. In 

mechanically ventilated patients, the physical signs of ventilator-associated 

pneumonia may be manifest primarily by purulent sputum on tracheal suction. 

A decrease in oxygenation may suggest pneumonia or pulmonary embolism. 

Catheter-related infection may be accompanied by redness and discharge from 

the insertion site but occurs in the absence of such signs. In surgical patients, 

wound dressings should be removed to inspect wounds if they have not been 

seen by clinical staff during a scheduled dressing on that day. Wounds may need 

to be opened in case of suspected infection. Drain fluids should be examined for 

turbidity. Clostridium difficile infection and pseudomembranous colitis should 

be considered in any patient with fever and diarrhoea. 

Q. Is fundoscopy or other specific physical examination procedure 

useful in the ‘septic work-up’ (see below) and why 

A. Fundoscopic and skin examination may point to evidence of (septic) emboli. 

Candidaemia may be more likely if there is widespread Candida infection and 

endophthalmitis. There may be evidence of decubitus ulcers and/or skin fold infection. 

The appearance of a new murmur may suggest endocarditis. 

The ‘septic work-up’ or diagnostic approach to new onset fever in the critically 

ill can be summarised as follows: 

[5]


Approach to 

the febrile 

patient in the 

ICU 

Dimopoulos G, Falagas ME. Approach to the febrile patient in the ICU. Infect Dis Clin 

North Am 2009; 23(3): 471–484. PMID 19665078 

This figure, and a number of the figures used below, are slides in the 

recommended ESICM Flash Conference: George Dimopoulos. Late fever in an 

ICU patient, Barcelona, 2010. 

Fever – notable features and measurement 

Prior to assessment, you will wish to confirm the presence of fever and 

determine its severity. Response to fever varies with age. Elderly patients are 

unable to regulate their body temperature to the same degree as young adults, 

making them susceptible to extremes of temperature – older patients with 

serious infections have a substantial prevalence of apyrexia (20% to 30%) and a 

lower febrile response than younger patients. A lack of fever may contribute to 

lower resistance to infection, delayed recovery, and suboptimal outcome while 

lower febrile responses to infection are associated with a higher mortality rate 

and poorer prognosis. In children between the ages of six months and six years, 

febrile convulsions may occur. 

Core temperature measurement is, of course, the gold standard and several 

methods may be used in the ICU, involving the placement of a thermistor or 

similar device in the pulmonary (or femoral) artery, the bladder or the 

oesophagus. In practice, however, surrogate site (rectal, oral or axillary) 

temperature measurement is often used (Table below). 

[6]


Measurement of fever using different techniques at different body sites 

Site Method Comments 

Pulmonary artery Mixed venous blood Core temperature but 

pulmonary artery 

catheter required 

Bladder measurement Thermometer Core temperature but 

‘Foley’ catheter 

required 

Infrared ear Thermometer Values a few tenths 

below values in the 

pulmonary artery 

catheter and brain 

Rectal temperature 

Mercury thermometer 

or electrical probe 

A few tenths higher 

than (and lags 

behind) core 

temperature. 

Unpleasant and 

intrusive for patients 

Oral measurement Thermometer Influenced by 

warmed gases 

delivered by 

respiratory devices, 

by eating and 

drinking 

Axillary measurement Thermometer Underestimates core 

temperature, lacks 

reproducibility 

Whether shell or ‘non-core’ temperature can be considered a 

practical equivalent to core temperature is controversial. Rectal 

temperature (although sometimes classified as a ‘core’ 

temperature) may lag behind rapid changes in actual core 

temperature and therefore is not regarded as a ‘real-time’ 

measurement of core temperature. Axillary and oral methods 

are less reliable in reflecting core temperature. Cold liquids, 

among others, may confound oral temperatures. Infrared 

tympanic membrane temperature measurement devices have 

gained some popularity but in the study below, oral 

thermometry was found to be more accurate when a pulmonary 

artery core measurement was not available. In addition, in 

patients with head injury or cerebral bleeding/stroke, brain and 

thus tympanic temperature may exceed core temperature, but 

the clinical significance is unknown. 

Core, ‘non-core’ 

or shell 

(peripheral) 

temperature – is 

core always 

better 

Review the practice in your department concerning temperature 

measurement. When exploring the pros and cons, how do you rate the practice in your 

ICU 

[7]


Stavem K, Saxholm H, Smith-Erichsen N. Accuracy of infrared ear thermometry 

in adult patients. Intensive Care Med 1997; 23(1): 100–105. PMID 

9037647 

Giuliano KK, Scott SS, Elliot S, Giuliano AJ. Temperature measurement in 

critically ill orally intubated adults: a comparison of pulmonary artery 

core, tympanic, and oral methods. Crit Care Med 1999; 27(10): 2188– 

2193. PMID 10548205 

Bridges E, Thomas K. Noninvasive measurement of body temperature in critically 

ill patients. Crit Care Nurse 2009; 29(3): 94–97. PMID 19487784 

An 86-year-old lady with multiple trauma receiving mechanical 

ventilation for two weeks develops fever (green line, °C), without tachycardia 

(blue line, b/min). The recording is from the bedside computer monitor, 

visualising continuous measurements (vertical lines are days). There is a diurnal 

pattern. The diagnosis made was ventilator-associated pneumonia attributable 

to Pseudomonas aeruginosa. The blue arrow indicates the day of starting 

piperacillin and tobramycin, and the ‘lytic’ resolution of the fever is illustrated. 

Heart rate b/min 

Body temperature C 

Laboratory appraisal 

The clinical assessment is supplemented by selected laboratory measurements. 

The commonest of these is the leukocyte and differential counts as signs of 

infection include leukocytosis and a ‘left shift’. Investigators have searched for 

specific ‘sepsis’ markers including circulating C-reactive protein, procalcitonin 

and the cytokine, interleukin-6. Although the exact predictive values remain 

uncertain, some of these plasma factors can help to forecast the likelihood of 

microbial infection in a patient with fever, before the results of Gram stains, and 

particularly microbiological cultures, are available. On day six after trauma or 

surgery, development of fever and persistently high circulating IL-6 and C- 

reactive protein levels may be predictive for nosocomial infection. Similarly, the 

detection of endotoxaemia by means of rapid assay techniques may be of some 

predictive value in Gram-negative infection/bacteraemia and its associated 

morbidity. 

The following papers address the (limited) value of surrogate indicators of 

microbial infection. 

[8]


Fassbender K, Pargger H, Müller W, Zimmerli W. Interleukin-6 and acute-phase 

protein concentrations in surgical intensive care unit patients: diagnostic 

signs in nosocomial infection. Crit Care Med 1993; 21(8): 1175–1180. 

PMID 8339583 

Ugarte H, Silva E, Mercan D, De Mendonça A, Vincent JL. Procalcitonin used as a 

marker of infection in the intensive care unit. Crit Care Med 1999; 27(3): 

498–504. PMID 10199528 

Heyland DK, Johnson AP, Reynolds SC, Muscedere J. Procalcitonin for reduced 

antibiotic exposure in the critical care setting: A systematic review and an 

economic evaluation. Crit Care Med 2011 [Epub ahead of print] PMID 

21358400 

Imaging 

Bedside chest radiography is routinely used to detect new pulmonary infiltrates 

in the ICU. In this condition and in sinusitis, CT scan is associated with fewer 

false negative results than plain radiography. The benefits of CT, however, only 

rarely outweigh the inconvenience and risk of transferring the patient to the 

radiology department (see later sections for further discussion). 

Other imaging techniques include ultrasonography. Transoesophageal 

echocardiography can be of help for detecting pulmonary emboli and valvular 

lesions in endocarditis. Nuclear techniques can supplement other imaging 

studies, including CT and ultrasound, but are rarely used in critically ill patients 

with fever of unknown origin. Nuclear techniques that may be helpful to 

supplement imaging in the critically ill are discussed in: 

Dumarey N, Egrise D, Blocklet D, Stallenberg B, Remmelink M, del Marmol V, et 

al. Imaging infection with 18F-FDG-labeled leukocyte PET/CT: initial 

experience in 21 patients. J Nucl Med 2006; 47(4): 625–632. PMID 

16595496 

PACT module on Clinical imaging 

Culture techniques 

Specimens from sites of suspected infection, together with blood samples when 

indicated, should be obtained as a matter of course for Gram stain, culture and 

sensitivity determinations. Taking cultures should precede the use of empirical 

antibiotics unless undue delays are anticipated. Aspiration of pleural fluid or 

ascites may indicate potential sites of infection. Aspiration of localised fluid 

collections or abscesses can be guided by ultrasonography or CT scans. 

THINK: What are the indications for these types of radiological investigations and 

who is the best person to approach for advice in your institution 

[9]


Blood should be obtained percutaneously via venipuncture (or via ‘clean-stick’, 

newly introduced arterial or central venous catheters), and 10 ml placed in each 

of two bottles for aerobic/anaerobic cultures. At least two to three sets, 10 min 

apart, should be taken, after proper skin preparation. 

Shafazand S, Weinacker AB. Blood cultures in the critical care unit: improving 

utilization and yield. Chest 2002; 122(5): 1727-1736. PMID 12426278 

Microbiology 

The commonly identified micro-organisms causing infections in 

the ICU include Gram-negative bacilli (mainly 

Enterobacteriaceae, Klebsiella, Pseudomonas, Acinetobacter 

and Serratia spp.), Gram-positive bacteria such as coagulasenegative 

staphylococci and S. aureus and Candida albicans. 

Organisms should 

always be 

considered in the 

specific clinical 

context when 

making ‘best guess’ 

therapeutic 

decisions 

Blood culture results with S. epidermidis are considered clinically ‘significant’, if 

present in more than one bottle and are rapidly growing in culture. Candida 

spp. may cause catheter-related blood stream infections, wound infections, and 

peritonitis. Culture of Candida spp. may, of course, represent colonisation as 

opposed to infection, but there are no commonly accepted criteria to separate 

these conditions. Candiduria exceeding 10 5 colony forming units/mL in two 

urine specimens taken before and after change of a bladder catheter in a patient 

with clinical signs of sepsis may point to Candida as the aetiology. A high 

Candida colony count in urine, recovery from two or more otherwise sterile sites 

(excluding urine and sputum) may point to Candida sepsis in the febrile ICU 

patient with leukocytosis (>12.0 x 10 9 /L). Candidaemia (e.g. after change of 

intravascular catheters) is indicative of infection. Candida endophthalmitis, 

oesophagitis, suppurative thrombophlebitis or wound infections/peritonitis 

(‘open abdomen’) may be the source of deep Candida infections. Further 

relevant details are to be found in the following reference. 

Holley A, Dulhunty J, Blot S, Lipman J, Lobo S, Dancer C, et al. Temporal trends, 

risk factors and outcomes in albicans and non-albicans candidaemia: an 

international epidemiological study in four multidisciplinary intensive 

care units. Int J Antimicrob Agents 2009; 33(6): 554. e1-7. Epub 2009 Jan 

22. PMID 19167196 

Viral infections causing pneumonia, even in critically ill, immunocompromised 

(or immunocompetent) patients, are rare. Herpes virus, cytomegalo, adeno- or 

respiratory syncytial viruses or Chlamydia spp. are considered the most 

frequent causes. 

[10]


Jaber S, Chanques G, Borry J, Souche B, Verdier R, Perrigault PF, et al. 

Cytomegalovirus infection in critically ill patients: associated factors and 

consequences. Chest 2005; 127(1): 233–241. PMID 15653989 

Limaye AP, Kirby KA, Rubenfeld GD, Leisenring WM, Bulger EM, Neff MJ, et al. 

Cytomegalovirus reactivation in critically ill immunocompetent patients. 

JAMA 2008; 300(4): 413–422. PMID 18647984 

Chiche L, Forel JM, Roch A, Guervilly C, Pauly V, Allardet-Servent J, et al. Active 

cytomegalovirus infection is common in mechanically ventilated medical 

intensive care unit patients. Crit Care Med 2009; 37(6): 1850–1857. PMID 

19384219 

THINK about the common clinical contexts relevant to these specific organisms. 

When do you think viral reactivation is harmful and when not 

Q. Why is viral reactivation important in clinical management 

A. In order to make an informed decision as to appropriate ‘best guess’ antibiotic 

treatment. 

Rare fungal infections developing in the critically ill may include 

Aspergillus fumigatus lung infections after near drowning or in the 

neutropenic/organ transplant patient with underlying haematologic malignancy 

or immunosuppression. A rare cause of bilateral sinusitis may be infection with 

Rhizopus (mucormycosis), particularly in diabetics, as illustrated in the 

references below. 

Gans RO, Strack van Schijndel RJ, Laarman DA, Stilma JS, Thijs LG. Fatal 

rhinocerebral mucormycosis and diabetic ketoacidosis. Neth J Med 1989; 

34(1-2): 29–34. PMID 2492643 

Dimopoulos G, Vincent JL. Candida and Aspergillus infections in critically ill 

patients. Clin Intens Care 2002; 13: 1–12 

Fishman JA. Infection in solid-organ transplant recipients. N Engl J Med 2007; 

357(25): 2601–2614. PMID 18094380 

For further insight into the evolution of microbiology of nosocomial 

bacteraemia in the ICU, see: 

Edgeworth JD, Treacher DF, Eykyn SJ. A 25-year study of nosocomial bacteremia 

in an adult intensive care unit. Crit Care Med 1999; 27(8): 1421–1428. 

PMID 10470744 

[11]

For opportunistic infections in surgical patients: 


Dunn DL. Diagnosis and treatment of opportunistic infections in 

immunocompromised surgical patients. Am Surg 2000; 66(2): 117–125. 

PMID 10695740 

Systemic inflammatory response syndrome (SIRS) 

In any patient with fever, one has to consider the likelihood of microbial 

infection and sepsis as opposed to SIRS which has been defined as: 

Fever (>38 °C) or hypothermia (90 b/min) 

Tachypnoea (>20/min), or fall in arterial PCO2 (12.0 x 10 9 /L) or leukopenia (10% 

immature (band) forms. 

See PACT module on Sepsis and MODS. 

Sepsis is defined as SIRS in the presence of a clinical or microbiologically 

proven infection. Infection is indicated by a host response to micro-organisms 

or the invasion of otherwise sterile host tissues by (replicating) microorganisms. 

However, the predictive value of SIRS for severe microbial infection may be 

poor; specificity is low and sensitivity high. For example, the criteria are often 

met in trauma patients even in the absence of microbial infection. Hence, the 

clinical value of SIRS is in doubt. Nevertheless, meeting severe sepsis (organ 

dysfunction associated with sepsis) and septic shock criteria (hypotension below 

90 mmHg in sepsis despite volume resuscitation) carries a higher mortality rate 

than meeting SIRS criteria alone, so that the latter classifications may have 

prognostic (rather than diagnostic) significance. 

THINK: The usefulness of SIRS and sepsis criteria in patients with fever remains 

unclear. The sensitivity of the syndrome may be too high and specificity too low for 

microbial infection, even when supplemented by other ‘sepsis signs’. 

You may wish to consider the following references. 

Pittet D, Rangel-Frausto S, Li N, Tarara D, Costigan M, Rempe L, et al. Systemic 

inflammatory response syndrome, sepsis, severe sepsis and septic shock: 

incidence, morbidities and outcomes in surgical ICU patients. Intensive 

Care Med 1995; 21(4): 302–309. PMID 7650252 

Bossink AW, Groeneveld AB, Koffeman GI, Becker A. Prediction of shock in 

febrile medical patients with a clinical infection. Crit Care Med 2001; 

29(1): 25–31. PMID 11176153 

[12]


Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 

International Sepsis Definitions Conference. 2001 

SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions 

Conference. Intensive Care Med 2003; 29(4): 530–538. Epub 2003 Mar 

28. PMID 12664219 

Determine the prevalence (number of cases per total number of patients) and 

incidence (number of new cases per total number in a given time period) of SIRS and 

sepsis in your ICU population over three days, assuming that criteria must be met 

within a six-hour time window. What percentage is associated with positive culture 

[13]

Task 2. Determining the cause of fever in the critically ill patient 

2/ DETERMINING THE CAUSE OF FEVER IN THE 

CRITICALLY ILL PATIENT 

Causes of fever of recent onset in the critically ill patient, in descending order of 

likelihood. 

Infective causes 

Ventilator-associated pneumonia 

Catheter-related infection 

Upper respiratory tract infection and sinusitis 

Gastrointestinal infection: Clostridium difficile 

Urinary tract infection 

Acalculous cholecystitis 

Endocarditis 

Primary Gram-negative bacteraemia 

Malaria. 

Non-infective causes 

Pulmonary aspiration 

Postoperative fever (


Vincent JL, Rello J, Marshall J, Silva E, Anzueto A, Martin CD, et al.; EPIC II 

Group of Investigators. International Study of the Prevalence and 

Outcomes of Infection in Intensive Care Units. JAMA 2009; 302(21): 

2323–2329. PMID 19952319 

O’Grady NP, Barie PS, Bartlett JG, Bleck T, Carroll K, Kalil AC, et al. Guidelines 

for evaluation of new fever in critically ill adult patients: 2008 update 

from the American College of Critical Care Medicine and the Infectious 

Diseases Society of America. Crit Care Med 2008; 36(4): 1330–1349. 

PMID 18379262 

Infective causes 


The longer the duration of mechanical ventilation the greater the risk of 

developing ventilator-associated pneumonia (VAP). Sinusitis is also a risk factor 

for VAP. Early diagnosis and effective treatment is associated with a lower 

morbidity and mortality. Confirmation of the clinical diagnosis in a ventilated 

patient developing fever, impaired oxygenaton and purulent sputum may be 

obtained by means of tracheal aspirates and a new infiltrate on chest 

radiography. The additional presence of pathogenic micro-organisms with 

leukocytes is diagnostic – see pulmonary infection score below. 

The preceding table is adapted from: 

Fartoukh M, Maitre B, Honoré S, Cerf C, Zahar JR, Brun-Buisson C. Diagnosing 

pneumonia during mechanical ventilation: the clinical pulmonary 

infection score revisited. Am J Respir Crit Care Med 2003; 168(2): 173– 

179. Epub 2003 May 8. PMID 12738607 

CT scanning of the thorax provides better visualisation of infiltrates than 

bedside chest radiography. In selected cases of suspected VAP, CT scanning can 

therefore be useful, even though transportation to the CT department is 

necessary. It may also help to recognise and allow drainage of empyema. 

[15]


Tracheal aspirates: their diagnostic significance is greater if (semi-) 

quantitative rather than qualitative cultures are performed; this helps to obviate 

false-positive results by colonisation of upper airways in the absence of lower 

respiratory tract infection by the bacteria. Usually, a cut-off point of 10 5 cfu/mL 

is taken. Indeed, some micro-organisms are obligatory pathogens while the lowgrade 

presence of others, such as Gram-negative bacilli, may merely represent 

colonisation. Microscopy of the aspirates is necessary to exclude saliva with 

many epithelial cells, and elastin staining may confirm a lower (versus upper) 

respiratory tract origin of the aspirate. In the case of VAP, the aspirate typically 

contains numerous neutrophils. 

Distal bronchial specimens: tracheal aspirate results are less specific than 

those of Gram stains, microscopy and cultures of lower (distal) pulmonary 

secretions obtained by bronchoscopy and bronchoalveolar lavage (BAL) or 

protected specimen brush (PSB). It remains unclear, however, whether 

antibiotic guidance based on the latter is associated with lower morbidity and 

mortality for suspected VAP than antibiotic treatment guided by tracheal 

aspirates. Nevertheless, utilisation of these invasive tools may prevent 

overtreatment by antibiotics and reduce antibiotic pressure. This may be 

increasingly relevant because of the increase in multiresistant pathogens 

causing VAP. 

Finally, positive blood or, when present, pleural fluid cultures with the same 

organism as recovered from the airway can be found in VAP. 

Before proceeding to the next section, consider searching the Web for 

evidence that treatment guided by BAL/PSB specimens is superior to that guided by 

conventional, less invasive techniques and also for more specific clinical indications for 

the use of the invasive technique. Then assess the views expressed below. 

The table below outlines a diagnostic approach to VAP when invasive 

procedures are performed. 

Criteria for the diagnosis of VAP 

I Three or more of the following: 

a) Rectal temperature >38.0 °C or 103 cfu/L) or >5% of leukocytes containing phagocytosed bacteria. 

b) Positive blood culture with the same micro-organism as that present in the airway 

c) Positive culture of pleural fluid 

[16]


The following reference is in favour of invasive technique for VAP diagnosis to 

reduce antibiotic usage and to improve outcome. 

Fagon JY, Chastre J, Wolff M, Gervais C, Parer-Aubas S, Stéphan F, et al. Invasive 

and noninvasive strategies for management of suspected ventilatorassociated 

pneumonia. A randomized trial. Ann Intern Med 2000; 132(8): 

621–630. PMID 10766680 

A CDC paper on the diagnosis of VAP favouring invasive techniques can be 

found on the following e-reference 

http://www.cdc.gov/ncidod/eid/vol7no2/mayhall.htm 

but the following papers oppose this approach: 

Ruiz M, Torres A, Ewig S, Marcos MA, Alcón A, Lledó R, et al. Noninvasive versus 

invasive microbial investigation in ventilator-associated pneumonia: 

evaluation of outcome. Am J Respir Crit Care Med 2000; 162(1): 119–125. 

PMID 10903230 

Heyland D, Dodek P, Muscedere J, Day A, Cook D; for the Canadian Critical 

Care Trials Group. A randomized trial of diagnostic techniques for 

ventilator-associated pneumonia. N Engl J Med 2006; 355(25): 2619– 

2630. PMID 17182987 

Muscedere J, Dodek P, Keenan S, Fowler R, Cook D, Heyland D; VAP Guidelines 

Committee and the Canadian Critical Care Trials Group. Comprehensive 

evidence-based clinical practice guidelines for ventilator-associated 

pneumonia: diagnosis and treatment. J Crit Care 2008; 23(1): 138–147. 

PMID 18359431 

Hence, choosing among strategies remains hard, is controversial and is 

dependent in part on local practices. 

For review: 

Morehead RS, Pinto SJ. Ventilator-associated pneumonia. Arch Intern Med 

2000; 160(13): 1926–1936. PMID 10888967 

Q. Look at the chest radiographs of this ICU patient who developed 

fever of 38.5 °C and a deterioration in his oxygenation. The initial 

CXR is on the left and shows right lower lobe volume loss and some 

patchy infiltrate; the CXR on the right was taken 48 hours later. The 

patient was a 72-year-old male with cerebellar haemorrhage, a 

[17]


tracheal tube for mechanical ventilation and his sputum had become 

purulent. Interpret the CXRs and give your presumptive diagnosis. 

Day 1 Day 3 

A. The chest radiograph two days later (right), shows the development of a pleural 

effusion. Taking the radiological infiltrates together with the clinical signs (purulent 

secretions, gas-exchange deterioration and fever), there is presumptive evidence of 

ventilator-associated pneumonia. 

Q. How would you prove a diagnosis of VAP 

A. Microbiological proof may be obtained by a positive culture from sputum or distal 

bronchial sample (see microbial diagnostic criteria above) or from the pleural fluid. In 

this instance, the cultures of tracheal aspirate and pleural fluid yielded Serratia 

marcescens which was subsequently successfully treated by ceftriaxone. 

Central venous catheter-related infections 

Catheter infection should be suspected in the febrile ICU patient with an 

intravascular catheter when 

There is fever or a positive blood culture in the absence of another 

evident source of infection 

The CVC dwell time exceeds 3 days 

Fever abates after catheter removal 

There are signs of local (exit-site) infection. 

Signs of exit-site infection include redness and purulent discharge. 

The definitive diagnosis of central venous catheter-related infection is normally 

made after CVC removal and demonstration of a positive catheter tip culture 

with an identical micro-organism grown from a percutaneous ‘clean-stick’ 

culture(s), drawn peripherally and simultaneously, from blood. 

Some important definitions 

Exit-site catheter infection is defined as the presence of positive quantitative 

catheter culture in the presence of symptoms of local infection (erythema, 

tenderness, induration, or purulence), in the absence of other foci. 

Catheter-related blood stream infection (CRBSI) is diagnosed when the same 

organism is isolated (at higher concentrations – see below) on quantitative 

culture of the distal catheter segment and from the blood of a patient with 

[18]


clinical symptoms of local or systemic infection and no other source of infection 

evident. 

In the absence of laboratory confirmation, defervescence after removal of an 

implanted catheter from a patient with blood stream infection is considered 

indirect evidence of CRBSI. 

Walshe CM, Boner KS, Bourke J, Hone R, Phelan D. Diagnosis of catheter-related 

bloodstream infection in a total parenteral nutrition population: inclusion 

of sepsis defervescence after removal of culture-positive central venous 

catheter. J Hosp Infect 2010; 76(2): 119–123. Epub 2010 Jun 16. PMID 

20554348 

Catheter colonisation is diagnosed when bacteria are cultured from catheter 

segments (more than 15 CFU for the semi-quantitative roller plate method or 

>100–1000 CFU for quantitative techniques) or blood drawn through the 

catheter; in the absence of local or systemic infection symptoms, or of positive 

cultures from peripherally taken blood. 

Catheter contamination is diagnosed when bacteria are cultured from catheter 

segment (more than 15 CFU bacteria for the roller plate method or >100–1000 

CFU for quantitative techniques) or blood drawn through the catheter, in the 

presence of systemic infection symptoms and positive cultures from 

peripherally obtained blood, that do not resolve after removal of the catheter. 

The following sets of diagnostic criteria are variously used: 

Catheter-related Blood Stream Infections (BSI) criteria 

CDC* criteria 

Criterion 1 

Patient has a recognised pathogen 

cultured from one or more blood 

cultures and organism cultured from 

blood not related to an infection at 

another site. 

Criterion 2 

Patient has at least one of the following 

signs or symptoms : fever (>38 C), 

chills, or hypotension 

and 

signs and symptoms 

and 

positive laboratory results not related to 

an infection at another site 

and 

common skin contaminant (i.e., 

diphtheroids [Corynebacterium spp.], 

Bacillus [not B. anthracis] spp., 

Propionibacterium spp., coagulase- 

[19] 

HELICS** criteria 

Criterion 1 – catheter-related infection 

(CRI) 1 

Local central venous catheter (CVC)- 

related infection (without positive blood 

cultures). 

Quantitative CVC culture 10 3 colony 

forming units (CFU)/ml 

or 

Semi-quantitative CVC culture >15 

CFU and 

Pus/inflammation at the insertion site 

or tunnel. 

Criterion 2 (CRI 2) 

General CVC-related infection (without 

positive blood cultures). 

 

Quantitative CVC culture 10 3 CFU/ml 

or


negative staphylococci [including S. 

epidermidis], viridans group 

streptococci, Aerococcus spp., 

Micrococcus spp.) cultured from two or 

more blood cultures drawn on separate 

occasions 

Criterion 3 

Patient 38 C core) hypothermia (15 

CFU and 

Clinical signs of systemic sepsis 

improve within 48 hours after 

catheter removal. 

Criterion 3 (CRI 3) 

Blood stream infection (BSI) 

occurring 48 hours before or after 

catheter removal 

and 

Positive culture with the same microorganism 

of either: 

 

 

Quantitative CVC culture 10 3 CFU/ml 

or 

Semi-quantitative CVC culture >15 

CFU or 

Pus from the insertion site 

Alternatively, if paired blood samples 

are taken from blood and from the 

CVC: 

Quantitative blood culture ratio CVC 

blood sample/peripheral blood 

sample >5 or 

Differential delay to positivity of 

blood cultures: CVC blood sample 

culture becomes positive 2 hours or 

more before peripheral blood culture 

(blood samples drawn at the same 

time) 

* Centers for Disease Control 

**Hospitals in Europe Link for Infection Control through Surveillance 

Causative micro-organisms include coagulase-negative staphylococci, S. aureus, 

Enterococcus, Gram-negative and Candida spp. The following references 

describe these criteria and preventable risk factors for infection. 

[20]


O'Grady NP, Alexander M, Dellinger EP, Gerberding JL, Heard SO, Maki DG, et 

al. Guidelines for the Prevention of Intravascular Catheter-Related 

Infections. Centers for Disease Control and Prevention. MMWR Recomm 

Rep 2002; 51(RR-10): 1–29. PMID 12233868 

HELICS: Surveillance of nosocomial infections in Intensive Care Units, Version 

6.1, September 2004, http://helics.univ-lyon1.fr/index.htm 

Maki DG, Kluger DM, Crnich CJ. The risk of bloodstream infection in adults with 

different intravascular devices: a systematic review of 200 published 

prospective studies. Mayo Clin Proc 2006; 81(9): 1159–1171. PMID 

16970212 

Mermel LA, Allon M, Bouza E, Craven DE, Flynn P, O'Grady NP, et al. Clinical 

practice guidelines for the diagnosis and management of intravascular 

catheter-related infection: 2009 update by the Infectious Diseases Society 

of America. Clin Infect Dis 2009; 49(1): 1–45. PMID 19489710 

CRI diagnosis without CVC removal 

Two recognised methods of catheter-related blood stream infection diagnosis 

are: 

 

Differential time to positivity: a relatively rapid (by at least 2h) onset of a 

positive culture in blood drawn via the CVC as compared to a paired 

sample from peripheral blood. 

Quantitative blood cultures: a quantitative organism ratio of at least 3:1 

colony forming units (CFU) per ml between paired samples drawn from 

the catheter hub and peripheral blood cultures respectively. 

Hence, paired blood cultures may be used, one taken through the catheter and 

the other via a ‘clean-stick’ (percutaneous) blood culture and presuming the 

patient is reasonably stable, the CVC may remain in situ until results from such 

cultures become available. 

Blot F, Nitenberg G, Chachaty E, Raynard B, Germann N, Antoun S, et al. 

Diagnosis of catheter-related bacteraemia: a prospective comparison of 

the time to positivity of hub-blood versus peripheral-blood cultures. 

Lancet 1999; 354(9184): 1071–1077. PMID 10509498 

Chen WT, Liu TM, Wu SH, Tan TD, Tseng HC, Shih CC. Improving diagnosis of 

central venous catheter-related bloodstream infection by using differential 

time to positivity as a hospital-wide approach at a cancer hospital. J Infect 

2009; 59(5): 317–323. Epub 2009 Sep 11. PMID 19748520 

Other methods that have also been reported to be able to help diagnose 

catheter-related BSIs without catheter removal include acridine orange/Gram 

staining of blood drawn through the catheter, brush specimens of endoluminal 

contents and cultures from the hub surface and skin at the catheter exit-site. 

Indeed, direct (acridine orange/Gram) staining techniques of bacteria (in 

[21]


leukocytes) from catheter segments and blood may give results earlier than 

cultures. 

Non-removal of the CVC is only acceptable in stable patients. If it is decided to 

remove the catheter, both the tip and the intradermal part should be cultured 

(semi-) quantitatively (roller plate method). A change of catheter over a guide 

wire in case of suspected catheter-related infection carries a risk of reinfecting 

the new catheter, such that infection does not resolve, and it is not 

recommended. The risk should be weighed against the mechanical risks 

associated with a new puncture and insertion site. 

For an overview and guidelines in diagnostics, see: 

Dobbins BM, Kite P, Wilcox MH. Diagnosis of central venous catheter related 

sepsis – a critical look inside. J Clin Pathol 1999; 52(3): 165–172. PMID 

10450173 

Polderman KH, Girbes AR. Central venous catheter use. Part 2: infectious 

complications. Intensive Care Med 2002; 28(1): 18–28. Epub 2001 Nov 

29. PMID 11818995 

Eggimann P. Diagnosis of intravascular catheter infection. Curr Opin Infect Dis 

2007; 20(4): 353–359. PMID 17609593 

Incidence of CRBSI 

Depending on the type of unit and patient, among other factors, 

the rate of catheter-related blood stream infection varies 

between 0 and 33% (mean 5%) of catheters, or an incidence 

density of 2.8 to 12.8 episodes per 1000 catheter days. 

Central venous 

catheter-related 

sepsis is 

preventable! 

Implementation of a multiple approach prevention strategy can decrease the 

number of catheter blood stream infections from 11.3 episodes per 1000 patient 

days to 3.8 episodes. 

Prevention of catheter-related infection is of utmost importance, see: 

Pearson ML. Guideline for prevention of intravascular device-related infections. 

Part I. Intravascular device-related infections: an overview. The Hospital 

Infection Control Practices Advisory Committee. Am J Infect Control 

1996; 24(4): 262–277. PMID 8870910 

Eggimann P, Harbarth S, Constantin MN, Touveneau S, Chevrolet JC, Pittet D. 

Impact of a prevention strategy targeted at vascular-access care on 

incidence of infections acquired in intensive care. Lancet 2000; 

355(9218): 1864–1868. PMID 10866442 

Pronovost P, Needham D, Berenholtz S, Sinopoli D, Chu H, Cosgrove S, et al. An 

intervention to decrease catheter-related bloodstream infections in the 

ICU. N Engl J Med 2006; 355(26): 2725–2732. PMID 17192537 

[22]


O'Grady NP, Alexander M, Burns LA, Dellinger EP, Garland J, Heard SO, et al.; 

Healthcare Infection Control Practices Advisory Committee (HICPAC). 

Guidelines for the prevention of intravascular catheter-related infections. 

Clin Infect Dis 2011; 52(9): e162–193. Epub 2011 Apr 1. PMID 21460264 

or 

http://www.cdc.gov/ncidod/dhqp/gl_intravascular.html 

What is the incidence of central venous catheter-related infection in your unit 

and does this necessitate a change of insertion and maintenance policy 

Complications of catheter-related infection other than severe sepsis 

and septic shock include endocarditis, and metastatic abscesses, thrombosis 

and suppurative phlebitis. 

Management of CRI 

Treatment is outside the scope of this module but the guidelines on the 

management of CRI of the Infectious Diseases Society of America (IDSA)is 

recommended and is available (http://www.idsociety.org/) – see Mermel LA et 

al below 

Mermel LA, Allon M, Bouza E, Craven DE, Flynn P, O'Grady NP, et al. Clinical 



of America. Clin Infect Dis 2009; 49(1): 1–45. PMID 19489710 

http://cid.oxfordjournals.org/content/49/1/1.full#sec-3 

Sinusitis 

Nasogastric and nasotracheal tubes are important risk factors, and should be 

removed if present during treatment of sinusitis. Prolonged nasotracheal 

intubation is particularly associated with sinusitis; this type of airway is often 

converted to an oral tube when prolonged intubation is anticipated. Sinusitis is 

a risk factor for ventilator-associated pneumonia (VAP). Hence, a search for and 

treatment of sinusitis in febrile patients may prevent VAP and associated 

mortality. Fever and purulent nasal discharge in the presence of nasal tubes 

may point to nosocomial sinusitis. The maxillary sinuses are most commonly 

affected, but sphenoidal or ethmoidal sinusitis, whether or not accompanied by 

maxillary sinusitis, is increasingly recognised. The diagnosis is difficult, even in 

the case of maxillary sinusitis, since bedside plain radiograms (Caldwell and 

Waters’ view) may not be sensitive and specific. This may be overcome by a CT 

scan. 

[23]


Konen E, Faibel M, Kleinbaum Y, Wolf M, Lusky A, Hoffman C, et al. The value of 

the occipitomental (Waters’) view in diagnosis of sinusitis: a comparative 

study with computed tomography. Clin Radiol 2000; 55(11): 856–860. 

PMID 11069741 

Vargas F, Bui HN, Boyer A, Bébear CM, Lacher-Fougére S, De-Barbeyrac BM, et 

al. Transnasal puncture based on echographic sinusitis evidence in 

mechanically ventilated patients with suspicion of nosocomial maxillary 

sinusitis. Intensive Care Med 2006; 32(6): 858–866. Epub 2006 Apr 14. 

PMID 16614810 

Q. Considering the risks involved in transporting a critically ill 

patient to the radiology department, in what circumstances might it 

be justified to perform a CT examination to confirm a diagnosis of 

sinusitis 

A. If the patient has to be transported for investigation of another major problem. Some 

institutions may also perform a CT scan when sinusitis persists or recurs despite 

adequate treatment (drainage and lavage) for 72h. 

Opacification or fluid-air levels necessitate needle aspiration, microscopy and 

culture of secretions to confirm a radiologic diagnosis. Only half the patients 

with a radiological diagnosis are confirmed to have sinusitis on aspiration. A 

radiological diagnosis is confirmed on aspiration if staining and culture yield 

neutrophils and micro-organisms at a concentration >10 3 CFU/ml. Gramnegative 

bacteria are often involved, and polymicrobial infections are relatively 

common. Treatment of maxillary sinusitis includes needle aspiration, lavage 

and, sometimes, systemic antibiotics. The clinical and radiographic features of 

infection should abate within a few days, following the start of appropriate 

treatment. Rarely, persistent or recurrent sinusitis may necessitate surgical 

exploration. 

Q. Examine this CT scan from a 54-year-old male admitted because 

of respiratory insufficiency in the course of Legionella pneumonia. 

Day 11 of mechanical ventilation was complicated by fever, 

leukocytosis, and purulent nasal discharge, in spite of systemic 

antibiotics. What is your presumptive diagnosis and how would you 

prove it 

[24]


A. The CT scan demonstrates fluid levels in both maxillary sinuses. The diagnosis of 

sinusitis can be confirmed if aspiration reveals microscopic and culture evidence for 

bacterial infection. Specimens obtained by aspiration revealed some leukocytes on 

Gram staining and Gram-positive cocci. Cultures from both sinuses grew Candida 

albicans. The condition cleared following repeated lavages. 

Urinary tract infections 

Most critically ill patients will have indwelling urinary bladder catheters. 

Nevertheless, a urinary tract infection is rarely the cause of fever in the critically 

ill, even though colonisation (bacteriuria at >10 5 CFU/ml) is common. Fever, 

leukocytosis, pyuria and a known pathogen in urine with the same pathogen 

cultured from blood points to a urinary tract infection in the febrile critically ill 

patient that should be treated by antibiotics. Obstructed catheters should be 

replaced. 

Rosser CJ, Bare RL, Meredith JW. Urinary tract infections in the critically ill 

patient with a urinary catheter. Am J Surg 1999; 177(4): 287–290. PMID 

10326844 

Tambyah PA, Maki DG. The relationship between pyuria and infection in patients 

with indwelling urinary catheters: a prospective study of 761 patients. 

Arch Intern Med 2000; 160(5): 673–677. PMID 10724053 

Hooton TM, Bradley SF, Cardenas DD, Colgan R, Geerlings SE, Rice JC, et al.; 

Infectious Diseases Society of America. Diagnosis, prevention, and 

treatment of catheter-associated urinary tract infection in adults: 2009 

international clinical practice guidelines from the Infectious Diseases 

Society of America. Clin Infect Dis 2010; 50(5): 625–663. PMID 20175247 

[25]

Acute acalculous cholecystitis 


After multiple trauma, burns, severe sepsis and major surgery, the gallbladder 

may become inflamed in the absence of gall stones. This inflammation, called 

acalculous cholecystitis, has an estimated incidence of 1.5% especially in septic 

patients or in patients recovering from abdominal sepsis. The low incidence is 

probably because of the non-specific clinical signs (pain in the right upper 

quadrant and nausea) and laboratory work-up. The detected wall thickness >3 

mm, intramural lucencies, gallbladder distension, pericholecystic fluid, and 

intramural sludge are helpful radiological findings while hepatobiliary 

scintigraphy is characterised by a high false-positive rate (>50%). Frequently, 

the diagnosis is delayed and the disease progresses to ischaemia, gangrene and 

perforation, indicating the necessary high index of suspicion while the 

treatment of choice is cholecystectomy. However, in very unstable patients, 

radiologic percutaneous drainage (cholecystostomy) may be preferred as a 

temporary measure and has replaced surgical cholecystectomy as a first choice 

treatment in many centres. In many patients, antibiotics will be prescribed, 

aimed at the causative organism, identified after percutaneous puncture and 

culture of the bile. For further details see: 

Boland G, Lee MJ, Mueller PR. Acute cholecystitis in the intensive care unit. New 

Horiz 1993; 1(2): 246–260. PMID 7922407 

Barie PS, Fischer E. Acute acalculous cholecystitis. J Am Coll Surg 1995; 180(2): 

232–244. PMID 7850064 

Other causes 

Be aware of central nervous system infections in patients with (internal or 

external) neurosurgical monitoring or draining devices. Coagulase-negative 

staphylococci are often involved. Suspected infection should prompt obtaining 

cerebrospinal fluid (CSF) for Gram stain and culture. 

Pseudomembranous colitis caused by Clostridium difficile has become a 

prevalent problem in many ICUs. In milder forms of the infection, diarrhoea 

may be the only feature. C. difficile -related diarrhoea is a relatively frequent 

occurrence in the critically ill, particularly if there has been treatment with 

multiple courses of broad-spectrum antibiotics. More severe forms of the 

disease are frequently characterised by a marked leukocytosis and elevated 

creatinine. Occasionally, an acute abdomen may result from C. difficile infection 

and surgical colectomy may be required. More virulent strains causing severe 

disease have recently emerged. 

The bacteria can be transmitted from patient to staff and vice 

versa, so that inadequate handwashing (alcohol gel is 

inadequate and soap and water is required for spore removal) 

may result in small outbreaks in the ICU. The diagnosis is 

established by a positive faecal toxin A and B (or tissue culture 

cytotoxicity) assays and increased faecal leukocytes. A negative 

[26] 

Rigorous attention to 

simple hygienic measures 

in the ICU is imperative. 

Alcohol hand cleansing is 

regarded as inadequate 

to clear C diff spores; a 

(traditional) physical 

handwash is required


immunoassay does not rule out the diagnosis. For further 

reading: 

Pépin J, Valiquette L, Alary ME, Villemure P, Pelletier A, Forget K, et al. 

Clostridium difficile-associated diarrhea in a region of Quebec from 1991 

to 2003: a changing pattern of disease severity. CMAJ 2004; 171(5): 466– 

472. PMID 15337727 

Pépin J, Saheb N, Coulombe MA, Alary ME, Corriveau MP, Authier S, et al. 

Emergence of fluoroquinolones as the predominant risk factor for 

Clostridium difficile-associated diarrhea: a cohort study during an 

epidemic in Quebec. Clin Infect Dis 2005; 41(9): 1254–1260. Epub 2005 

Sep 20. PMID 16206099 

Bartlett JG. Narrative review: the new epidemic of Clostridium difficileassociated 

enteric disease. Ann Intern Med 2006; 145(10): 758–764. 

PMID 17116920 

Ticehurst JR, Aird DZ, Dam LM, Borek AP, Hargrove JT, Carroll KC. Effective 

detection of toxigenic Clostridium difficile by a two-step algorithm 

including tests for antigen and cytotoxin. J Clin Microbiol 2006; 44(3): 

1145–1149. PMID 16517916 

Cohen SH, Gerding DN, Johnson S, Kelly CP, Loo VG, McDonald LC, et al.; 

Society for Healthcare Epidemiology of America; Infectious Diseases 

Society of America. Clinical practice guidelines for Clostridium difficile 

infection in adults: 2010 update by the Society for Healthcare 

Epidemiology of America (SHEA) and the Infectious Diseases Society of 

America (IDSA). Infect Control Hosp Epidemiol 2010; 31(5): 431–455. 

PMID 20307191 

Non-infective causes 

Half of fever episodes in the ICU are of non-infective origin without the 

temperature usually exceeding 38.3 ºC. The medical history, including recent 

interventions along with the physical examination aids the clinician in 

narrowing down the differential diagnosis. However, the type of ICU population 

(e.g. medical vs surgical patients), the specific type of patients (e.g. 

immunocompromised, elderly), the history of recent epidemics and the local 

epidemiology must be taken into account. 

Drug-fever: has an unknown incidence (3%–7% of febrile episodes are 

attributed to drug reactions but many cases remain undiagnosed), a 

temperature range from 38.8 C (102 F) to 40 ºC(104 ºF) and is a difficult 

diagnosis (usually established by exclusion because of the non-specific signs and 

laboratory tests), shaking chills and spiking temperatures. A concomitant 

maculopapular rash makes the diagnosis simple but accompanies fever in only 

5%–10% of cases. Rarely an increased leukocyte count with a left shift, 

peripheral eosinophilia, a moderate elevation of serum transaminases, and a 

markedly elevated erythrocyte sedimentation rate (>100 mm/h) are recorded. 

The signs that are associated with drug-fever are a lack of appropriate pulse rate 

response and a relative bradycardia in the absence of intrinsic conduction 

[27]


defects or beta-blockade. Any drug can cause fever due to hypersensitivity 

producing fever alone, with local inflammation at the site of administration 

(phlebitis, sterile abscess, soft tissue reaction) or because of the delivery systems 

(diluent intravenous fluid, intravascular delivery devices).The high-risk agents 

for drug-fever are all antibiotics (especially β-lactams), anti-epileptic drugs 

(especially phenytoin), antiarrhythmics (mainly quinidine and procainamide), 

antihypertensives (α-methyldopa), diuretics, and stool softeners. Antibiotics 

with a lower risk for drug-fever development are: clindamycin, vancomycin, 

chloramphenicol, aztreonam, doxycycline, erythromycin, imipemen, 

quinolones, and aminoglycosides. 

The time between initiating a drug and fever appearance is estimated to be 21 

days (median 8 days) while the fever resolves usually within 72 hours after 

removing the offending drug. When a rash is present it may persist for days or 

weeks. The usual scenario of drug-fever in the ICU includes a patient in whom 

an already diagnosed infection is resolving and after an initial defervescence in 

temperature, a recurrence of fever is noticed. In this type of patient, the 

antibiotics should be discontinued if the infection has resolved or another 

infected site has not been detected. If the patient is stable, but the infection has 

not resolved, then the presumed offending agent should be removed and a 

modification to antibiotics, without potential sensitising, according to the 

spectrum of pathogens should be performed. 

Wood AJ. Adverse Drug Reactions. In: Fauci AS, Braunwald E, et al., editors. 

Harrison’s Principles of Internal Medicine. 14th ed. NY: McGraw-Hill; 

1998. ISBN 007912013X. pp. 422–430 

In cardiac care units (CCUs), the main causes of non-infective fever include: 

myocardial infarction, Dressler’s syndrome with pericarditis, 

thromboembolism, thrombolytic therapy with haemorrhagic complications and 

antiarrhythmic medication (e.g. procainamide, quinidine), and deep venous 

thrombosis. 

In a neurosurgical ICU, the posterior fossa syndrome is a common cause of noninfective 

fever that mimics meningitis with stiff neck, low level of 

glucose/increased level of protein in cerebrospinal fluid, and predominance of 

polymorphonuclear leukocytes in cerebrospinal fluid (CSF) as a result of blood 

leakage into CSF. The differential diagnosis from bacterial meningitis is based 

on the negative cultures and the gradual lessening of meningeal symptoms as 

the number of red blood cells decreases in the CSF with time. 

Other causes are: central fever (caused by intracranial lesion or trauma affecting 

the brain or hypothalamus that is resistant to antipyretics, exceeds 39 C or 

106 F and is characterised by absence of perspiration); the use of 

anticonvulsive medications; deep venous thrombosis and fat embolism 

syndrome in trauma patients. In the acute phase after head injury, the 

appearance of pyrexia is extremely frequent and deleterious to cerebral 

perfusion (CCP) and intracranial pressure (ICP); while lack of treatment by 

[28]


antipyretics has been correlated with a longer ICU stay. Other causes of 

nosocomial fever in the ICU include adrenal insufficiency, acute pancreatitis, 

decubitus ulcers and gastrointestinal haemorrhage. 

Cunha BA, Tu RP. Fever in the neurosurgical patient. Heart Lung. 1988; 17(6 Pt 

1): 608-11. PMID 3056881 

Stein PD, Afzal A, Henry JW, Villareal CG. Fever in acute pulmonary embolism. 

Chest 2000; 117(1): 39–42. PMID 10631196 

Stocchetti N, Rossi S, Zanier ER, Colombo A, Beretta L, Citerio G. Pyrexia in 

head-injured patients admitted to intensive care. Intensive Care Med 

2002; 28(11): 1555–1562. Epub 2002 Oct 4. PMID 12415441 

Blood transfusions may elicit acute febrile reactions, even in the 

absence of bacterial contamination or haemolysis. Both acute, 

but more often, delayed haemolytic transfusion reactions may 

also be responsible for the fever. (Sub) acute febrile reactions 

without haemolysis (negative direct antiglobulin test) are 

caused by antibodies present in the recipient’s plasma and 

directed against HLA antigens on leukocytes in the donor’s 

blood. Occasionally, the donor’s blood is contaminated by 

micro-organisms and elicits fever in the recipient. Acute or 

subacute febrile reactions during red cell and platelet infusions 

should lead to discontinuation of the transfusion. Both donor 

and recipient blood should be sent for culture and 

haematological investigation. 

Onset of fever 

during blood 

transfusion is a 

medical 

emergency 

Alcohol withdrawal is often seen in the first 48 to 72 hours following 

hospitalisation and can often result in confusion with infection. Infection, 

hepatitis or pancreatitis should therefore be sought and excluded. Alcohol 

withdrawal leads to autonomic disturbances with sweating and fever. 

Benzodiazepines are the drug of choice for treatment. An overview is given in: 

Spies CD, Rommelspacher H. Alcohol withdrawal in the surgical patient: 

prevention and treatment. Anesth Analg 1999; 88(4): 946–954. PMID 

10195555 

Contrary to common belief, atelectasis does not cause fever. However, 

atelectasis may become infected and constitute a nidus of pneumonia. 

[29]

Task 3. Fever in specific categories of critically ill patient 

3/ FEVER IN SPECIFIC CATEGORIES OF CRITICALLY 

ILL PATIENT 

Some categories of critical care patients deserve special mention. In some 

centres, surgical ICU is a separate entity but even where surgical patients are 

part of a general critical care population, some distinctive considerations 

pertain particularly in the early postoperative period. Fever in the immune 

suppressed and in neurological patients is also included here. 

The surgical critical care patient – determining the 

cause of fever 

The normal response to trauma and surgery includes release of proinflammatory 

mediators and an elevation of body temperature that usually does 

not exceed 38.5 °C and does not last longer than two days, unless the surgery 

was done for infection itself, e.g. peritonitis. Hence, any elevation of 

temperature above 38.5 °C, lasting longer than two days or developing on the 

third day, may indicate concomitant microbial infection and sepsis. About 10% 

of trauma patients develop a nosocomial infection. 

Trauma has some immunodepressant effect thereby increasing the risk 

for infection. 

Other risk factors relate to advanced age, underlying morbidity and extent of 

trauma and surgery. Risk factors also include prolonged hypotension, 

haematoma, foreign bodies and blood transfusion. Repeated and careful 

searches for a source and micro-organisms are mandatory in these patients. 

Gram-negative pneumonia and wound infection are among the most common 

sources. Careful search should be made for an infective focus, including removal 

of dressings and wound inspection. Bear in mind, however, that at least 35% of 

episodes of fever after trauma or surgery are of non-infective origin, and 

thromboembolism may lead the list of causes. 

Causes of fever of recent onset and infection in descending order of 

likelihood are: 

 

 

 

 

 

 

 

Nosocomial pneumonia (and rarely empyema) 

Urinary tract infection 

Wound infection 


Sinusitis 

Gram-negative bacteraemia 

Miscellaneous. 

Empirical antibiotics should only be given after appropriate clinical assessment 

and provisional diagnosis, imaging and obtaining specimens for culture. 

Antibiotic therapy if started should be reviewed, in the clinical context, once 

staining and culture results become available. 

[30]


Wound infection 

Obviously, wound infection is a major cause of fever after trauma and surgery. 

The frequency after surgery varies between 3.5 and 5%. 

Q. Name at least four risk factors for surgical site infection (SSI). 

A. Risk factors include advanced age, prolonged hypotension, haematoma, 

malnutrition, diabetes mellitus, foreign bodies, blood transfusion, and ‘dirty’ surgery. 

The infections are mostly caused by coagulase-negative staphylococci (and S. 

aureus), enterococci, Gram-negative bacteria and Candida albicans. Redness, 

discharge, swelling and pain are features of infection, and have a peak on the 

seventh postoperative day. Early (within three days) wound infections include 

necrotising fasciitis caused by virulent Streptococcus species, gas gangrene and 

abscesses with S. aureus. Later infections may harbour mixed flora or Gramnegative 

rods, either as cellulitis, necrotising fasciitis or abscess. 

Wound infections may thus include clostridial cellulitis (caused by Clostridium 

perfringens), synergistic necrotising soft tissue infections including Meleney’s 

synergistic gangrene (advancing gangrene of the abdominal wall) caused by 

microaerophilic streptococci and S. aureus, and Fournier’s gangrene (of 

perineum and scrotum) caused by obligate anaerobic streptococci and 

facultative anaerobes. Necrotising fasciitis can be polymicrobial (aerobic and 

anaerobic Gram-negative bacilli, enterococci, S. aureus) or can be caused by 

Streptococcus pyogenes Group A. 

A wound infection by toxin producing Staphylococcus or Streptococcus spp. can 

cause a toxic shock (like) syndrome. Any wound should be opened and cultured 

if there is a high index of suspicion of wound infection. Necrotising fasciitis is an 

emergency, the management of which includes necessary wide excision and 

drainage. 

Stevens DL, Bisno AL, Chambers HF, Everett ED, Dellinger P, Goldstein EJ, et al. 

Practice guidelines for the diagnosis and management of skin and softtissue 

infections. Clin Infect Dis 2005; 41(10): 1373–1406. Epub 2005 Oct 

14. PMID 16231249 

Q. How would you treat toxic shock syndrome 

A. The toxic shock syndrome is treated by wound drainage (or wide excision in the case 

of necrotising fasciitis), removal of foreign bodies associated with development of the 

syndrome, and an extended spectrum penicillin combined with clindamycin. The 

treatment of shock consists of fluid loading and, if needed, inotropic agents. 

[31]


Q. What is the role of immunoglobulins in the treatment of toxic 

shock syndrome 

A. Their value of immunoglobulins and hyperbaric oxygen is unclear. In lifethreatening 

cases, the potential advantages may outweigh the current lack of positive 

randomised clinical trials. For further reading on this subject, see the references below. 

A study comparing intravenous immunoglobulins in toxic shock syndrome with 

historical controls, not receiving such treatment, suggested a reduced mortality rate. 

Kaul R, McGeer A, Norrby-Teglund A, Kotb M, Schwartz B, O'Rourke K, et al. 

Intravenous immunoglobulin therapy for streptococcal toxic shock 

syndrome – a comparative observational study. The Canadian 

Streptococcal Study Group. Clin Infect Dis 1999; 28(4): 800–807. PMID 

10825042 

The abdomen 

In patients with suspected intra-abdominal sepsis, the decision to proceed with 

laparotomy is hard, particularly after a prior laparotomy, with or without 

peritonitis. Leakage of anastomoses and ischaemia/perforation leading to intraabdominal 

abscesses, are most often involved, followed by postoperative 

cholecystitis and diverticulitis. Conversely, shock in the course of extraabdominal 

sepsis may occasionally result in ischaemia, necrosis and perforation 

of (small) bowel segments, necessitating laparotomy, resection and drainage, if 

the patient is considered appropriate for surgery. 

After a primary laparotomy, persistent fever and leukocytosis, 

in spite of antibiotics and in the absence of an overt extraabdominal 

source, may be indicative of intra-abdominal 

infection. Other signs that may point to (recurrent) intraabdominal 

sepsis and should lead to consideration of 

relaparotomy are respiratory failure, renal failure, ileus, wound 

infection, mental changes, abdominal pain, and previous 

contaminated surgery. 

Submitting a patient 

to 

diagnostic/therapeuti 

c relaparotomy is 

largely a matter of 

clinical judgment 

based on experience 

and may be necessary 

for clinical recovery 

On the basis of these variables, scoring systems have been developed to help 

decision-making for relaparotomy. Careful imaging procedures (CT scan) are 

helpful in guiding surgery, while laboratory tests are not. Nevertheless, 

laparotomy without specific indications appears diagnostic in more than half of 

cases of suspected intra-abdominal sepsis. The mortality risk of a negative 

laparotomy in the critically ill is about 15%, while unrecognised intra-abdominal 

sepsis may have a mortality rate exceeding 50%. A laparotomy should not be 

withheld in such patients when there is a high index of suspicion for intraabdominal 

sepsis. Most common micro-organisms include Gram-negative and 

Enterobacteriaceae 

[32]


The decision for relaparotomy should, on every occasion, be based on a 

careful consideration of possible benefits and harm in the individual patient. 

Scheduled or planned relaparotomies may be of doubtful value. You may wish to 

discuss this statement with surgical colleagues. 

Determine retrospectively the frequency of a correct diagnosis for the last ten 

(re)laparotomies performed in your unit on the basis of suspected intra-abdominal 

sepsis. 

Solomkin JS, Mazuski JE, Bradley JS, Rodvold KA, Goldstein EJ, Baron EJ, et al. 

Diagnosis and management of complicated intra-abdominal infection in 

adults and children: Guidelines by the Surgical Infection Society and the 

Infectious Diseases Society of America. Surg Infect (Larchmt) 2010; 11(1): 

79–109. PMID 20163262 

Fever in immunocompromised patients 

See PACT module on Immunocompromised patients 

The immunocompromised host includes patients with immunological or 

haematological disease or those receiving immunosuppressive therapy, 

including steroids and anticancer drugs, for autoimmune disease or after organ 

transplantation. Moreover, the hypo-gammaglobulinaemic or post-splenectomy 

patient may have some immune defect, as shown by their susceptibility to 

pneumococcal sepsis and shock, often with a fatal outcome. Cellular immune 

defects predispose to infections with intracellular micro-organisms. Other risk 

factors for infection in cancer patients include the invasive procedures these 

patients undergo and the devices they receive. Neutropenic 

cancer/haematologic patients (neutrophil count


Gruson D, Hilbert G, Vargas F, Valentino R, Chene G, Boiron JM, et al. Impact of 

colony-stimulating factor therapy on clinical outcome and frequency rate 

of nosocomial infections in intensive care unit neutropenic patients. Crit 

Care Med 2000; 28(9): 3155–3160. PMID 11008974 

Quadri TL, Brown AE. Infectious complications in the critically ill patient with 

cancer. Semin Oncol 2000; 27(3): 335–346. PMID 10864221 

Review the practice in your own unit concerning the administration of growth 

factors in neutropenic patients. What evidence is available for giving this Is there a 

difference in practice worldwide 

Patients with human immunodeficiency virus (HIV) infection, whether or not 

complicated by AIDS, may require admission to the ICU, mainly because of 

respiratory insufficiency associated with pneumonia. Pneumonia is the major 

source of infection in AIDS patients with fever in the ICU. The latter still most 

often involves Pneumocystis jirovecii. Identification of the micro-organism on 

methenamine silver stains of respiratory secretions or with help of 

immunofluorescent monoclonal antibody techniques establishes the diagnosis. 

HIV infection also increases the risk of (penicillin-resistant) pneumococcal 

pneumonia. Opportunistic infections further include Myocobacterium avium 

complex disease, tuberculosis, cytomegalovirus infection, histoplasmosis, 

toxoplasmosis and leishmaniasis. 

Non-infective causes of fever include lymphomas and thrombophlebitis. Other 

conditions requiring intensive care may include the recently recognised 

complications of extensive and aggressive antiretroviral therapy including 

hepatic steatosis and lactic acidosis. The immune reconstitution syndrome at 

the start of antiretroviral therapy and pancreatitis associated with these drugs 

may be causes of fever in HIV-infected persons in the ICU. 

Cappell MS, Marks M. Acute pancreatitis in HIV-seropositive patients: a case 

control study of 44 patients. Am J Med 1995; 98(3): 243–248. PMID 

7872340 

Stenzel M. The HIV-infected patient in the intensive care unit. Curr Opin Crit 

Care 2000; 6: 330-336 

http://journals.lww.com/cocriticalcare/Abstract/2000/10000/The_HIV_infected_patient_in_the_i 

ntensive_care.5.aspx 

Q. How would you treat Pneumocystis jirovecii pneumonia in a 

mechanically ventilated patient 

A. The standard treatment includes cotrimoxazole (sulfamethoxazole and 

trimethoprim) at high doses intravenously and systemic steroids. Oxygen therapy or 

mechanical ventilation may be required for supportive care. Compliance of the lung is 

often quite low, so that permissive hypercapnia may be necessary to limit high inflation 

pressures. 

[34]

Fever in neurological disease 


Fever in patients with cerebral disease is especially dangerous. After acute 

stroke and head injury, brain temperature (directly measured) may even exceed 

body temperature. After stroke or neurotrauma, fever may be associated with 

increased damage and a worse neurological outcome. It is believed that 

increasing oxygen demand of the brain augments susceptibility to 

hypoperfusion. Fever in the neurological critically ill patient therefore warrants 

a rapid assessment of potential infective causes, including meningitis. This may 

be particularly difficult, since fever may have a non-infective origin, i.e. reset of 

the hypothalamic thermostat by the cerebral damage. Only a minority of febrile 

patients with serious cerebral disease have an identifiable infective source. 

Regardless of the origin, it is generally recommended that a febrile patient with 

serious cerebral disease be treated with antipyretics, in order to lower cerebral 

oxygen demand, decrease secondary brain damage, and improve (neurological) 

outcome. 

Antipyretics may include paracetamol (acetaminophen) by the oral, nasogastric, 

intravenous or rectal routes and/or by enteral or intravenous diclofenac (0.04 

mg/kg/hour by continuous i.v. infusion – max daily dose 150 mg – not available 

in the US). The benefit of lowering the temperature to prevent secondary brain 

damage is probably greater than the disadvantage of losing a marker for the 

response to antimicrobial drugs. Obviously, antimicrobial agents are required in 

the presence of infection. 

Examine the changing opinions over the past decade of the value of 

therapeutic cooling to reduce/prevent secondary brain damage. Is there now a 

consensus Examine current practice in your institution. 

In the following references the authors suggest that body temperature be 

maintained in a safe, normothermic range (e.g. 36.7 °C to 37.0 °C) for at least 

the first several days after acute stroke or head injury. 

Reith J, Jørgensen HS, Pedersen PM, Nakayama H, Raaschou HO, Jeppesen LL, 

et al. Body temperature in acute stroke: relation to stroke severity, infarct 

size, mortality, and outcome. Lancet 1996; 347(8999): 422–425. PMID 

8618482 

Ginsberg MD, Busto R. Combating hyperthermia in acute stroke: a significant 

clinical concern. Stroke 1998; 29(2): 529–534. PMID 9472901 

This recent ESICM Flash Conference addresses the current knowledge of the 

role of hypothermia in a variety of brain injury circumstances. Peter Andrews, 

Edinburgh, UK. Resuscitating the brain from global ischemia: hypothermia and 

beyond. Summer conference, Dublin, 2010. 

[35]

Identifying special forms of fever 


Heat stroke is included in the module on Environmental hazards. Other 

circumstances where the fever may be especially high are malignant 

hyperthermia, neuroleptic malignant syndrome and lethal catatonia 

See PACT module on Environmental Hazards 

Malignant Hyperthermia (MH) can be caused by succinylcholine and inhaled 

anaesthetics administration (especially halothane). MH is more common in the 

operating room than in the ICU and occurs after general anaesthesia with 

depolarising agents in patients with a mutation in the calcium channel of 

sarcoplasmic reticulum. 

The Neuroleptic Malignant Syndrome (NMS) resembles malignant 

hyperthermia and may occur after single or repeated doses of neuroleptic drugs 

in any patient. It is thought to be a consequence of blockade of dopamine 

receptors from antipsychotic agents such as phenothiazines, thioxanthenes or 

butyrophenones e.g. haloperidol. 

Both MH and NMS inhibit hypothalamic heat-regulation mechanisms 

generating high fever, muscular rigidity, and increased creatine phosphokinase 

concentrations. The main difference between these two clinical entities is the 

initial muscle contraction which is centrally mediated in NMS and peripherally 

mediated in MH. The Serotonin syndrome caused by excessive stimulation of 

the 5-HTIA-receptor by various psychiatric medications may be confused with 

NMS. It may be exacerbated by the concomitant linezolid therapy. 

The central problem of these syndromes is calcium overload in muscle 

consequent on certain stimuli. There is an in vitro (caffeine-halothane 

contracture) muscle test to detect susceptibility to malignant hyperthermia. The 

family of a patient with malignant hyperthermia should be counselled. 

Malignant hyperthermia is a dangerous syndrome and since core temperature 

may exceed 41 °C, severe brain injury and even brain death may occur if the 

condition is unrecognised and untreated. 

The onset of MH is sudden and, in addition to fever, is associated with 

tachycardia, arrhythmias, arterial desaturation and metabolic acidosis; oxygen 

consumption and carbon dioxide production (as measured via capnography) are 

increased. These abnormalities may persist after transfer to the ICU. Mental 

disturbances (progressing to coma), muscular cramps or rigidity and signs of 

altered autonomic function (tachycardia and hypertension) may predominate. 

Creatine phosphokinase (CK) levels are elevated (particularly the muscular MM 

band) as a consequence of rhabdomyolysis, and dangerous hyperkalaemia may 

ensue. Dantrolene is a recognised therapy in addition to cooling and supportive 

measures – see Task 4. 

THINK: Given the lethality of malignant hyperthermia and the existence of a test for 

susceptibility, consider the ways in which the condition might be prevented. What is 

your local practice How would you advise a patient, and the family, after suffering 

from malignant hyperthermia 

[36]


The Lethal Catatonia syndrome is a neuropsychiatric disorder that may overlap 

the neuroleptic malignant syndrome. Fever, stupor, mutism, coma, muscle 

rigidity and autonomic instability are amongst the most prominent features. 

The differential diagnosis includes thyrotoxic crisis, anticholinergic drug 

overdose (anticholinergic syndrome), sepsis, pheochromocytoma and central 

nervous system abnormalities. 

A 47-year-old woman was admitted to the neurological ward 

because of mutism. She had a history of paranoid psychosis. The psychiatrist 

prescribed haloperidol 2.5 mg bd, and a few days later she developed fever 

without a clinically evident infective focus. There was muscle rigidity and 

hypertonia. CK was 300 U/l, slightly elevated. After discontinuation of 

haloperidol and start of oxazepam and dantrolene orally, hyperthermia and 

coma persisted and the patient was transferred to the ICU. Dantrolene was 

continued by the intravenous route, and after some days the patient’s condition 

improved sufficiently for her to be returned to the ward. During the subsequent 

neurological course, discontinuation of dantrolene again elicited a syndrome of 

coma, hyperthermia, hypertonia, accompanied by arterial hypertension and 

hyperglycaemia which was attributed to neuroleptic malignant syndrome. 

However its severity led lethal catatonia to be considered and, at one stage, 

electroconvulsion therapy was considered. Nonetheless, dantrolene was 

recommenced with an efficacious clinical response and a subsequent slow 

withdrawal of the drug was achieved without untoward effect. The case 

illustrates the prolonged nature of the condition and the disadvantages of too 

rapid a discontinuation of dantrolene. 

In a patient with at risk of malignant hyperthermia, the following drugs can be 

given safely: 

Paracetamol (acetaminophen) 

Propofol 

Thiopentone (thiopental) 

Droperidol 

Non-depolarising muscle relaxants 

Neostigmine 

Atropine 

Nitrous oxide 

Benzodiazepines 

Lignocaine. 

For an overview, see: 

Heiman-Patterson TD. Neuroleptic malignant syndrome and malignant 

hyperthermia. Important issues for the medical consultant. Med Clin 

North Am 1993; 77(2): 477–492. PMID 8095087 

Saper CB, Breder CD. The neurologic basis of fever. N Engl J Med 1994; 330(26): 

1880–1886. PMID 7832832 

Hopkins PM. Malignant hyperthermia: advances in clinical management and 

diagnosis. Br J Anaesth 2000; 85(1): 118–128. PMID 10928000 

[37]


Lawrence KR, Adra M, Gillman PK. Serotonin toxicity associated with the use of 

linezolid: a review of postmarketing data. Clin Infect Dis 2006; 42(11): 

1578–1583. Epub 2006 Apr 27. PMID 16652315 

The Malignant Hyperthermia Association of the US, www.mhaus.org, has 

educational material on the website. 

[38]

Task 4. Understanding and treating fever 

4/ UNDERSTANDING AND TREATING FEVER 

Knowledge concerning fever and its influence on disease progression, its relative 

importance in infective and non-infective circumstances and the merit of 

controlling fever is incompletely understood. Although fever control is 

commonplace in medicine and in ICU, there is conflicting evidence on the 

efficacy of such therapy, using paracetamol or NSAID (non-steroidal antiinflammatory 

drugs) for example. Furthermore, infective and non-infective 

fever may differ in their effect. However, an understanding of its pathogenesis is 

useful and its control in highly febrile states is unquestioned. 

Jefferies S, Weatherall M, Young P, Eyers S, Perrin KG, Beasley CRW. The effect 

of antipyretic medications on mortality in critically ill patients with 

infection: a systematic review and meta-analysis. Crit Care Resusc 2011; 

13(2): 125-131. PMID 21627583 

Pathogenesis and pathophysiology 

Several conditions, such as malnutrition, uraemia and immunosuppression, 

may modify the body’s thermoregulatory responses. Uraemic or 

immunosuppressed patients (such as those receiving steroids which are 

inherently antipyretic) may manifest little febrile response to bacteraemia. 

Hence, bacteraemia or infection in these patients may be associated with a less 

marked elevation of body temperature. Modest elevations, or even a fall, in body 

temperature in patients with these conditions may therefore be worthy of 

prompt investigation. 

Rigors can be associated with a sudden rise in body temperature and energy 

expenditure, leading to cardiorespiratory instability. In these circumstances, 

increased dependency on inotropic or ventilatory support may develop and ICU 

patients may suddenly begin to trigger the ventilator or manifest tachycardia 

and hypotension. Sudden wheezing may also point to bacteria entering the 

blood stream, mirroring the effect of increased airway resistance and 

bronchoconstriction seen in animal experiments during the infusion of 

endotoxin. 

The transfusion of contaminated blood products (now rare in regulated 

environments) has been known to result in sudden bacteraemia and thus serve 

as an unintentional model of sepsis. You may wish to read more about 

transfusion-related sepsis in: 

Palavecino E, Jacobs M, Yomtovian R. Bacterial contamination of blood products. 

Curr Hematol Rep 2004; 3(6): 450–455. PMID 15496280 

Rigors are unpleasant for the patient, and the symptoms can be alleviated with 

opioid drugs. 

[39]


Although fever can be regarded as a beneficial component of the 

host response to invading micro-organisms and tissue injury, 

deleterious endocrine and metabolic side effects can occur, 

including increased protein breakdown (catabolism) and 

cerebral damage. The latter is imminent when temperature 

exceeds 42 °C for one hour or longer. The effect of raised 

ambient temperature on the endocrine and metabolic response 

is displayed in the figure: 

Fever is 

normally good 

for the patient 

but can be bad 

Responses of 

different organs to 

fever 

From: Dimopoulos G, Falagas M. Approach to the febrile patient in ICU. Infect Dis Clin 

North Am 2009;23:471–484 

A warm nursing environment (e.g. a fluidised bed at 32 °C) after trauma, burns 

or surgery, for instance, may attenuate these effects. The diminished gradient 

between body and ambient temperature decreases heat production, metabolic 

stress and the catabolic response. See the following reference: 

Ryan DW, Clague MB. Nitrogen sparing and the catabolic hormones in patients 

nursed at an elevated ambient temperature following major surgery. 

Intensive Care Med 1990; 16(5): 287–290. PMID 2212251 

A number of exogenous stimuli (e.g. infections, inflammatory or autoimmune 

diseases, vascular occlusive diseases, drugs) lead to the release of large proteins 

called ‘endogenous pyrogens’ which bind to specific receptors in the preoptic 

region of the anterior hypothalamus where a blood-brain barrier acts as a valve 

permitting the entrance of a limited quantity of these proteins into the brain. 

Pyrogen access and contact with neurons is achieved with the aid of small 

neuronal cells with fenestrated capillaries called ‘circumventricular organs’. 

[40]


This results in a direct response of the neurons within the organum vasculosum 

of the lamina terminalis or of the astrocytes or microglia to circulating cytokines 

such as interleukin-1 and -6, resulting in arachidonic acid metabolite production 

(prostaglandin E2 and thromboxane A2) and upregulation of the thermostatic 

set point. The brain responds by sending signals that activate effector 

mechanisms (through the spinal/supraspinal motor system or throughout the 

sympathetic nervous system), which in turn generate heat production (in brown 

adipose tissue), reduce heat loss and increase core body temperature to match 

the upregulation of the thermostatic set point (see diagram). 

The activation of arachidonic acid metabolites acts as a substrate for the cyclooxygenase-2 

(COX-2) pathway, which in turn leads to further elevation of 

prostaglandin levels, a decreased rate of firing of sensitive neurons and an 

increased heat production. The role of COX-2 is important for the development 

of fever although its activity is inhibited by selective inhibitors, namely nonsteroidal 

anti-inflammatory drugs (NSAIDS) and acetaminophen. Thermal 

imbalance may of course occur in the absence of a reset of the thermostat, when 

heat loss is less than heat production as, for example, in hyperthyroidism, 

salicylate and anticholinergic drug overdose, skin disease and heat stroke. 

Saper CB, Breder CD. The neurologic basis of fever. N Engl J Med 1994; 330(26): 

1880–1886. PMID 7832832 

Luheshi GN. Cytokines and fever. Mechanisms and sites of action. Ann N Y Acad 

Sci 1998; 856: 83–89. PMID 9917868 

Simmons DL, Wagner D, Westover K. Nonsteroidal anti-inflammatory drugs, 

acetaminophen, cyclooxygenase 2, and fever. Clin Infect Dis 2000; 31 

Suppl 5: S211–218. PMID 11113025 

Leon LR. Invited review: Cytokine regulation of fever: studies using gene 

knockout mice. J Appl Physiol 2002; 92(6): 2648–2655. PMID 12015385 

The following figure illustrates the time course of thermostat resetting and the 

consequences for the time course of body temperature (green line) as a function 

of heat gain, when the thermostat is set at a higher level (grey line), and of heat 

[41]


loss, when the thermostat is set again to the normal level. Shivering occurs in 

the first period (heat gain) and sweating in the second one (heat loss). 

For every degree centigrade increase in temperature, oxygen demand 

and energy expenditure are said to increase by about 6–10%. Sedation, 

anaesthesia and cooling may decrease oxygen demand while shivering has the 

opposite effect. When cooling results in shivering, oxygen demand may also 

increase. 

Treating fever 

There is ongoing debate as to the benefit of physical or pharmacologic lowering 

of core temperature exceeding 40 °C with the attendant risk of cerebral damage. 

Indeed, we previously noted that brain damage may be anticipated when the 

duration of fever of >42 ° C exceeds one hour; higher temperatures may be 

tolerated for much shorter periods. In patients with sepsis, NSAIDs may 

attenuate the febrile response. In patients with severe sepsis, ibuprofen has 

been shown to lower body temperature, tachycardia and lactic acidosis but no 

other haemodynamic, respiratory, or survival benefit was demonstrated. 

Although this type of drug may be associated with renal dysfunction, 

particularly in hypovolaemic subjects, the administration appeared safe. 

Antipyretics may also include paracetamol (acetaminophen) (orally or rectally) 

or diclofenac (0.04 mg/kg/h continuous i.v. infusion) if available. Overzealous 

treatment with NSAIDs may augment inflammatory responses. 

Mackowiak PA. Physiological rationale for suppression of fever. Clin Infect Dis 

2000; 31 Suppl 5: S185–189. PMID 11113022 

Manthous CA. Toward a more thoughtful approach to fever in critically ill 

patients. Chest 2000; 117(3): 627–628. PMID 10712983 

[42]


Gozzoli V, Treggiari MM, Kleger GR, Roux-Lombard P, Fathi M, Pichard C, et al. 

Randomized trial of the effect of antipyresis by metamizol, propacetamol 

or external cooling on metabolism, hemodynamics and inflammatory 

response. Intensive Care Med 2004; 30(3): 401–407. Epub 2004 Jan 13. 

PMID 14722642 

Thompson HJ. Fever: a concept analysis. J Adv Nurs 2005; 51(5): 484–492. 

PMID 16098165 

THINK: Are there any specific infective conditions in which NSAIDs contribute to 

the cause 

Malignant hyperthermia, neuroleptic malignant 

syndrome and lethal catatonia 

The treatment of these syndromes consists of eliminating underlying causes, 

cooling (initially by infusion of cold intravenous fluids – see below), rehydration 

and benzodiazepines. Intravenous dantrolene sodium to treat muscle 

contraction (1 mg/kg bolus) which may be repeated during the course of the 

illness is followed by 4–8 mg/kg per day in 3–4 doses, orally, for 3–4 days as 

ongoing prophylaxis. Bromocriptine, other dopamine agonists (levodopa) and 

amantadine may also be used for neuroleptic malignant syndrome. 

Electroconvulsive therapy may be necessary for catatonia if refractory to drug 

treatment. NSAIDs are not efficacious. 

Cooling techniques 

In highly febrile states(usually quoted as >40 C), treatment of the fever per se 

(in addition to managing the underlying cause) may be required. Cooling is also 

a recognised therapy of coma following cardiac arrest. 

The cooling techniques in practice in ICU include the conventional cooling 

(rapid infusion of 30 ml/kg cold fluids, ice and/or coldpacks), cooling with 

water circulating blankets, air circulating blankets, water circulating gel-coated 

pads and intravascular heat exchange systems. Cooling blankets are used mainly 

in patients with hyperthermia. The use of such devices lowers oxygen 

requirement in paralysed or heavily sedated patients, thereby reducing demand 

on the cardiorespiratory system. In unparalysed patients, cooling may have the 

undesirable effect of inducing shivering and a rise in oxygen uptake to augment 

heat production and compensate for the increased loss. Otherwise, temperature 

fluctuations and rebound hyperthermia may occur. There are some reports on 

the maintaining of the target temperature where intravascular cooling was 

superior to all other cooling methods. See the references below. 

Plaisance KI, Mackowiak PA. Antipyretic therapy: physiologic rationale, 

diagnostic implications, and clinical consequences. Arch Intern Med 

2000; 160(4): 449–456. PMID 10695685 

Hasday JD, Garrison A. Antipyretic therapy in patients with sepsis. Clin Infect 

Dis 2000; 31 Suppl 5: S234–241. PMID 11113029 

[43]


Hoedemaekers CW, Ezzahti M, Gerritsen A, van der Hoeven JG. Comparison of 

cooling methods to induce and maintain normo- and hypothermia in 

intensive care unit patients: a prospective intervention study. Crit Care 

2007; 11(4): R91. PMID 17718920 

Unintended cooling may occur during continuous renal replacement therapy 

(CRRT) techniques, when the extracorporeal circuit is exposed to environmental 

(room temperature) and heat is lost via the ultrafiltrate. The decrease in core 

temperature may contribute to vascular stability, and hypotensive episodes may 

be shortened or prevented. These effects may explain the reversal of the 

hypermetabolic and vasodilated state sometimes noted in patients with septic 

shock and acute renal failure on whom CRRT is initiated, and may be 

independent of possible filtration or absorption of the toxic mediators of sepsis, 

including cytokines. 

Q. Does cooling, by continuous haemofiltration techniques, decrease 

oxygen requirements of the body Is the cooling effect dependent on 

the ultrafiltration rate Explain your answer. 

A. Yes, the fall in temperature during high (and less so during low) ultrafiltration rates 

results in a decrease in O2 uptake. This may partly explain the beneficial 

haemodynamic effect of the technique in patients with septic shock, when clearance of 

absorption via the artificial membrane of cytokines and other inflammatory mediators 

is minimal. 

For information on haemofiltration techniques: 

Kamijo Y, Soma K, Sugimoto K, Tsuruta H, Ohwada T. The effect of a hemofilter 

during extracorporeal circulation on hemodynamics in patients with SIRS. 

Intensive Care Med 2000; 26(9): 1355–1359. PMID 11089764 

John S, Griesbach D, Baumgärtel M, Weihprecht H, Schmieder RE, Geiger H. 

Effects of continuous haemofiltration vs intermittent haemodialysis on 

systemic haemodynamics and splanchnic regional perfusion in septic 

shock patients: a prospective, randomized clinical trial. Nephrol Dial 

Transplant 2001; 16(2): 320–327. PMID 11158407 

[44]


CONCLUSION 

This module helps to recognise and manage the cause of febrile diseases in 

critically ill patients. Early recognition and proper diagnosis is dependent on the 

differentiation between infective and non-infective causes. There will be 

differences of emphasis between the surgical and the medical patient and also in 

the immune suppressed patient. A full clinical evaluation and supplementary 

imaging or other testing as indicated by clinical evaluation, is the most 

important process in identifying the cause of the fever. A supplementary step in 

identifying an infective origin of a fever is having knowledge of surveillance 

cultures from the patient and also of the local microbial epidemiology and 

antibiogram pattern. Such knowledge can facilitate early adequate and 

appropriate empiric treatments, pending specific information from targeted, 

patient diagnostic sampling. Cooling, when indicated, is best accomplished in 

an easy, controllable, minimally invasive and well-tolerated way. Little is known 

about the optimal method of temperature control. The cooling devices in 

practice include the water circulating external cooling methods, the air 

circulating external cooling device, the water circulating external cooling device 

using self-adhesive gel-coated pads and the intravascular heat exchange system. 

[45]

Self-assessment 

SELF-ASSESSMENT 

EDIC-style Type K 

1. Clinical signs or measurements associated with fever include: 

A. Rigor 

B. Increased oxygen consumption 

C. Decreased energy expenditure 

D. Tachycardia 

2. Fever in a patient on the second day after major trauma: 

A. Is a common phenomenon 

B. Is abnormal when it occurs 

C. Is a strong indication of an infection 

D. Usually does not exceed 38.5 C 

3. Fever above 39 C in a patient with head injury 

A. Is normal and does not need treatment 

B. May increase cerebral damage 

C. Should immediately be treated with broad-spectrum antibiotics 

D. Acetaminophen (paracetamol) is the first treatment of choice 

4. Clinical manifestations of heat stroke include 

A. Core temperature above 42 C required for diagnosis 

B. Reduced consciousness 

C. Convulsions 

D. Hypertension 

5. The Neuroleptic Malignant syndrome is caused by: 

A. Mutation in calcium channels of sarcoplasmatic reticulum 

B. Excessive stimulation of beta receptors in the musculature 

C. Drug-induced blockade of dopamine receptors 

D. Inactive Na/K channels in myocytes 

6. Safe drugs to use in a patient at risk of malignant hyperthermia 

include: 

A. Thiopentone (thiopental) 

B. Propofol 

C. Succinylcholine 

D. Lignocaine 

7. Important pathways responsible for increasing heat production and 

temperature include: 

A. Activation of arachidonic acid metabolites 

B. Down-regulation of prostaglandin levels 

C. Activation of α receptor activity 

D. Increase in cyclo-oxygenase-2 (COX-2) pathway activity 

[46]


EDIC-style Type A 

8. The definition of fever in an ICU patient is temperature above: 

A. 38.0 

B. 38.3 

C. 38.5 

D. 38.8 

E. 39.0 

9. The part of the brain mainly involved in the pathogenesis of fever is: 

A. Hypothalamus 

B. Hippocampus 

C. Cerebellum 

D. Pons 

E. The 4th ventricle 

10. Core temperature is measured in all of the following anatomical sites 

EXCEPT the: 

A. Pulmonary artery 

B. Bladder 

C. Oesophagus 

D. Rectum 

E. Femoral artery 

11. Non-infective causes of fever in an ICU patient include all of the 

following EXCEPT: 

A. Thromboembolism 

B. Drugs 

C. Central venous catheter 

D. Myocardial infarction 

E. Acute pancreatitis 

12. Central venous catheter (CVC)-related sepsis is a frequent cause of 

sudden fever in the ICU. This complication is most commonly reported to 

occur at a rate of: 

A. 0–2 episodes/1000 CVC days 

B. 2–15 episodes/1000 CVC days 

C. 16–30 episodes/1000 CVC days 

D. 30 episodes/1000 CVC days 

E. 15–20% of CVCs 

13. Effective prevention strategies to reduce central venous catheter 

(CVC)-related sepsis includes all of the following EXCEPT: 

A. Prophylactic antibiotic intravenously 

B. Use of alcohol-based skin disinfection 

C. Use of sterile gloves and gown 

D. Protocol-based care and maintenance of the CVC 

E. Antibiotic coating of the CVC 

[47]

14. The most likely cause of fever (of new onset) and infection in a 

critically ill surgical patient is: 

A. Wound infection 

B. Catheter-related infection 

C. Nosocomial pneumonia 

D. Urinary tract infection 

E. Sinusitis 

15. Treatment of heat stroke includes the following EXCEPT 

A. Infusion of cold fluids 

B. External cooling 

C. Intravascular cooling 

D. Dantrolene 

E. Supportive cardiovascular therapy 


[48]


Self-assessment answers 

Type K 

Q1. Q2. Q3. Q4. Q5. Q6. Q7. 

A. T 

A. T 

A. F 

A. F 

A. F 

A. T 

A. T 

B. T 

B. F 

B. T 

B. T 

B. F 

B. T 

B. F 

C. F 

C. F 

C. F 

C. T 

C. T 

C. F 

C. F 

D. T 

D. T 

D. T 

D. F 

D. F 

D. T 

D. T 

Type A 

8. Answer B is correct 

9. Answer A is correct 

10. Answer D is correct 

11. Answer C is correct 

12. Answer B is correct 

13. Answer A is correct 

14. Answer C is correct 

15. Answer D is correct 

[49]

Patient Challenges 

PATIENT CHALLENGES 

A 63-year-old woman, previously operated on for ischaemic heart 

disease, underwent elective redo coronary artery bypass graft (CABG) and mitral 

valve repair. Because of persistent hypoxaemia after 24h, she remained intubated and 

ventilated. Haemodynamic evaluation revealed an elevated pulmonary capillary 

wedge/occlusion pressure (PCWP/PAOP) of 22 mmHg. The ejection fraction was 

50% but mild mitral regurgitation, as judged by echocardiography, had been present 

since the day of surgery. Vasodilators and diuretics were administered with a 

definitive decrease in wedge pressure and there was an improvement in the blood gas 

values and ventilatory parameters – at an inspiratory O 2 fraction of 0.3, a positive 

end-expiratory pressure (PEEP) of 5 cm H 2 O and a pressure support of 10 cm H 2 O, 

the arterial PO 2 was 9.31 kPa (66 mmHg) and PCO 2 5.05 kPa (36 mmHg). However 

on the fourth postoperative day, the patient developed a fever. 

The chest X-ray was abnormal: right lower lobe atelectasis and bilateral infiltrations 

were evident but no clear differences from the previous examinations were noted. 

The white blood cell count was 13.0 x 10 9 /l (previously 8.5 x 10 9 /l) with 80% 

neutrophils and 9% band forms. The tracheal aspirate was not macroscopically 

abnormal. The Gram stain registered the presence of bacteria but the sample was 

judged unreliable because of the presence of numerous epithelial cells. The figures 

below show the chest X-ray appearance on admission to the hospital and on the day 

when fever appeared. 

Assessment of fever 

Evaluation of bronchial aspirates 

X-ray on admission 

X-ray when fever appeared 

Q. The patient is probably developing an infective problem. Which type of further 

examination (if any) would you do 

A. A strategy of identifying the source of infection should be undertaken followed by 

appropriate treatment. First and foremost, clinical examination must take into 

account all the possible sources of infection in a postoperative patient. 

[50]


Q. Outline the particular features of the clinical evaluation of sepsis that would be 

appropriate in this surgical patient. 

A. Wound margins are examined for erythema or infiltration of the soft tissues and 

drains (including their content) are checked. Tracheal and pharyngeal secretions are 

reviewed together with catheter sites for possible cellulitis. 

Q. In this intubated patient, might any additional respiratory investigation be 

indicated 

A. Bronchoalveolar lavage or protected specimen brush specimens may be examined 

if pneumonia is suspected and initial tracheal aspirates are unhelpful. 

Since the patient is still invasively monitored, catheter infection has to be considered. 

Even in the absence of external signs of infection (local inflammation or pus 

formation) catheter infection should always be strongly suspected especially if there 

is no other evident source of infection or a positive blood culture has been obtained. 

Blood cultures, urinalysis and, even though fever has occurred only a few days 

postoperatively, an echocardiogram may be useful. If endocarditis is suspected, 

transoesophageal echocardiography is utilised when transthoracic (valvular) images 

are suboptimal. 

Ryan EW, Bolger AF. Transesophageal echocardiography (TEE) in the evaluation 

of infective endocarditis. Cardiol Clin 2000; 18(4): 773–787. PMID 

11236165 

While waiting for the microbiological results, the temperature rose to 39.3 °C and the 

patient developed tachycardia and respiratory distress. Ventilatory assistance was 

increased (inspiratory O 2 fraction 0.4; pressure support 15 cm H 2 O). Antibiotics were 

started after a septic work-up which included bronchoscopy and distal bronchial 

sampling by bronchoalveolar lavage (BAL). 

Q. Are there particular indications in this patient to treat fever 

A. Patients with impaired myocardial reserve tolerate the increase in oxygen demand 

poorly because of the increased myocardial workload. While keeping clearly in mind 

that the key issue is the treatment of the underlying infection, a symptomatic 

approach to fever may be indicated. Several drugs have been proposed for the 

treatment of fever and no clear consensus has been reached as to the drug of choice. 

However, it is mandatory to consider their potential side effects. 

Antipyretic treatment 

Antipyretic drugs 

[51]


The urgent Gram stain and microscopy of the protected bronchial specimen has now 

revealed Gram-negative rods >10 4 CFU/ml together with leukocytes and bronchial 

cells; no epithelial cells are detected. 

Assessment of bronchial aspirates 

Q. Would you consider the specimen suggestive of infection 

A. In an intubated patient, in the presence of systemic and radiological signs 

suggestive of infection and with a microbiological specimen compatible with the 

diagnosis of pneumonia, the presumptive diagnosis of ventilator-associated 

pneumonia is appropriate and antibiotic therapy should be immediately initiated. 


After a further three days, the fever appears to be well controlled below 38 °C and 

does not require continued antipyretic therapy. The patient was progressively weaned 

from the ventilator, the chest X-ray which had developed obvious consolidation was 

now clearing (figure below); she had stable cardiovascular function and she was 

extubated successfully. Pseudomonas aeruginosa grew in the BAL culture. 

The chest radiograph of the patient with residual basal infiltrate (right greater than 

left) 

While transfer to the surgical ward was being planned, the patient became confused 

and agitated, her temperature rose to 39 °C and she became hypotensive: 

70/40 mmHg. The electrocardiogram did not show a change compared to previous 

ones. The echocardiogram showed a normally contracting ventricle without wall 

motion abnormalities. Fluid resuscitation was started and despite a fluid load of 

1500 ml, the arterial blood pressure rose to only 80/40 mmHg, and the patient 

became anuric. The oxygen saturation started to decrease despite supplemental O 2 by 

mask. Nasotracheal intubation was performed and vasoactive drugs were 

administered. 

[52]


A cardiogenic cause to explain the clinical picture was excluded by previous 

examinations. In a patient with reduced cardiovascular reserve however, a pulmonary 

artery catheter was inserted to guide the titration of fluid and haemodynamic 

therapy. With dopamine at 10 mcg/kg/min and a further 1000 ml of colloids to 

obtain a wedge pressure of 18 mmHg, the arterial blood pressure rose to 120/80 

mmHg. However, the patient remained anuric for three additional hours and despite 

its known lack of efficacy as a renal conservation measure, frusemide was 

administered. 

See PACT module on Oliguria and anuria (AKI part 1) 

Blood tests were as follows: white blood cell count 35 x 10 9 /l (band forms 20%; 

neutrophils 80%); platelets 150 x 10 9 /l; haematocrit 35%; aspartate 

aminotransaminase (ASAT AST) 30 U/l; alanine aminotransaminase (ALAT ALT) 50 

U/l; bilirubin 1.4 mg/dl; creatinine 176.8 micromol/l; blood urea nitrogen 14.28 

mmol/l; lactate 10 mmol/l; sodium 143 mmol/l; potassium 4.3 mmol/l. Blood was 

taken for culture. 

Haemodynamic monitoring during septic shock: see PACT module on 

Haemodynamic monitoring. 

Q. What other source of infection do you look for 

A. The development of such a dramatic clinical course in a postoperative patient 

should always remind us that infection can arise from the surgical wound or at the 

valvular level. Moreover the patient still has devices in situ such as a central venous 

catheter and urinary catheter. Clostridium difficile infection (non-diarrhoeal) needs 

to be included in the differential diagnosis. 

SIRS, infection, sepsis and shock definition 

Despite the administration of paracetamol (acetaminophen), the patient remained 

febrile (39 °C) and although restoration of blood pressure stability allowed some 

weaning of dopamine, tachycardia (135 bpm) persisted. The electrocardiogram 

showed diffuse ST segment depression which appeared to be tachycardia-associated. 

Because the patient has been in hospital for a number of days, had not responded to 

current antimicrobial therapy, and catheter or surgical site infection were 

possibilities, intravenous vancomycin was added and deeper sedation and analgesia 

were instituted. 

The presence of fever and leukocyte abnormalities together with shock and organ 

dysfunction were consistent with a diagnosis of septic shock. The appropriate 

response included: first, resuscitation and treatment of the haemodynamic 

derangement; second, rapid investigation to detect the source of infection and third, 

treatment and consideration of any additional issues. 

The first goal had been successfully achieved. The second goal in this complicated 

patient still requires further consideration. 

[53]


Treatment of septic shock 

Complications of pneumonia 


Q. The patient had pulmonary infection that is apparently resolving. Which further 

respiratory diagnostic test would you consider and why 

A. It is possible that a pulmonary abscess or empyema is complicating the original 

pneumonia. A CT scan may be indicated. 

In the presence of such complex radiological findings where abnormal features exist, 

a CT scan may be helpful to diagnose or exclude complications (see figure below). A 

CT scan revealed bilateral pleural effusion and posterior bilateral consolidation. The 

pleural effusion was drained and cultured and therapy with vancomycin, ceftazidime 

and amikacin is continued. 

CT scan indications 

Q. The pleural fluid proves not to be infected and the source of the infection remains 

unclear. What additional diagnoses might you now consider 

A. The presence of a nasotracheal tube should raise the possibility of sinusitis and CT 

scan of the head might be considered. Other possibilities include: 

Endocarditis 

Has sufficient time elapsed to see an improvement in infective endocarditis, if 

present 

Has serial physical examination shown evidence of new murmurs; further 

transoesophageal echocardiographic examination (TOE) should help in detecting 

abnormal valvular vegetations if suspected 

Vascular CVCs are always a possible source of infection 

Wound infection including mediastinitis 

Non-diarrhoeal Clostridium difficile is a possibility 

Urinary tract infection is kept under review 

Q. Which type of further therapeutic manoeuvre might be considered to treat the 

remaining fever (of 39 °C) 

A. External cooling should be considered. 

[54]


Cooling techniques 

External cooling is started and temperature is decreased to 36 °C. The heart rate 

decreased to 100 bpm. ECG disturbances disappear. Arterial blood pressure stabilises 

at 120/60 mmHg. Cardiac output is 3.5 l/min. Mixed venous oxygen saturation is 

72%. 

Staphylococcus aureus was cultured in the blood culture. The central venous catheter 

was removed and catheter tip sent for culture. Transoesophageal echocardiography 

was performed and was negative for vegetations on the cardiac valves. The culture of 

the tip of the catheter also (in addition to the blood) grew Staphylococcus aureus, 

thus complying with the criteria for catheter-related blood stream infection (CRBSI) 

or CRI3 by HELICS criteria. 

Diagnostic criteria for catheter-related infection 

The antibiotic therapy is rationalised to target Staphyloccoccus aureus CRBSI. The 

haemodynamic support is progressively decreased and cooling is stopped after 24 

hours and deep sedation and paralysing drugs are discontinued. Two days after the 

shock episode, the patient was weaned off the ventilator. After three days the patient 

was afebrile, the white blood cell count was 12 x 10 9 /l and she was eventually 

transferred to the ward to continue out her course of Flucloxacillin for 

Staphylococcus aureus CRBSI. 

Cobb DK, High KP, Sawyer RG, Sable CA, Adams RB, Lindley DA, et al. A 

controlled trial of scheduled replacement of central venous and 

pulmonary-artery catheters. N Engl J Med 1992; 327(15): 1062–1068. 

PMID 1522842 

Renaud B, Brun-Buisson C.; ICU-Bacteremia Study Group. Outcomes of primary 

and catheter-related bacteremia. A cohort and case-control study in 

critically ill patients. Am J Respir Crit Care Med 2001; 163(7): 1584–1590. 

PMID 11401878 

Mermel LA, Allon M, Bouza E, Craven DE, Flynn P, O'Grady NP. et al. Clinical 



of America. Clin Infect Dis 2009;49:1-45. PMID 19489710 

[55]


Following a motorcycle accident, a 24-year-old man was admitted to the 

Emergency Department. Pelvic and femoral fractures were confirmed on 

radiographic examination. On admission the patient was hypotensive; resuscitation 

was promptly instituted. Subsequently his haematocrit fell to 22%. Blood alcohol was 

negative. Twelve hours later, while awaiting femoral stabilisation, he became 

severely dyspnoeic and arterial blood gas analysis (breathing room air) revealed an 

arterial PO 2 of 4.65 kPa (33 mmHg), PCO 2 of 2.66 kPa (19 mmHg) and pH of 7.5. 

Oxygen was administered; the chest X-ray revealed diffuse bilateral interstitial 

infiltrates. Platelet count was 55 x 10 9 /l. He was transferred to the ICU where after 

an initial trial of non-invasive ventilation (pressure support 15 cm H 2 O, inspiratory 

O 2 fraction 1.0 with PEEP of 15 cm H 2 O), his oxygenation improved. No 

haemodynamic derangement developed but the patient became febrile (38.8 °C). 

Non-invasive ventilation (see PACT Module on Mechanical ventilation) 

Q. Are there reasonable grounds to suspect a non-infective problem 

A. In this trauma patient, fever could have several non-infective explanations 

including: 

Haematoma 

Drug reaction 

Fat embolism 

The patient probably has extensive retroperitoneal haematoma due to the pelvic 

fracture and this, per se, could explain the febrile response. 

Other causes of fever 

Blood platelets are, however, considerably decreased and although this could be 

explained by dilution following resuscitation, the fat embolism syndrome should also 

be suspected. Clinical diagnosis and physical examination is mandatory looking for 

petechiae, characteristically located near the axilla, around the nipples, and in the 

conjunctivae and fundi. 

Q. Is the distinction between infective and non-infective causes easy If infection is 

still suspected, are there additional blood tests which might be helpful in the 

diagnosis of infection 

A. There can be lack of clarity initially as infection is associated with non-specific 

symptoms and signs. Bacteriological evidence of infection can be absent despite the 

presence of relevant symptoms and signs. On the other hand, positive microbiological 

results may be due to contamination. The diagnosis in these situations can be further 

supported by specific markers including procalcitonin, C-reactive protein, urine test 

for lipiduria and examination of the tracheal aspirate for macrophages with lipid 

globules. 

CRP was within normal range and the fever gradually disappeared. The patient was 

stable and a weaning protocol was initiated. Microbiological results were all negative. 

After three days, the patient was afebrile, the clinical diagnosis of the fat embolism 

[56]

syndrome was clear and antibiotics, which had been started for the fever, were 

stopped. 


Value of laboratory diagnosis of fever 

[57]


A 64-year-old female psychiatric patient was scheduled for 

cholecystectomy. There was no history of previous surgical procedures, abnormal 

reactions to drugs, asthma or other cardiorespiratory diseases. The arterial blood 

pressure was 130/70 mmHg and heart rate 96 bpm. The electrocardiogram and chest 

X-ray were normal. The haemoglobin content was 14 g/dl and serum electrolytes and 

arterial blood gases were in the normal range. On the day of surgery, the patient 

received diazepam 10 mg and atropine. Anaesthesia was induced with propofol and 

suxamethonium (succinylcholine). Desflurane 2%, nitrous oxide and fentanyl were 

used to maintain anaesthesia. Twenty minutes after induction of anaesthesia, the 

arterial blood pressure rose to 170/100 mmHg and the heart rate to 120 bpm. 

Additional fentanyl did not control the hypertension. The end-tidal PCO 2 increased 

from 4.79 kPa (34 mmHg) to 5.72 kPa (40 mmHg) without apparent technical 

problems with the anaesthetic apparatus. The arterial blood gas analysis was as 

follows: PO 2 7.19 kPa (51 mmHg), PCO 2 7.19 kPa (51 mmHg) and pH 7.21 

Q. Pneumothorax was immediately excluded on clinical grounds. The increase in 

arterial PCO 2 was at first thought to be responsible for the clinical findings. Could the 

hypercarbia account for the physiological pattern evident 

A. Yes. Hypertension and tachycardia frequently accompany hypercarbia and 

acidosis. 

Arterial PCO 2 may increase for two reasons: decreased elimination and increased 

production. The former and more common is due to reduction in alveolar ventilation, 

possibly from some malfunction of the anaesthetic breathing circuit. Increased 

production can happen during a laparoscopic procedure when CO 2 is used to distend 

the abdominal cavity or when endogenous CO 2 production is increased (e.g. 

associated with fever, adrenal hormone excess and malignant hyperthermia). 

Technical reasons for decreased CO 2 elimination were excluded, as were surgical 

explanations for increased CO 2 production. 

Denborough M. Malignant hyperthermia. Lancet 1998; 352(9134): 1131–1136. 

PMID 9798607 

Inspiratory O 2 fraction and ventilation were increased. Muscle rigidity became 

evident and core temperature rose to 40 °C. 

Q. What is the most likely diagnosis at this time 

A Malignant hyperpyrexia 

Anaesthesia was discontinued. Iced bags were positioned over the patient and 

dantrolene was administered. Core temperature started to decrease after iced gastric 

lavage was initiated. The patient was admitted to the ICU and the temperature 

aggressively controlled. She later awoke and was eventually extubated without 

untoward sequelae. A more detailed history taken from a brother living abroad 

revealed that malignant hyperthermia was diagnosed in the patient’s mother during a 

hysterectomy performed 15 years earlier. 

[58]


An 18-year-old male with congenital hydrocephalus, permanent celioperitoneal 

drainage and epilepsy controlled with lamotrigine and 

oxcarbazepine, was admitted to the hospital because of respiratory failure due to 

lower respiratory tract infection. He needed ICU admission for endotracheal 

intubation, mechanical ventilation and central venous cannulation. During the 

seventh ICU day, he developed septic shock and blood cultures revealed three 

positive consecutive cultures for both methicillin-resistant Staph. aureus (MRSA) 

and Candida albicans. The tip of the central venous catheter (after removal) was also 

positive for Candida albicans. 

The main risk factors for invasive fungal infection development were considered the 

prolonged use of antibiotics and catheter devices, and the moderately prolonged stay 

in the ICU. The course of the fever, despite the administration of broad-spectrum 

antimicrobials including vancomycin and antifungals, is shown below. No specific 

temperature lowering measures were required. 

The main findings from the detailed diagnostic evaluation of the patient included: 

CT scan of the lungs – infiltrates compatible with pneumonia 

CT scan of the abdomen – negative 

Cerebrospinal analysis – no signs of infection 

WBC: 12.4 x 10 9 /L, Biochemistry: unremarkable 

Ophthalmology – negative for Candida endophalmitis 

[59]


 

Τransthoracic and transoesophageal echocardiography – no signs of 

endocarditis. 

The final diagnosis from thoracic CT (see below) was: Septic thromboplebitis of the 

left subclavian and jugular vein due to MRSA and Candida albicans. 

The thrombophlebitis with its septic complication was not associated with a 

prolonged central venous cannulation while no MRSA colonisation that might have 

predisposed to this infection has been detected from patient’s ICU admission 

surveillance screening. 

The patient received treatment with linezolid 600mg x 2 and anidulafungin 100mg x 

1 IV (after a loading dose of 200 mg) for three and two weeks (for MRSA and C. 

albicans respectively). A step down modification of the treatment to linezolid at the 

same dose and voriconazole (4 mg/kg x 2) orally was performed at the time of 

discharge from ICU to the ward with an anticipated total duration of antimicrobial 

therapy of eight and 12 weeks respectively. She remained under the ongoing 

consultative advice (including monitoring for drug toxicity – especially for linezolid) 

of the microbiological/infectious diseases team. 

The patient was also anticoagulated but this was stopped at six months after his 

discharge after follow up evaluation was negative for signs and symptoms of 

thrombophlebitis. 

On reflection, pyrexia is a frequent observed clinical sign in the ICU. The 

four patients described represent examples of the diagnostic and therapeutic 

challenge to ICU clinical staff when confronted with this problem. Although pyrexia is 

[60]


often a transient, minor sign it may be the harbinger of serious illness. Rapid 

identification of the underlying cause is imperative if life-threatening complications 

are to be avoided and differentiation between infective and non-infective causes is a 

priority. It is important to obtain samples for microbiological evidence in patients 

with suspected infection before starting antibiotic treatment although in the critically 

ill, when severe sepsis is present, a ‘best guess’ approach, based on thorough clinical 

evaluation, will be indicated pending results from the microscopy/culture 

examinations. Source control of infection may also require consideration of surgical 

or radiological intervention. 

You will have noted the importance of clinical acumen in reaching diagnostic and 

therapeutic decisions. No fool proof laboratory tests are available although selected 

investigations can be helpful, particularly when part of a collaborative strategy with 

colleagues from other disciplines. 

[61]

Pyrexia - PACT - ESICM

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?