19th April 2013, Volume 126 Number 1373

Robert Weinkove, Jennifer Clay, Catherine Wood

Severe neutropenia is a risk factor for sepsis.1 Febrile neutropenia is commonly defined as a single fever of ≥38.3°C or a temperature of ≥38°C for at least 1 hour, in the context of a neutrophil count of <0.5×109/L, or <1.0×109/L with the expectation of a decline to <0.5×109/L in subsequent days.2

Febrile neutropenia occurs in the majority of patients undergoing acute leukaemia induction and autologous and allogeneic stem cell transplantation,3,4 and carries a high mortality without prompt antibiotic treatment.1 Guidelines for the investigation and antimicrobial treatment of neutropenic fever have been published.2,5,6

Fever is a natural response to infection, and may be beneficial to the outcomes of sepsis: compared to 37°C, temperatures within the febrile physiological range inhibit in vitro growth of some bacteria,7 and enhance antimicrobial sensivitity.8

On the other hand, antipyretics such as paracetamol improve patient comfort,9 and may have favourable haemodynamic effects.10 The authors’ experience suggests that the attitudes of haematology clinicians to cooling measures in patients with neutropenic fever vary, with some encouraging, and others discouraging, the use of cooling measures.

We aimed to assess the attitudes of haematology clinicians in New Zealand to the management of fever in patients with febrile neutropenia using a practice survey.


A survey asking respondents about their management approach to three clinical vignettes was designed. The first scenario was of a neutropenic patient at home asking their clinician whether they could use paracetamol as an analgesic: respondents were asked to state whether they would advise the patient to take paracetamol regularly, as needed, or to avoid paracetamol.
The second and third scenarios related to a patient with severe neutropenia and fever who has already commenced first-line antimicrobials: respondents were asked at which temperature they would intervene with cooling measures (with options in 0.5°C bands) if the patient were asymptomatic, or symptomatic with rigors. Respondents were then asked which physical or pharmacological cooling measures they would use, and whether paracetamol would be administered as needed or regularly, and via which route.
Respondents were asked at what time point they would use fever as a determinant of empiric antimicrobial change in a patient with febrile neutropenia. Finally, respondents were asked whether they felt a clinical trial of antipyretic treatment in febrile neutropenia was warranted, and whether they would be willing to enrol their patients in such a trial. The survey is in Appendix 1.
The survey was piloted in the authors’ own haematology department, and changes made to improve clarity, and add to response options. A link to the online survey was then emailed to all current medical and nursing members of the New Zealand branch of the Haematology Society for Australia and New Zealand (HSANZ), to all haematology trainees in New Zealand, and to a nursing representative at each haematology centre in New Zealand, for distribution to other nursing staff.
Survey responses were collected using an online survey tool (SurveyMonkey.com, Palo Alto, California, USA), exported to an Excel spreadsheet (Version 12.3.0, Microsoft Corporation, Redmond, Washington, USA), and analysed using Prism statistical software (Version 5.0d, GraphPad Software, La Jolla, CA, USA). All data were analysed using non-parametric statistical tests. The a priori subgroups of doctor and nurse professionals were analysed separately. Temperature thresholds (in 0.5°C bands) were treated as continuous variables for analysis. A probability value of p < 0.05 was considered significant. This online practice survey was classified as low-risk, and did not require formal ethical review according to the guidelines of the Central Regional Ethics Committee of New Zealand.


Eighty-eight responses were received, from 20 consultant haematologists, 14 haematology trainee doctors, 18 senior nurses (charge nurses, clinical nurse specialists or nurse educators), and 36 ward or day unit nurses. Approximately 45 consultant haematologists and 25 haematology registrars are practising in New Zealand (Dr Bart Baker, personal communication), giving an overall response rate of 48% among doctors. The total number of nurses practising in haematology in New Zealand is not known. The number of respondents from each professional group, and duration of practice in clinical haematology, is given in table 1.

Table 1. Survey respondents
Professional group
Respondents (n)
Haematology experience (years); median (range)
Consultant Haematologists
14 (6–30)
Haematology trainee doctors
1 (0.33–5)
Charge nurses, Clinical nurse specialists, nurse educators
11.5 (5–27)
Ward and day unit nurses
5.5 (0.25–20)

Sixty-eight percent of doctors and 44% of nurses indicated that they would allow a neutropenic patient to take paracetamol as needed for pain, and a further 15% of doctors and 7% of nurses would allow a neutropenic outpatient to take regular paracetamol. Nurses were significantly more likely than doctors to advise a neutropenic patient to avoid paracetamol (p < 0.05, Fisher’s exact test).

Figures 1A and 1B show the temperature thresholds at which respondents would intervene to lower temperature in an asymptomatic and a symptomatic patient with febrile neutropenia, respectively.

For asymptomatic patients, the temperature threshold for intervention varied widely between respondents, but the threshold was significantly higher for doctors than for nurses (median 39°C for doctors, 38.5°C for nurses; p<0.01, Mann Whitney test).

For symptomatic patients, the median intervention threshold was 38°C for both doctors and nurses (difference not significant). Considering all survey respondents, there was no significant correlation between years of experience in haematology and temperature intervention threshold for either symptomatic or asymptomatic patients (data not shown).

Figure 1. Reported temperature intervention thresholds in severe neutropenia


Fifty-five percent of respondents favoured paracetamol as the first line intervention to reduce temperature, with a further 44% selecting physical cooling measures. Forty-five percent of respondents favoured physical cooling as the second line intervention, with 30% selecting paracetamol. Sixty-eight percent of respondents selected pethidine as the third line intervention. Other measures, such as non-steroidal anti-inflammatory drugs (NSAIDS), cyclo-oxygenase 2 (COX-2) inhibitors, and other opiates, were favoured by fewer respondents. These data are presented in Figure 2.

Figure 2. Preferred temperature-lowering interventions in febrile neutropenia


Eighty-three respondents answered the question about frequency of paracetamol administration in inpatients with febrile neutropenia. Of these, seventy-two (87%) reported that they would administer or prescribe paracetamol only as needed, and three (4%) would prescribe or administer paracetamol regularly. Eight (10%) reported that they would never use paracetamol in this setting.

Regarding the route of paracetamol use in patients with febrile neutropenia, all 85 respondents to this question reported that they would use oral paracetamol as the first choice. Forty-nine respondents selected an alternative route of administration in case oral paracetamol could not be given, of which 39 respondents favoured intravenous paracetamol and ten favoured rectal paracetamol. Sixty-four percent of respondents to this question (54/85) reported that they would avoid rectal paracetamol in patients with neutropenic fever.

Among the 83 respondents who selected at least one choice of physical cooling method, the most frequently selected options were removal of clothes (83% of respondents), provision of a fan (81%), tepid sponging (58%), and provision of a wet towel or flannel (53%). Fewer than ten percent of respondents selected the use of ice packs, cooling blankets or intravenous fluids for physical cooling.

Seventy-five respondents answered the question about time to change of antimicrobials. Of these, the majority stated that they would consider an antibiotic change at either 48 or 72 hours, with 39% (29 respondents) selecting each of these time points. A further 9% would change at 24 hours, 4% at 36 hours and 1% at 96 hours. Eight percent (6/75) indicated that they had no fixed time for antimicrobial change in this situation.

Sixty-nine percent of respondents (51/74) stated that they would be willing to enter their patients into a randomised study of antipyretic management in febrile neutropenia. A further 27% (20/74) were unsure. Three respondents to this question (4%) stated that they would not be willing to enter their patients in such a study. Asked whether there were specific groups of patients they would not be willing to enter into a randomised study, respondents nominated the following groups: elderly patients (3% of all respondents), children (3%), stem cell donors (5%), autologous stem cell transplant recipients (8%), allogeneic stem cell transplant recipients (18%).


This practice survey reports the attitudes of haematology doctors and nurses to antipyretic treatment for patients with neutropenic fever.

The survey indicates that overall, most respondents would advise a neutropenic patient to take paracetamol as needed for pain, but that nurses were significantly more likely than doctors to advise patients to avoid paracetamol. In febrile neutropenia, thresholds for temperature-lowering interventions varied widely, but nurses reported they would intervene at a significantly lower temperature than doctors in a patient without rigors.

In a symptomatic patient, both professional groups would intervene at a median of 38.0°C. Most clinicians would use either paracetamol or physical measures as a first-line intervention, with a narrow preference for paracetamol. Physical measures were the favoured second-line, and pethidine was the favoured third-line, cooling intervention.

Oral paracetamol was preferred over the intravenous route, and most respondents would avoid the rectal route. Finally, the majority of respondents believed a clinical study of antipyretic treatment in febrile neutropenia was warranted, and would consider entering their patients in such a study.

To the authors’ knowledge, this is the first survey of temperature management in febrile neutropenia. In collaboration with the New Zealand branch of the HSANZ, we were able to survey nearly half of all haematology doctors working in New Zealand. The majority of respondents who commenced the online survey completed it, with 95% and 84% response rates to the penultimate and final survey questions, respectively. This study employed scenarios to assess clinical practice, an approach that has been validated in a variety of settings.11,12

Although we were able to calculate a response rate for doctors, the survey response rate among nurses is unknown due to a lack of a central register of haematology nurses in New Zealand, and a large number of haematology patients being cared for in mixed-specialty wards. Nonetheless, we believe the response rate among nurses to be lower than that among doctors.

Among both doctors and nurses, the voluntary nature of survey participation may result in bias, so the current findings may not be representative of all clinicians. Finally, reported preferences in a survey may not reflect actual clinical practice. We intend to address some of these points by undertaking an observational study of paracetamol usage among inpatients with febrile neutropenia.

The authors are not aware of any published surveys of fever management in neutropenic fever, but clinicians’ attitudes to fever have been investigated in other infection scenarios. In a review of fever management among critical care clinicians in Australia and New Zealand,13doctors reported a significantly higher mean temperature intervention threshold than nurses (39.0°C vs 38.5°C), similarly to this study. The temperature intervention thresholds in the critical care study were higher than in the current study, possibly because many intensive care patients are sedated, so fever-related symptoms are less of a concern.

A study of fever management by paediatric junior doctors indicated a mean antipyretic treatment threshold of 38.6°C, with alternating aspirin and paracetamol as the favoured intervention.14 Very few respondents favoured non-steroidal anti-inflammatory drugs in the current study, possibly due to concerns about impairing platelet function in neutropenic patients, many of whom are also thrombocytopenic.

In a survey of fever management among Swiss paediatricians, a temperature threshold of 38.5°C was identified as a threshold for treatment, with the vast majority favouring paracetamol;15 in this study, the favoured routes of paracetamol administration were rectal for 18 month olds, and oral for older children. The widespread reluctance to use rectal paracetamol in the current study is likely to reflect concern about inducing bacteraemia in the neutropenic patient.

The role of antipyretics in the management of infection remains unclear. Observational studies have found that the absence of fever is associated with increased mortality in patients admitted to intensive care units with infection,16 and that the use of antipyretics is associated with increased mortality in septic, but not in non-septic patients.17

A number of interventional studies support the notion that fever is an important physiological response to infection: in children with falciparum malaria, the administration of regular paracetamol was associated with delayed resolution of parasitaemia,18 and in healthy volunteers infected with rhinovirus, regular paracetamol administration was associated with a reduced antibody response and a prolongation of symptoms.19

A single randomised trial comparing aggressive temperature control to permissive hyperthermia in patients with sepsis in intensive care found a trend towards reduced mortality in the permissive hyperthermia group.20 This finding is yet to be replicated.21

The present study assesses clinician preferences regarding antipyretic treatment in febrile neutropenia using a scenario-based survey. This survey demonstrates a lack of clear consensus on thresholds for, and methods of, lowering temperature, which is understandable given the lack of evidence. Establishing the role of antipyretics in neutropenic fever would require a prospective randomised controlled trial, for which the majority of respondents to this survey indicate support.


Infections are a common problem after chemotherapy, and can be life-threatening. Patients with infection after chemotherapy must be promptly treated with antibiotics. These patients often have high fevers. This paper is a survey of haematologists (blood specialists) in New Zealand, which asks how they manage fever in patients who have had chemotherapy. The reason for asking this question, is that some researchers believe that having a high fever helps to eradicate an infection, and that measures to reduce fever (such as taking paracetamol) might be detrimental. However, we also know that paracetamol makes patients more comfortable. We found that the way clinicians manage fever among their patients is very variable, but that most do give their patients treatments to reduce their temperature. Nurses have a preference for intervening at a lower temperature than doctors do. The variability in clinical practice we found reflects the fact that there is no good evidence (from clinical trials) to guide us. In our survey, most clinicians supported the idea of a randomised trial to show whether lowering temperature during post-chemotherapy infections is beneficial or not.



To assess the attitudes of clinicians to temperature management in haematology patients with febrile neutropenia.


An online scenario-based survey was circulated to consultant members of the New Zealand branch of the Haematology Society of Australia and New Zealand, to haematology advanced trainees, and to nursing representatives at each haematology department in New Zealand.


Eighty-eight responses were obtained, from 34 doctors and 54 nurses. Most respondents would advise a neutropenic patient to take paracetamol as needed for pain. Median temperature intervention threshold for an asymptomatic patient with febrile neutropenia was higher for doctors than for nurses (38.5 versus 38.0C), despite considerable heterogeneity. Both groups indicated they would intervene at a median 38.0C for a patient with rigors. Paracetamol was the preferred first-line cooling measure, with physical methods second-line, and pethidine third-line. All respondents favoured oral over intravenous or rectal paracetamol. Most believed a clinical trial of antipyretic treatment for febrile neutropenia was warranted, and indicated willingness to enrol their patients in such a study.


This survey documents clinicians’ preferred temperature intervention thresholds and methods for haematology patients with neutropenic fever, and shows considerable variation in practice. Most respondents supported a trial of antipyretic management in febrile neutropenia.

Author Information

Robert Weinkove, Consultant Haematologist, Wellington Blood & Cancer Centre, Wellington Hospital, Wellington—and Clinical Research Fellow, Vaccine Research Group, Malaghan Institute of Medical Research, Wellington; Jennifer Clay, Haematology Registrar, Wellington Blood & Cancer Centre, Wellington Hospital, Wellington; Catherine Wood, Research Nurse, Vaccine Research Group, Malaghan Institute of Medical Research, Wellington—and Clinical Nurse Specialist, Wellington Blood & Cancer Centre, Wellington Hospital, Wellington


The authors would like thank Professor Richard Beasley and Dr Paul Young for valuable suggestions and discussions, and Drs Bartrum Baker and Hugh Goodman of the New Zealand branch of the Haematology Society of Australia and New Zealand for assistance with distribution of this survey.


Dr Robert Weinkove, Wellington Blood & Cancer Centre, Wellington Hospital, Private Bag 7902, Wellington 6242, New Zealand Fax: +64 (0)4 3855984

Correspondence Email


Competing Interests



  1. Bodey GP, Buckley M, Sathe YS, Freireich EJ. Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Ann Intern Med. 1966;64:328-40.
  2. Tam CS, O'Reilly M, Andresen D, et al. Use of empiric antimicrobial therapy in neutropenic fever. Australian Consensus Guidelines 2011 Steering Committee. Intern Med J. 2011;41:90-101.
  3. Savoie ML, Nevil TJ, Song KW, et al. Shifting to outpatient management of acute myeloid leukemia: a prospective experience. Ann Oncol. 2006;17:763-8.
  4. Weissinger F, Auner HW, Bertz H, et al. Antimicrobial therapy of febrile complications after high-dose chemotherapy and autologous hematopoietic stem cell transplantation-guidelines of the Infectious Diseases Working Party (AGIHO) of the German Society of Hematology and Oncology (DGHO). Ann Hematol. 2012.
  5. Freifeld AG, Bow EJ, Sepkowitz KA, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the infectious diseases society of america. Clin Infect Dis. 2011;52:e56-93.
  6. Penack O, Buchheidt D, Christopeit M, et al. Management of sepsis in neutropenic patients: guidelines from the infectious diseases working party of the German Society of Hematology and Oncology. Ann Oncol. 2011;22:1019-29.
  7. Mackowiak PA. Direct effects of hyperthermia on pathogenic microorganisms: Teleologic implications with regard to fever. Rev Infect Dis. 1981;3:508-20.
  8. Mackowiak PA, Marling-Cason M, Cohen RL. Effects of temperature on antimicrobial susceptibility of bacteria. J Infect Dis. 1982;145:550-3.
  9. Pernerstorfer T, Schmid R, Bieglmayer C, et al. Acetaminophen has greater antipyretic efficacy than aspirin in endotoxemia: a randomized, double-blind, placebo-controlled trial. Clin Pharmacol Ther. 1999;66:51-7.
  10. Manthous CA, Hall JB, Olson D, et al. Effect of cooling on oxygen consumption in febrile critically ill patients. Am J Respir Crit Care Med. 1995;151:10-4.
  11. Peabody JW, Luck J, Glassman P, et al. Comparison of vignettes, standardized patients, and chart abstraction: a prospective validation study of 3 methods for measuring quality. JAMA. 2000;283:1715-22.
  12. Peabody JW, Luck J, Glassman P, et al. Measuring the quality of physician practice by using clinical vignettes: a prospective validation study. Ann Intern Med. 2004;141:771-80.
  13. Saxena MK, Hammond NE, Taylor C, et al. A survey of fever management for febrile intensive care patients without neurological injury. Crit Care Resusc. 2011;13:238-43.
  14. Weiss J, Herskowitz L. House officer management of the febrile child. A survey. Clin Pediatr (Phila). 1983;22:766-9.
  15. Lava SA, Simonetti GD, Ramelli GP, et al. Symptomatic management of fever by Swiss board-certified pediatricians: results from a cross-sectional, Web-based survey. Clin Ther. 2012;34:250-6.
  16. Young PJ, Saxena M, Beasley R, et al. Early peak temperature and mortality in critically ill patients with or without infection. Intensive Care Med. 2012.
  17. Lee BH, Inui D, Suh GY, et al. Association of body temperature and antipyretic treatments with mortality of critically ill patients with and without sepsis: multi-centered prospective observational study. Crit Care. 2012;16:R33.
  18. Brandts CH, Ndjave M, Graninger W, Kremsner PG. Effect of paracetamol on parasite clearance time in Plasmodium falciparum malaria. Lancet. 1997;350:704-9.
  19. Graham NM, Burrell CJ, Douglas RM, et al. Adverse effects of aspirin, acetaminophen, and ibuprofen on immune function, viral shedding, and clinical status in rhinovirus-infected volunteers. J Infect Dis. 1990;162:1277-82.
  20. Schulman CI, Namias N, Doherty J, et al. The effect of antipyretic therapy upon outcomes in critically ill patients: a randomized, prospective study. Surg Infect (Larchmt). 2005;6:369-75.
  21. Jefferies S, Weatherall M, Young P, et al. The effect of antipyretic medications on mortality in critically ill patients with infection: a systematic review and meta-analysis. Crit Care Resusc. 2011;13:125-31.