Access provided by
London Metropolitan University
• To increase your awareness of the risk of neutropenic sepsis in patients receiving anticancer treatment
• To acknowledge the importance of staff education on neutropenic sepsis and its timely management
• To recognise the importance of patient education regarding the prompt reporting of adverse effects
Neutropenic sepsis is a potentially life-threatening complication of systemic anticancer therapy and must be recognised and treated rapidly. National guidance recommends, among other things, that patients with suspected neutropenic sepsis receive intravenous antibiotics within one hour of presentation. Initial observations, assessment and investigations are key to inform management.
This article discusses the findings of a service evaluation conducted in a cancer unit in Northern Ireland to assess the initial management of patients with suspected neutropenic sepsis against the targets set by national guidance. The authors identify areas of optimal practice and areas for improvement, and make recommendations on how to enhance the management of neutropenic sepsis, notably through staff education.
Cancer Nursing Practice. doi: 10.7748/cnp.2023.e1835
Peer reviewThis article has been subject to external double-blind peer review and checked for plagiarism using automated software
Correspondence Conflict of interestNone declared
McCauley L, Donovan MM, McCaughey C et al (2023) Neutropenic sepsis: evaluating the timeliness of initial patient management on presentation to hospital. Cancer Nursing Practice. doi: 10.7748/cnp.2023.e1835
Published online: 06 July 2023
Patients receiving systemic anticancer therapy (SACT), particularly chemotherapy, are at risk of developing an oncological emergency known as neutropenic sepsis. Neutropenic sepsis is a common and potentially life-threatening complication of neutropenia (low neutrophil count) and is defined as a temperature of ≥38°C or any symptoms and/or signs of sepsis in a person with an absolute neutrophil count of ≥0.5x109/L. Potential complications include invasive and atypical infection, coagulopathy, encephalopathy, delirium, organ failure and death (National Institute for Health and Care Excellence (NICE) 2012, 2020).
The incidence of neutropenic sepsis in the UK is challenging to determine due to variations in definitions and coding. White and Ybarra (2017) noted that neutropenic sepsis is only documented in 20% to 30% of patients who develop it. Another term for neutropenic sepsis is febrile neutropenia, which is the presence of fever in a person with neutropenia and is the most common complication of anticancer treatment (NICE 2020). Ba et al (2020) cited a 23% to 26% incidence of febrile neutropenia during the first cycle of chemotherapy but also said the incidence of febrile neutropenia is likely much higher in the ‘real world’ than reported in trials.
According to the Northern Ireland Cancer Network (NICaN) (2022), early recognition of neutropenic sepsis is essential and all patients who present as an emergency to a healthcare setting within six weeks of having received SACT must be assumed to have neutropenic sepsis until proven otherwise. In patients with neutropenic sepsis, the urgent administration of intravenous (IV) fluids and antibiotics has demonstrated benefits in terms of clinical outcomes and survival (Clarke et al 2013). NICE (2012) and the UK Sepsis Trust (2022) recommend the administration of antibiotics within one hour of presentation with suspected neutropenic sepsis.
• Neutropenic sepsis is a common and potentially life-threatening complication of neutropenia
• Neutropenic sepsis is common in patients receiving systemic anticancer therapy, particularly chemotherapy
• It is essential that neutropenic sepsis is recognised and treated early
• National guidance recommends antibiotic administration within one hour of presentation with suspected sepsis
• Hospital staff likely to care for patients with cancer presenting with suspected neutropenic sepsis need education on neutropenic sepsis and its initial management
To meet the target of antibiotic administration within one hour, several components are required, including medical assessment, investigations and observations using a tool such as the National Early Warning Score (NEWS)2 (Royal College of Physicians 2017). The target of antibiotic administration within one hour is part of what has become known as the Sepsis 6, which should be delivered ‘as quickly as possible’ and, in the sickest patients, ‘always within the first hour following recognition of sepsis’ (UK Sepsis Trust 2022).
The UK Sepsis Trust (2022) lists the Sepsis 6 as follows:
• Ensure senior clinician attends.
• Give oxygen if required.
• Send bloods including cultures.
• Give IV antibiotics and consider source control.
• Give IV fluids.
• Monitor patients using the NEWS2.
However, according to Perron et al (2014), meeting the target of antibiotic administration within one hour of presentation – also referred to as the ‘door-to-needle’ time – has proved challenging, mainly due to prolonged assessments, a lack of awareness of neutropenic sepsis among staff and a lack of clear protocols.
To assess the immediate management (in the first hour of admission to a hospital in Northern Ireland) of patients with suspected neutropenic sepsis presenting within six weeks of receiving chemotherapy.
A prospective service evaluation was conducted over a 12-month period. Patient management in the first hour of admission to the cancer unit was assessed against the recommendations made by NICE (2012) in its guideline on the prevention and management of neutropenic sepsis in people with cancer and by NICaN (2022) in its acute oncology clinical guidelines. The ultimate aim was to identify areas of optimal practice and areas that required improvement.
The evaluation was conducted in all 37 patients admitted to the cancer unit between January and December 2018 with suspected neutropenic sepsis who had presented within six weeks of receiving chemotherapy. Among the 37 patients, 46% (n=17) had a central line already in place by the time they were admitted to the cancer unit. The source of data was patients’ healthcare records, including chemotherapy helpline proformas, medical admission notes, medicine prescriptions and administration records (Kardex), NEWS2 charts, and fluid balance charts.
The authors developed a password-protected electronic data collection tool. To enhance the validity and reliability of data, the tool had been developed in collaboration with stakeholders (Healthcare Quality Improvement Partnership 2021) including the hospital’s service development lead, the general manager for cancer services and the acute oncology team.
The first author (LM) and other members of the acute oncology team filled in the data collection tool as soon as possible after each care intervention during each patient’s stay in the cancer unit, thereby ensuring that data were recorded contemporaneously, as required by the 2018 Data Protection Act 2018. The collected data were transferred onto a spreadsheet, which was subsequently checked separately by LM and by the service development lead for accuracy and comprehensiveness.
Breast cancer was the most common type of cancer, accounting for 81% (n=30) of patients. The remaining cancer types were colorectal cancer (n=3, 8%), lung cancer (n=2, 5%), oesophageal cancer (n=1, 3%) and ovarian cancer (n=1, 3%). Table 1 shows a breakdown of the 37 patients according to cancer type and chemotherapy regimen.
The 37 patients had been admitted to the cancer unit after varying numbers of chemotherapy cycles: 27% (n=10) after their first cycle, 14% (n=5) after their second cycle, 19% (n=7) after their third cycle, 22% (n=8) after their fourth cycle and the remaining 19% (n=7) of patients after their sixth cycle or beyond.
Among the 37 patients, 27% (n=10) had been admitted to the cancer unit between day seven and day ten after chemotherapy. The largest group of patients admitted on a single day comprised 16% (n=6) admitted on day 14 after chemotherapy, while 19% (n=7) of patients had been admitted by day six after chemotherapy.
Figure 1 shows the number of days since the 37 patients had received chemotherapy before they were admitted to the cancer unit.
Before presenting to hospital, 92% (n=34) of patients had contacted the local chemotherapy helpline, one patient (3%) had contacted the Northern Ireland Cancer Centre, and two patients (5%) had not contacted either the local helpline or the Northern Ireland Cancer Centre. The data further showed that 35% (n=13) of patients had been admitted to the cancer unit via the emergency department (ED), 35% (n=13) via the medical assessment unit and 30% (n=11) via the chemotherapy treatment unit.
Regarding the time of presentation to hospital, 51% (n=19) of the 37 patients had presented out of hours and 41% (n=15) had presented during service hours. For the remaining 8% (n=3) of patients, the time of presentation to hospital had not been recorded. Over 46% (n=6) of patients who had been admitted to the cancer unit via the ED (n=13) had presented to hospital out of hours.
All 37 patients (100%) had been medically assessed on admission to the cancer unit, but the exact timing of the assessments could not be determined because times had been either unclearly documented or not documented at all. All patients except one (n=36, 97%) had NEWS2 scores recorded within one hour of admission. For the one outstanding patient (n=1, 3%), the NEWS2 chart did not specify the time of the NEWS2 assessment.
Baseline blood tests including a full blood count had been carried out within one hour of presentation to hospital in 95% (n=35) of patients. Baseline blood tests had been carried out in the remaining 5% (n=2) of patients but not within the one-hour time frame.
An additional blood sample had been taken within one hour of presentation to hospital in 35% (n=13) of patients to measure venous lactate levels. In the remaining 65% (n=24) of patients, venous lactate levels had not been measured within one hour of admission or had not been measured at all.
Central blood cultures had been obtained within one hour of presentation to hospital in 41% (n=15) of patients. Central blood cultures had been obtained in the remaining 59% (n=22) of patients but not within the one-hour time frame.
Peripheral blood cultures had been obtained within one hour of presentation to hospital in 41% (n=15) of patients. Peripheral blood cultures had been obtained but not within the one-hour time frame in 46% (n=17) of patients. In the remaining 14% (n=5) of patients there were no records of peripheral blood cultures. Among the 17 patients who had a central line in place, peripheral blood cultures had been obtained within one hour of admission in 41% (n=7) of these patients but had not been obtained within the one-hour time frame in 59% (n=10) of them.
The data demonstrated that the door-to-needle time target had been achieved in the cancer unit in 57% (n=21) of patients. It had been achieved in 100% (n=11) of patients admitted via the chemotherapy treatment unit, in 62% (n=8) of patients admitted via the medical assessment unit and in 38% (n=5) of patients admitted via the ED.
In 32% (n=12) of the 37 patients, the door-to-needle time target had not been achieved in the cancer unit, the main documented reason being that staff were waiting for blood test results from the laboratory to confirm the diagnosis of neutropenic sepsis. The length of time it took to administer IV antibiotics to those 12 patients ranged from one hour and three minutes to 14 hours and 20 minutes.
The remaining 11% (n=4) of patients were already receiving antibiotics for suspected sepsis on admission to the cancer unit.
All patients (n=37, 100%) had received an appropriate antibiotic regimen, as determined by checking the Kardex, albeit not all within the one-hour time frame. The data showed that 86% (n=32) had received piperacillin with tazobactam, which is indicated as the first-line antibiotic treatment for potential neutropenic sepsis (Vossen et al 2018). The remaining 14% (n=5) had received other antibiotic regimens to account for their allergy status (for example, allergy to penicillin) or the potential presence of line sepsis (where the infection has entered the bloodstream through a central line).
Breast cancer, colorectal cancer and lung cancer were the three most common cancers in this service evaluation, with breast cancer accounting for 81% of patients. These are among the most common cancers (Simmons 2012), so there will be higher numbers of patients with one of these cancers who present with neutropenic sepsis. However, it is important that nurses are aware of the risk of neutropenic sepsis in patients with less common cancers, such as oesophageal cancer and ovarian cancer.
The European Organisation for Research and Treatment of Cancer (EORTC) reviewed the occurrence of neutropenic sepsis in patients receiving chemotherapy regimens deemed to pose an intermediate or high risk of causing neutropenic sepsis (Aapro et al 2011). Patients who had received FEC100-D (fluorouracil, epirubicin hydrochloride, cyclophosphamide and docetaxel with the epirubicin component at 100mg/m2) were found to be at increased risk compared with patients who had received other chemotherapy regimens. This increased risk of neutropenic sepsis in patients receiving FEC-100D is reflected in this service evaluation, where 19% of patients had received FEC-D.
Culakova et al (2014) found that neutropenic sepsis was more likely to develop in the first cycle of chemotherapy and that the number of cases decreased in subsequent cycles. That decrease in subsequent cycles was associated with an increased use of granulocyte-colony stimulating factor (G-CSF) and antibiotics and with reductions in the dose of chemotherapy agents (Culakova et al 2014). G-CSF stimulates the production of neutrophils in bone marrow and can be prophylactically prescribed to patients receiving chemotherapy to reinforce their immune system and reduce the risk of infection (Baig et al 2019). While neutropenic sepsis is more likely to develop in the first cycle of chemotherapy (Culakova et al 2014), nurses need to be aware that it can develop in later cycles as well. In this service evaluation, only 27% of patients had presented to hospital after their first cycle of chemotherapy.
In this service evaluation, 27% of patients had been admitted to the cancer unit between seven and ten days after chemotherapy. Patients are most at risk of neutropenic sepsis between the seventh and tenth day after chemotherapy because, according to NICaN (2022), their neutrophil count is at its lowest point (neutrophil nadir) during that period. However, 19% of patients had been admitted by day six, which shows that there is a risk of neutropenic sepsis developing before the high-risk period of seven to ten days after chemotherapy.
The largest group of patients admitted on a single day were those on day 14 after chemotherapy. This seems to match the pattern described by Ford and Marshall (2014), who suggested that the typical neutrophil nadir occurs between day ten and day 14 after chemotherapy. However, this contradicts the timing of the neutrophil nadir according to NICaN (2022) and to NICE (2020), which state that the neutrophil nadir is typically between five and ten days after the last chemotherapy dose.
The local chemotherapy helpline contacted by 92% of patients in this service evaluation is staffed by nurses who assess and prioritise patients based on the UK Oncology Nursing Society (UKONS 2016) 24-hour triage toolkit. The helpline is supported by the Acute Oncology Service with the aim of streamlining patient care in the case of unplanned cancer-related admissions (Marshall et al 2013, UKONS 2016).
Patients receiving SACT are encouraged to contact the local chemotherapy helpline if they have any concerns, for example if they think they are experiencing adverse effects of treatment. Advising patients to contact their local helpline is an integral component of patient education (Jivraj et al 2018), the aim being to ensure that patients report adverse effects promptly (UKONS 2016). Whenever possible, family and informal carers should also receive education.
The proportion of patients who had contacted the local chemotherapy helpline before presenting to hospital (92%) was high, which could have been due to effective patient education by staff. However, nurses need to be aware that not all patients will report adverse effects promptly. Patients may underestimate the adverse effects of SACT and/or be slow at reporting them, which can result in treatment delays, an increased number of hospitalisations, reduced quality of life and death (McKenzie et al 2011).
Oakley (2023) found that patients’ anxiety about death and dying could be heightened when they were given information about neutropenic sepsis. Patients’ fears that their cancer treatment may have to be stopped if they develop neutropenic sepsis can lead them to avoid contacting healthcare professionals. Oakley (2023) listed reasons for patients delaying the reporting of symptoms of neutropenic sepsis such as denial, lack of motivation, advanced disease, fatalism and suboptimal relationships with healthcare staff. Other potential barriers to patients contacting the chemotherapy helpline include the anxiety associated with a cancer diagnosis (Jivraj et al 2018) and a lack of understanding of the urgency of reporting treatment toxicities (Borsellino and Young 2011).
Similar proportions of patients had been admitted via the chemotherapy treatment unit (30%), the medical assessment unit (35%) and the ED (35%), which reflects the processes in place to ensure round-the-clock service provision. However, the door-to-needle time target was only achieved in the cancer unit in 38% of patients admitted via the ED, compared with 62% of those admitted via the medical assessment unit and 100% of those admitted via the chemotherapy treatment unit.
Over half of patients admitted via the ED had presented to hospital out of hours, which according to Scullin and Hurwitz (2014) has implications in terms of staff availability and level of specialist competence. Kyriacou et al (2014) found that patients presenting to the ED with febrile neutropenia experienced significant delays in antibiotic treatment. Patients may present to the ED in a hospital that does not have a cancer unit or an acute oncology service, in which case they will be more likely to be under the care of staff who lack experience in oncological emergencies (Navani 2014). The acute oncology service at the trust where this service evaluation was conducted operates from Monday to Friday between 9am and 5pm and is open on bank holidays.
In this service evaluation, the door-to-needle time target was not achieved after admission to the cancer unit in 32% of patients, which matches the delays in antibiotic administration mentioned in the literature (Forde and Scullin 2017). The length of time it took to administer IV antibiotics to those 32% of patients ranged from one hour and three minutes to 14 hours and 20 minutes, emphasising a clear need for improvement.
In the case of medical emergencies such as neutropenic sepsis, it is essential to carefully and accurately record all care interventions, including their time, as well as patients’ clinical markers, since patient care may need to be escalated quickly. For example, it is vital to accurately record patients’ fluid balance, since a negative fluid balance is an indicator of organ dysfunction (Dellinger et al 2017).
All patients in the service evaluation had been medically assessed on admission to the cancer unit but the time of assessment had either been unclearly documented or not documented at all. In 14% of patients, there were no records of peripheral blood cultures. The fluid balance charts for all 37 patients were incomplete or illegible. These issues with recordkeeping indicate that there is an urgent need to educate staff regarding the importance of keeping accurate and comprehensive patient records.
In patients with neutropenic sepsis, blood tests are used to identify the presence of a blood infection (septicaemia) that can lead to septic shock. They enable to be determined the presence, in the blood, of bacteria or fungi that can significantly increase the mortality rate of patients with neutropenic sepsis (Bodey 2009). In this service evaluation, baseline blood tests had been carried out within one hour of presentation to hospital in 95% of patients.
A patient’s venous lactate level can support the diagnosis of neutropenic sepsis and is an important prognostic indicator (Kochanek et al 2019). A venous lactate level of ≥4mmol/L is associated with a 27% mortality rate, and mortality significantly decreases in parallel with the venous lactate level (Farkas 2015). Additionally, the venous lactate level can help to determine the severity of neutropenic sepsis. According to NICaN (2022), venous lactate levels should be measured immediately, based on the assumption that patients have developed neutropenic sepsis until proven otherwise. NICE (2012) recommends measuring venous lactate levels as part of the initial investigation of potential neutropenic sepsis. However, in this service evaluation, venous lactate levels had been measured within one hour of presentation to hospital in only 35% of patients.
Blood cultures support clinicians to identify the source of infection and guide their decision-making regarding antibiotic treatment. Obtaining blood cultures before the administration of IV antibiotics is essential to rule out bloodstream infections. When blood cultures are not obtained before the administration of antibiotics but only afterwards there is a risk of a false-negative result (Puerta-Alcalde et al 2019). However, in this service evaluation, peripheral blood cultures and central blood cultures had been obtained within one hour on presentation to hospital in only 41% of patients.
The service evaluation demonstrated substandard adherence to guidance in terms of measuring venous lactate levels and obtaining peripheral and central blood cultures. Continuous education of staff is required to ensure that they understand the rationale for measuring venous lactate levels and obtaining blood cultures before antibiotic administration. These elements should also be included in the routine auditing process in each ward or unit.
Based on the findings of this service evaluation, the authors recommend that cancer services:
• Develop an integrated care pathway specific to neutropenic sepsis and ensure it is implemented and regularly audited.
• Ensure guidelines for the management of neutropenic sepsis are readily accessible in all clinical areas.
• Develop a trust-wide standardised training programme for all front-line medical and nursing staff who may encounter patients receiving SACT and include it in the induction programme for junior doctors.
• Develop an action plan to address substandard practice, including strategies, such as regular reviews and audits, to support accurate and comprehensive documentation.
• Ensure that the acute oncology team provides leadership, including being available during service hours to offer guidance and training to all medical and nursing staff.
• Further explore this area of practice and disseminate outcomes regionally and nationally.
Furthermore, in line with NICE (2012), the authors recommend a prospective national cohort study to assess the incidence of suspected and confirmed neutropenic sepsis in patients receiving SACT. Such a study could also generate hypotheses concerning service delivery for neutropenic sepsis that could be tested regionally and nationally.
One limitation was that this service evaluation only explored the management of patients with suspected neutropenic sepsis in the first hour of admission in one cancer unit. It did not capture data for patients presenting with suspected neutropenic sepsis who had been discharged, which would have added to the richness of the data. The patient sample was small and did not include any patient treated in the haematology department. However, the service evaluation provided baseline data and a framework to make recommendations on how to enhance the management of neutropenic sepsis management at the trust. Furthermore, the data collection tool developed by the authors can be used by other regional services and NHS trusts.
SACT, particularly chemotherapy, comes with a high risk of neutropenic sepsis, a potentially life-threatening adverse effect of treatment. National guidance recommends, among other measures, that patients with suspected neutropenic sepsis receive antibiotics within on hour of presentation. In this evaluation of the initial management of 37 patients admitted to a cancer unit with suspected neutropenic sepsis, the one-hour target of antibiotic administration was only achieved in the cancer unit in 57% of patients.
The service evaluation also demonstrated substandard adherence to guidance in terms of measuring venous lactate levels and obtaining blood cultures, as well as suboptimal recordkeeping, notably regarding patients’ fluid balance. The service evaluation provided a framework to make recommendations on how to enhance the management of neutropenic sepsis, notably through staff education.
02 September 2024 / Vol 23 issue 5
Developing an e-learning package to provide chemotherapy updates
Cytotoxic chemotherapy is potentially carcinogenic,...
Exploring the role and expertise of ward-based oncology clinicians
Aim The aim of this study was to involve ward-based band 6...
Effectiveness of a neutropenic sepsis clinical pathway
This article reports the findings of a re-audit across a...
Benefits of attending nurse-led pre-chemotherapy group sessions
Before the start of chemotherapy, cancer patients and their...
Observational prospective cohort study on fluid infusion rate and cisplatin-associated complications
Aim To evaluate the influence of hydration infusion rate on...
