A multicentre, randomised controlled trial comparing the clinical effectiveness and cost-effectiveness of early nutritional support via the parenteral versus the enteral route in critically ill patients (CALORIES)

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 07/52/03. The contractual start date was in July 2010. The draft report began editorial review in July 2015 and was accepted for publication in November 2015. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Michael G Mythen reports grants from Smiths Medical Endowment and Deltrex Medical, personal fees from Edward Lifesciences and Fresenius-Kabi for speaking, consultation or travel expenses, and personal fees from AQIX (start-up company with novel crystalloid solution – pre-clinical), patent pending for the QUENCH pump and patent issued for Gastrotim outside the submitted work. Ella Segaran reports grants from Abbott Nutrition to attend a UK Intensive Care conference outside the submitted work. Danielle E Bear reports grants from the UK National Institute for Health Research Comprehensive Local Research Network, and personal fees/other support from Nestle Nutrition for speaker fees, Nutricia for speaker and consultancy fees plus payment of conference attendance, accommodation and travel expenses, Baxter for payment of course fees, travel and accommodation expenses and Corpak MedSystems UK for research support paid to Guy’s and St Thomas’ NHS Foundation Trust, outside the submitted work. Geoff Bellingan reports grants from the National Institute for Health Research for Selective Decontamination of the Digestive tract in critically ill patients treated in Intensive Care Unit (The SuDDICU study) (09/01/13) during the trial.

Copyright statement

© Queen’s Printer and Controller of HMSO 2016. This work was produced by Harvey et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Background and rationale

Malnutrition is a common problem in critically ill patients in UK NHS critical care units. 1 The consequences of malnutrition include vulnerability to complications, such as infection, which can lead to delays in recovery. Early nutritional support is therefore recommended for critically ill patients to address both deficiencies in nutritional state and related disorders in metabolism.

However, evidence is conflicting regarding the optimum route (parenteral or enteral) of delivery. 2–4 Meta-analyses of the trials comparing nutritional support via the enteral and parenteral route in critically ill patients have been published, but interpretation of their results is complicated by small sample size, poor methodological quality, select groups of critically ill patients studied, lack of standardised definitions for outcome measures and interventions combining more than one element of nutritional support, for example timing and route.

In 2003, Heyland et al. 2 reported no difference in mortality between patients given parenteral and enteral nutritional support, but enteral was associated with a significant reduction in infections. Safety, cost and feasibility led them to recommend enteral over parenteral in the critically ill adult patient. In 2004, Gramlich et al. 3 also found no difference in mortality but a significant reduction in infections with enteral nutrition (EN). 3 In addition, they reported no difference in length of unit stay or days on ventilation, but indicated that there were insufficient data to analyse these statistically. Using a different methodological approach to assessing quality of included studies (one less biased towards including the poorer-quality studies), Simpson and Doig,4 in 2005, found a significant reduction in mortality but a significant increase in infections with parenteral nutritional support compared with the enteral nutritional support. However, the significant mortality benefit with parenteral nutrition (PN) appeared to exist when compared with the provision of delayed, rather than early enteral nutritional support and thus this was not a like-for-like comparison. Similar time-based analyses for infections were not possible as a result of insufficient data.

All of the meta-analyses highlighted the problems of combining data from poor-quality studies conducted on heterogeneous patient populations (all were on select subgroups, such as head trauma, acute pancreatitis, etc.) plus variation in the timing of measurement of mortality and, perhaps more importantly, the nature and definitions for infections included and pooled (pneumonia, urinary tract, bacteraemia, wound, line sepsis, etc.). Owing to incomplete reporting, it was not possible to classify and combine infections based on risk of outcome (e.g. severe infection, moderate infection, subclinical infection).

The enteral route is the mainstay of nutritional support in critical care2,5,6 but it is frequently associated with gastrointestinal intolerance and underfeeding. 7,8 In contrast, the parenteral route though more invasive and expensive is more likely to secure delivery of the intended nutrition. 7 Historically, nutritional support via the parenteral route has been associated with more risks and complications (e.g. infectious complications) compared with the enteral route2–4 but recent improvements in the delivery, formulation and monitoring of PN justify further comparison and evaluation of these routes of nutritional support, particularly in the early phase of the illness. 9,10 Economic evidence surrounding the optimum route of delivery of nutritional support is largely based on evidence of effectiveness of questionable methodological quality and narrow focus on upfront acquisition costs, and full economic evaluation is lacking. 11

In view of this, in late 2007, the National Institute for Health Research (NIHR) Health Technology Assessment programme put out a call for a large pragmatic randomised controlled trial to be conducted to determine the optimal route of delivery of early nutritional support in critically ill adults. In response to this call, in 2008, we updated the most recent systematic review by Simpson and Doig (see Figure 1). Highly sensitive search criteria identified a further 570 potentially relevant studies since May 2003. Following detailed review of these studies, one additional randomised controlled trial comparing PN and EN was identified. 12 This paper reported the full results of a trial previously identified by Simpson and Doig as having only interim results reported,13 but was excluded from their meta-analysis, as the enteral nutritional support arm included immune-enhancing supplements. As the use of immune-enhancing supplements was to be permitted in our study, we repeated the meta-analysis to include studies with supplementation of either arm. This resulted in the inclusion of this trial and one additional randomised controlled trial excluded from the Simpson and Doig meta-analysis on this criterion. 14 The results of the updated meta-analysis, including a total of 13 studies,12,14–25 indicated a non-significant survival benefit for parenteral support [relative risk 0.82, 95% confidence interval (CI) 0.60 to 1.11] but an increased risk of infection (relative risk 1.77, 95% CI 1.19 to 2.63) compared with enteral nutritional support (Figure 1). Consequently, parenteral nutritional support in the critical care unit remained controversial and no clear evidence existed as to the optimum route for delivery of early nutritional support to critically ill patients.

FIGURE 1.

Updated meta-analysis of randomised trials comparing PN with EN. (a) Relative risk of mortality; and (b) relative risk of infectious complications. RR, risk ratio. Based on the criteria of Simpson and Doig,4 updated to 24 January 2008, and with exclusion criteria relaxed to include trials that had been excluded previously because of the use of immune-enhancing supplements.

Aim

The aim of the CALORIES trial was to compare the clinical effectiveness and cost-effectiveness of early nutritional support in critically ill patients, delivered via the parenteral compared with the enteral route.

Objectives

Trial design

The CALORIES trial was a pragmatic, open, multicentre, parallel-group, randomised controlled trial with an integrated economic evaluation.

The trial was nested in the Case Mix Programme, the national clinical audit of adult general critical care units in England, Wales and Northern Ireland, established in 1995 and co-ordinated by Intensive Care National Audit & Research Centre (ICNARC) (Scotland has its own separate national clinical audit). 26 The Case Mix Programme is listed in the Department of Health’s ‘Quality Accounts’ for 2013–14 as a recognised national audit by the National Advisory Group for Clinical Audit and Enquiries.

Nesting the CALORIES trial in the Case Mix Programme ensured an efficient design (with respect to participating units and data collection) and facilitated efficient management of the study, including monitoring recruitment.

Research governance

The trial was sponsored by ICNARC and co-ordinated by the ICNARC Clinical Trials Unit (CTU).

An ethics application was submitted to the North West London Research Ethics Committee on 28 October 2010 and the CALORIES trial received a favourable opinion on 16 December 2010 (reference number: 10/H0722/78). The protocol is available via www.nets.nihr.ac.uk/.

To ensure transparency, the trial was registered with the International Standard Randomised Controlled Trial Number (ISRCTN) Registry on 25 March 2009. Registration was confirmed on 9 April 2009 (ISRCTN17386141).

The NIHR Clinical Research Network (CRN) Portfolio details high-quality clinical research studies that are eligible for support from the NIHR CRN in England. The trial was adopted on to the NIHR CRN Portfolio on 24 March 2011 and was issued the NIHR CRN Portfolio number 10098.

Global NHS permissions were obtained from Central and East London Comprehensive Local Research Network (CLRN) on 10 March 2011 and local NHS permissions were obtained from each participating NHS trust. A clinical trial site agreement, based on the model agreement for non-commercial research in the health service, was signed by each participating NHS trust and the sponsor (ICNARC).

Following guidelines from the NIHR, a Trial Steering Committee (TSC), with a majority of independent members, was convened to oversee the trial on behalf of the funder (NIHR) and the sponsor (ICNARC). The TSC met at least annually during the trial and comprised an independent chairperson; independent lay members (representing patient perspectives); independent clinicians (specialising in critical care medicine); the chief investigator (KR); and the lead clinical investigator (MM) representing the Trial Management Group (TMG).

Additionally, an independent Data Monitoring and Ethics Committee (DMEC) was convened to monitor trial data and ensure the safety of trial participants. The DMEC met at least annually during the trial; it comprised two expert clinicians specialising in critical care medicine and was chaired by an experienced statistician.

Management of the trial

The trial manager was responsible for day-to-day management of the trial with support from the data manager, trial statistician and research assistant. The TMG, chaired by the ICNARC CTU manager (SH), was responsible for overseeing day-to-day management of the trial and comprised the chief investigator (KR), trial dietitians (DB, ES) and co-investigators (GB, RB, DH, RL and MM). The TMG met regularly throughout the trial to ensure adherence to the trial protocol and monitor the conduct and progress of the trial.

Network support

To maintain the profile of the trial, regular updates on trial progress were provided at quarterly meetings of the NIHR CRN Critical Care Specialty Group and at local CLRN meetings. In addition, updates were provided at national meetings, such as the Annual Meeting of the Case Mix Programme and the UK Critical Care Research Forum.

Design and development of the protocol

Clinicians – including doctors, nurses and dietitians, from NHS critical care units across the UK – were invited to a meeting in May 2010 to discuss the trial protocol. The purpose of the meeting was to provide a forum for clinicians who had expressed an interest in taking part in the trial to discuss the trial protocol in detail with the trial investigators. Following the meeting, minor changes were made to the trial protocol to ensure clarity.

Amendments to the trial protocol

Following receipt of a favourable opinion of the trial protocol from the Research Ethics Committee on 16 December 2010, four substantial amendments were submitted and received favourable opinion. In summary, these were:

• Amendment 1 (13 May 2011) – the personal/professional consultee consent form was replaced with a personal/professional consultee agreement form to clarify that consultees were being asked for agreement (not consent) for patients to participate in the trial according to the Mental Capacity Act (2005)27 and the National Research Ethics Service Guidance for Researchers & Reviewers. The patient information sheets (prospective and retrospective for the patient and for the personal/professional consultee) and consent forms (patient consent form and retrospective patient consent form) had minor administrative changes. A personal/professional consultee telephone agreement form was produced to ensure that, in the situation that a personal/professional consultee was contacted via the telephone for their opinion, this contact was documented. The trial protocol was amended to clarify the aim of the trial – to compare early nutritional support delivered via the parenteral with early nutritional support delivered via the enteral route – and to incorporate the most up-to-date version of the EuroQoL 5-dimension (5-level version) questionnaire (EQ-5D-5L) to evaluate health-related quality of life at 90 days and at 1 year.

• Amendment 2 (19 December 2011) – the letter to the patient’s general practitioner (GP) informing them of the patient’s participation in the trial was amended for use in cases when the patient was known to have died. The patient follow-up letters were amended to be specific to the follow-up time point, that is 90 days and 1 year post-randomisation, and minor semantic changes were made to the patient information sheets and consent/agreement forms.

• Amendment 3 (4 October 2012) – following requests from research teams at sites, a CALORIES trial information leaflet for family and friends was produced. The leaflet was placed in the critical care unit relatives’ room, with the aim of providing relatives/friends of patients in critical care with a brief overview of the trial. The exclusion criterion ‘known to be participating in an interventional study’ was removed following review by the TMG; it was agreed that patients could be co-enrolled into two interventional studies if, after careful consideration, there were no concerns about patient safety, risk of biological interaction or the scientific integrity of the trial. Local principal investigators (PIs) were advised to contact the ICNARC CTU on a case-by-case basis to discuss co-enrolment of patients. In addition, minor semantic changes were made to the trial protocol and consent/consultee agreement forms.

• Amendment 4 (23 October 2013) – a patient newsletter was added to the follow-up questionnaire pack that was sent to each patient at 90 days and 1 year post recruitment into the CALORIES trial.

NHS support costs

Trials in critical care are challenging and expensive to conduct. Unlike other areas of health care, such as oncology, recruitment cannot take place solely within usual office hours. Resources are needed to enable screening and recruitment 24 hours per day, 7 days per week. Patients with a critical illness can be admitted to the critical care unit at any time of day or night, including weekends. Another challenge of critical care research is the informed consent process, which often has to be completed within a short time frame, as treatments are often time limited. Furthermore, critically ill patients usually lack the mental capacity to be able to provide informed consent prior to randomisation, in which case it is necessary to involve a personal or professional consultee in accordance with the Mental Capacity Act. 27 Senior, experienced staff are needed to be able to assess the patient’s mental capacity and to be able to effectively communicate information about the trial to the patient and/or their relatives in a stressful situation.

To this end, resources equivalent to 0.5 whole-time equivalent (WTE) band 7 research nurse, 0.1 WTE critical care consultant, 0.1 WTE band 7 dietitian and 1.4 hours per week of a band 6 pharmacist were successfully agreed with the lead CLRN on 21 June 2011.

Resources were based on an estimated 175 eligible admissions per unit per year, of whom approximately 60 would be recruited and 30 randomised to receive early nutritional support via the parenteral route. Using these recommendations, participating sites, assisted by the TMG, negotiated resources required locally for the trial with their respective research and development departments and CLRNs.

Patient and public involvement

Engagement with patients was vital to the successful conduct of the trial. The original study proposal was reviewed and endorsed by Patients on Intravenous and Nasogastric Nutritional Therapy support group. Two former critical care patients were independent members of the TSC and they provided input into the conduct of the trial, including reviewing literature to be given to patients and their families (e.g. patient information sheets and patient newsletters).

Participants: sites

The trial aimed to recruit a representative sample of at least 20 adult, general critical care units from the UK. Adult, general critical care units were defined as intensive care units (ICUs) or combined intensive care/high-dependency units. Stand-alone high-dependency units and specialist critical care units (e.g. neurosciences, cardiothoracic, etc.) were not eligible for participation in the trial. The criteria for inclusion were:

• active participation in the Case Mix Programme – defined as submission of data no later than 6 weeks after the end of each quarter and returning corrected data validation reports no later than 6 weeks after receipt

• pre-existing, established protocols for PN and EN reflecting mainstream practice (reviewed and approved by the TMG)

• pre-existing implementation of bundles as promoted by the NHS (NHS Saving Lives: reducing infection, delivering clean and safe care – ‘High Impact Intervention No. 1: Central venous catheter’ and ‘High Impact Intervention No. 5: Ventilator’) to prevent the development of bloodstream infection and ventilator-associated pneumonia28,29

• pre-existing prophylaxis protocol for the prevention of venous thromboembolism

• glycaemic control protocol in line with international guidelines30

• agreement to incorporate the CALORIES trial into routine unit practice, including prior agreement – from all consultants in the unit – to adhere to the patient’s randomly allocated route (parenteral or enteral route) for delivery of early nutritional support

• agreement to recruit all eligible patients into the CALORIES trial and to maintain a screening log of eligible patients who were not randomised, and patients who fulfilled the inclusion criteria but met one or more of the exclusion criteria

• sign up from the unit clinical director, senior nurse manager, dietitian/clinical nutritionist and pharmacist, and

• identification of a dedicated CALORIES trial research nurse.

All of the units actively participating in the Case Mix Programme were invited for expressions of interest to take part in the trial. In addition, the trial was promoted through presentations at relevant national meetings of professional organisations, such as the Intensive Care Society and at the UK Critical Care Research Forum.

A PI, who was responsible for the conduct of the trial locally, was identified at each participating site.

Participants: patients

The trial procedures for recruitment and follow-up of patients are summarised in Figure 2.

FIGURE 2.

Summary of trial procedures for recruitment and follow-up of patients.

Screening and recruitment

Following attendance at a site initiation meeting, screening and recruitment was commenced at participating units once the clinical trial site agreement had been signed and all of the necessary approvals were in place.

To promote awareness of the trial and facilitate recruitment, posters providing information about the CALORIES trial were displayed in the critical care unit and in family/visitor waiting rooms.

Potentially eligible patients were identified and approached by authorised members of staff about taking part in the trial. Information about the trial was provided to the patient, which included the purpose of the trial, the consequences of taking part or not, data security and funding of the trial. This information was also provided in a patient information sheet (see Appendix 1), along with the name and contact details of the local PI, which was given to the patient to read before making their decision to take part, or not, in the trial.

If the patient lacked mental capacity (because of their acute illness) to understand the information about the trial then, in accordance with the Mental Capacity Act,27 a personal consultee, who could be a relative or close friend, was identified with whom to discuss the patient’s participation in the trial. If there was no personal consultee available then the patient was provided with a professional consultee, for example an independent mental capacity advocate appointed by the NHS hospital trust, with whom to discuss the patient’s participation in the trial. The personal/professional consultee was provided with the same information as for patients (see Appendix 1) along with an explanation that they were being asked for their agreement for the patient to take part in the trial. Patients and personal/professional consultees were provided with an opportunity to ask questions before being invited to sign the consent form or personal/professional consultee agreement form, as appropriate.

Informed consent

Staff members, who had received training on the background, rationale and purpose of the CALORIES trial, and on the principles of the International Conference on Harmonisation Good Clinical Practice guidelines, were authorised by the PI to take informed consent from patients or informed agreement from a personal/professional consultee.

Once the staff member who was taking informed patient consent or consultee agreement was satisfied that the patient or personal/professional consultee had read and understood the patient information sheet, and that all of his/her questions about the trial had been answered, the patient or personal/professional consultee was invited to sign the patient consent form or personal/professional consultee agreement form, as appropriate.

For patients who had lacked mental capacity prior to randomisation, informed consent to continue participating in the trial was sought as soon as possible after the patient had regained mental capacity. If a patient did not regain mental capacity then, if possible, for patients entered via professional consultee agreement, agreement from a personal consultee was obtained for the patient to continue participating in the trial.

Randomisation and allocation procedure

Following informed consent from the patient or informed agreement from a personal/professional consultee, eligible patients were randomised via a central 24 hours per day, 7 days per week, telephone randomisation service hosted by Sealed Envelope Ltd. Patients were allocated, 1 : 1, to early nutritional support via either the parenteral route or the enteral route, by minimisation with a random component (each patient being allocated with 80% probability to the treatment group that would minimise imbalance). Minimisation was based on the following factors: site; age (< 65 years or ≥ 65 years); surgical status (surgery within 24 hours prior to unit admission or not); and malnutrition status (based on clinical judgement) (yes or no). A manual randomisation list (using permuted blocks, with block lengths of 4, 6 and 8) was prepared in advance of the trial by the trial statistician for use if the central telephone randomisation service was not available for any reason. Staff at participating sites were advised to call the 24 hours per day, 7 days per week, telephone support service if they experienced any problems with the central telephone randomisation service. Manual randomisation was carried out, as required, by the on-call member of the TMG. Details of any patients manually randomised were passed to the randomisation service for inclusion in the minimisation algorithm for subsequent allocations.

Screening log

To enable full and transparent reporting for the trial, brief details of all patients who met eligibility criteria or who met all of the inclusion criteria plus one or more of the exclusion criteria were recorded in the screening log. The reasons for eligible patients not being recruited were recorded, which included the patient declining the invitation to take part, the patient being excluded by the treating clinician, logistical reasons, etc. No patient identifiers were recorded in the screening log.

Treatment groups

As a pragmatic trial, the CALORIES trial did not dictate the use of specific nutritional products or protocols for delivery of nutritional support via the parenteral and enteral routes. However, existing established protocols for delivery of PN and EN at participating units were reviewed and approved by the TMG to ensure that they fell within common boundaries.

Early nutritional support was delivered via either the parenteral or enteral route for the 5-day (120 hours) intervention period, unless the patient transitioned to exclusive oral feeding or was discharged from the critical care unit before this time. Patients were able to start oral feeding if clinically indicated during the 5 days.

Outcome measures

The primary clinical effectiveness outcome was all-cause mortality at 30 days following randomisation and the primary cost-effectiveness outcome was the incremental net benefit (INB) gained at 1 year following randomisation, at a willingness-to-pay of £20,000 per quality-adjusted life-year (QALY).

The secondary outcomes were:

• number of days alive and free from advanced respiratory support, advanced cardiovascular support, renal support, neurological support and gastrointestinal support up to 30 days following randomisation

• number of new, treated, confirmed or strongly suspected infectious complications, classified according to clinical diagnosis, which occurred in the critical care unit

• non-infectious complications of episodes of hypoglycaemia, elevated levels of liver enzymes, nausea requiring treatment, abdominal distension and vomiting, collected through adverse event reporting up to 30 days following randomisation, and new or significantly worsened pressure ulcers while in the critical care unit

• duration of critical care unit stay (from dates and times) and acute hospital length of stay (in whole days) following randomisation

• duration of survival at 90 days and at 1 year following randomisation

• all-cause mortality at discharge from the critical care unit and from the acute hospital

• all-cause mortality at 90 days and at 1 year following randomisation

• nutritional and health-related quality of life at 90 days and at 1 year following randomisation

• resource use and costs at 90 days and at 1 year following randomisation, and

• estimated lifetime incremental cost-effectiveness (INB).

Safety monitoring

Patients were monitored for adverse events occurring between randomisation and 30 days following randomisation. Specified adverse events were defined as follows:

• abdominal distension was defined as any new, clinically significant change in appearance; abdominal distension was considered severe if there was an acute obstruction

• abdominal pain was defined as any new episode of abdominal pain, localised to the abdomen and requiring more than just simple analgesia; abdominal pain was considered severe if it was not controlled with opiates

• episodes of electrolyte disturbance were defined as any new, clinically significant electrolyte disturbance requiring active monitoring or treatment

• haemopneumothorax was defined as any new haemopneumothorax requiring insertion of a chest drain

• hepatomegaly was defined as any new or increased hepatomegaly on clinical examination

• hyperosmolar syndrome was defined as any new, clinically significant osmolar gap requiring active monitoring or treatment

• hypersensitivity reaction (anaphylactic reaction) was defined as any new anaphylactic reaction

• episodes of hypoglycaemia were defined as any new episode of clinically significant hypoglycaemia requiring active monitoring or treatment

• ischaemic bowel was defined as any new episode of ischaemic bowel inferred on radiology or diagnosed visually, for example during surgery or by endoscopy

• jaundice was defined as any new, clinically significant jaundice requiring active monitoring or treatment

• nausea requiring treatment was defined as any new episode of nausea requiring treatment with anti-emetic drugs

• pneumothorax was defined as any new pneumothorax requiring insertion of a chest drain

• raised levels of liver enzymes were defined as any new, clinically significant rise in liver enzyme levels requiring active monitoring or treatment

• regurgitation/aspiration was defined as any episode of regurgitation/aspiration

• vascular catheter-related infection was defined as any new vascular catheter-related infection for which a vascular catheter, such as a central venous catheter, was identified as the primary source of infection and associated with signs and symptoms of infection requiring antimicrobial drugs and/or removal of catheter, and

• vomiting was identified as any episode of vomiting.

Unspecified adverse events were defined as an unfavourable symptom or disease that was temporally associated with the use of the trial treatments, whether or not it was related to the trial treatment, which was not deemed to be a direct result of the patient’s medical condition and/or standard critical care treatment.

All adverse events were recorded in the electronic case report form and reported, as part of routine reporting throughout the trial, to the DMEC and research ethics committee. Adverse events that were assessed to be serious (i.e. prolonging hospitalisation or resulting in persistent or significant disability/incapacity), life-threatening or fatal – collectively termed serious adverse events – were reported to the ICNARC CTU and reviewed by a clinical member of the TMG. Serious adverse events that were unspecified and considered to be possibly, probably or definitely related to the trial treatment were reported to the research ethics committee within 15 calendar days of the event being reported.

Data collection

A secure, dedicated electronic case report form, hosted by ICNARC, was set up to enable trial data to be entered by staff at participating sites. The electronic case report form was accessible only to authorised users and access was approved centrally by the trial manager, data manager or research assistant (after cross-checking the site delegation of trial duties log). Each individual was provided with a unique username and password and had access to data only for patients recruited at their site.

The data set for the CALORIES trial included the minimum data required to confirm patient eligibility, to describe the patient population, to monitor and describe delivery of the intervention, to assess primary and secondary outcomes and to enable linkage to the ICNARC Case Mix Programme (see Appendix 2). 26

Longer-term follow-up

Following randomisation, a letter was sent to the patient’s GP informing them of the patient’s participation in the trial and a request for assistance with follow-up, if required. All patients who survived to leave hospital were followed up at 90 days for the secondary outcomes (all-cause mortality, duration of survival, health-related quality of life and resource use), and at 1 year for secondary outcomes (all-cause mortality, duration of survival, health-related quality of life and resource use) and to calculate the primary cost-effectiveness outcome (INB).

Data linkage with death registration

Follow-up of patients was carefully monitored to prevent any potential distress to those who care for the patient receiving a letter addressed to a deceased relative, partner or friend. The follow-up process started at 75 days for the 90-day follow-up and at 350 days for the 1-year follow-up to allow for the administrative processes. Each week a list of all patients who had been discharged alive from hospital and who were either 75 days or 350 days post-randomisation was sent to the HSCIC Data Linkage and Extract Service to confirm their mortality status. Patients indicated as having died were logged and the follow-up process ended.

Follow-up procedure

Patients identified by the HSCIC Data Linkage and Extract Service as not having died started the follow-up process summarised in Figure 3. A questionnaire pack was sent from the ICNARC CTU, by post, to the patient. Following evidence-based practice for maximising responses to postal surveys,37 the questionnaire pack included a cover letter (see Appendix 5); the patient information sheet (see Appendix 1) or patient newsletter (which replaced the patient information sheet in November 2013); two questionnaires – the Health Questionnaire and the Health Services Questionnaire (see Appendix 6); a stamped-addressed return envelope; and a pen. The Health Questionnaire (see Appendix 6) included the required questions from the EQ-5D-5L to evaluate health-related quality of life38 and the Satisfaction with Food-related Life Questionnaire to evaluate the patient’s nutritional quality of life. 39 It is a measure of satisfaction developed by Grunert and the Food in Later Life team. The five items exhibit good reliability (as measured by Cronbach’s alpha), good temporal stability, convergent validity with two related measures, and construct validity as indicated by relationships with other indicators of quality of life. 39 The Health Services Questionnaire (see Appendix 6) included questions about the patient’s use of health services following discharge from the acute hospital and was used to cost subsequent use of health services. The cover of the questionnaires included a ‘do not wish to participate’ tick box.

FIGURE 3.

Patient follow-up process at 90 days and at 1 year.

If no response was received after 2 weeks, a reminder letter was sent with another questionnaire pack. If no response was received after a further 2 weeks, the patient was telephoned, if contact details were available. Telephone calls were made at various times from Monday to Friday between 08.30 and 20.30 to maximise the chances of contacting the patient. Patients who were successfully contacted by telephone were asked if they had received the questionnaire pack and were invited to complete the questionnaires over the telephone, if it was convenient. In addition, patients were reminded about completing the questionnaire when they attended hospital follow-up appointments.

Follow-up ended on receipt of a completed (or blank) questionnaire; a questionnaire with a ticked ‘do not wish to participate’ box; notification to the ICNARC CTU by telephone or e-mail that the patient wished to withdraw from the trial; or if there was no response to telephone follow-up. For questionnaire packs returned indicating that the recipient was not known at the address, the contact details for the patient were checked with the recruiting hospital and/or GP.

For patients who were identified as either being a hospital inpatient, or resident in a care home or rehabilitation centre, the relevant institution was contacted to establish the status of the patient and the most appropriate way to proceed with follow-up. If the patient had mental capacity to consent but required assistance in reading and/or completing the questionnaire then health-care professionals usually assisted the patient. For patients who lacked mental capacity to consent, institutions advised on the most appropriate person to contact to complete the questionnaires.

If patients were identified as having no fixed abode but were registered with a GP or had regular contact with a homeless shelter then the questionnaire pack was sent to be passed (when appropriate) to them at their next appointment or visit.

Data linkage with the Case Mix Programme

Linkage of patient identifiable trial data to the Case Mix Programme Database provided information on the baseline characteristics of patients and subsequent admission to the critical care unit following discharge from the critical care unit.

Data for the Case Mix Programme are collected by trained data collectors to precise rules and definitions. The data then undergo extensive local and central validation for completeness, illogicalities and inconsistencies prior to pooling.

Data management

Data management was an ongoing process. Data entered by sites on to the electronic case report form were monitored and checked throughout the recruitment period to ensure that data were as complete and accurate as possible.

Two levels of data validation were incorporated into the electronic case report. The first was to prevent obviously erroneous data from being entered, for example entering a date of birth that occurred after the date of randomisation. The second level involved checks for data completeness and any unusual data entered, for example a physiological variable, such as blood pressure, which was outside of the predefined range. Site staff could generate data validation reports, listing all outstanding data queries, at any time via the electronic case report form. The site PI was responsible for ensuring that all of the data queries were resolved. Ongoing data entry and validation at sites was closely monitored by the data manager (JT) and any concerns were raised with the site PI.

The contact details for patients and their GPs (name and postal address) were checked weekly for completeness to avoid unnecessary delays in sending out questionnaire packs at 90 days and at 1 year.

Adherence to the trial protocol was closely monitored, including adherence to all elements of early nutritional support delivered via the parenteral and the enteral route. Any queries relating to adherence were generated in a separate monthly report, which was sent to the site PI. For each query, the PI was asked to explain the reason for any non-adherence to the protocol. If deemed necessary, a teleconference was arranged with the site to ensure that effective plans were put in place to improve future adherence.

Data received from completed Health and Health Services questionnaires were entered centrally into a secure database at the ICNARC CTU following a standard operating procedure. All identifiable information, such as names (e.g. of patients, family members or hospital staff members) were removed. All queries relating to data entry were reviewed by two members of the TMG (SH/BM) and any disagreement was reviewed and discussed with a third (KR).

To ensure that data were entered accurately, all questionnaire data entered into the database were cross-checked by a second member of the CTU team. Any errors that were found were logged and corrected on the database.

Sample size

Applying the trial entry criteria to over 500,000 admissions to adult, general critical care units in the Case Mix Programme Database, the 30-day mortality for unplanned, ventilated adult admissions staying ≥ 3 days was 32%. As the enteral route is the predominant choice for nutritional support, this mortality was used as the basis to estimate control group mortality.

A meta-analysis of existing randomised controlled trials of PN compared with EN indicated a potential relative risk reduction associated with PN of around 20% (see Figure 1). To have 90% power, with a type I error rate of 5% (two sided), to detect a 20% relative risk reduction (6.4% absolute risk reduction) from 32% in the enteral route group to 25.6% in the parenteral route group, requires a sample size of 1082 per treatment group (Stata/SE version 10.1, StataCorp LP, College Station, TX, USA). To allow 2% for crossover/protocol violation (in each direction) and 2% for loss to follow-up/withdrawal prior to 30 days (based on observed rates from the PAC-Man Study40), a sample size of 1200 per treatment group (2400 total) was required. No adjustment to the sample size calculation was made to account for subgroup analyses.

Interim analysis

Unblinded, comparative data on recruitment, withdrawal, adherence (to the allocated treatment) and serious adverse events were regularly reviewed by an independent DMEC.

Without specific analysis of the primary outcome, the DMEC reviewed data from the first 37 trial participants and continued to review data at least 6-monthly to assess potential safety issues and to review adherence. A single, planned, formal, interim analysis was performed at the point that 30-day outcome data for the first 1200 patients enrolled were available. A Haybittle–Peto stopping rule (p < 0.001) was used to guide recommendations for early termination as a result of harm. Following the planned interim analysis, the DMEC recommended that the trial continue with no changes.

Analysis principles

All analyses were based on the intention-to-treat principle. Patients were analysed according to the treatment group to which they were randomised, irrespective of whether or not the allocated treatment was received (i.e. regardless of whether they did or did not adhere to the allocated route). All tests were two-sided, with significance levels set at p < 0.05 and with no adjustment for multiplicity. All a priori subgroup analyses were carried out irrespective of whether or not there was strong evidence of a treatment effect associated with the primary outcome. As missing data for the clinical effectiveness primary outcome were anticipated to be minimal, a sensitivity approach was taken when the primary outcome was missing (see Secondary analyses of the primary outcome). Missing data for the cost-effectiveness analysis, as well as missing baseline data for adjusted analysis of clinical outcomes, were handled by multiple imputation.

Multiple imputation

Missing data in baseline covariates, resource use and health-related quality-of-life variables at 90 days and 1 year were handled with multivariate imputation by chained equations. 41 Under this approach, each variable was imputed conditional on fully observed baseline variables, such as age, sex, ICNARC Physiology Score, presence of cancer, length of stay in critical care and general medical wards, and all other imputed variables. When addressing the missing data, multiple imputation assumes that the data are missing at random, conditional on the observed data.

Patients who did not return or fully complete the EQ-5D-5L questionnaire administered at 90 days had their EQ-5D-5L utility scores imputed from those survivors who did fully complete the questionnaire. Similarly, for those eligible patients who did not return the Health Services Questionnaire, information on the use of outpatient services up to 90 days following randomisation was imputed from those patients who did complete this questionnaire. Health Services Questionnaire costs and quality-of-life end points were conditional on survival status; as such, the imputation was conducted in two stages. In the first stage, imputation models were specified for mortality at 90 days according to baseline covariates and auxiliary variables, including hospital length of stay of up to 90 days. In the second stage, for each of the imputed data sets from stage 1, imputation models were specified for Health Services Questionnaire costs and quality of life at 90 days for those patients who were missing these but were known to be alive at 90 days, or were predicted to be alive by the first stage imputation model.

Patients who did not return or fully complete the EQ-5D-5L questionnaire or the Health Services Questionnaire administered at 1 year also had their information imputed from those survivors who did fully complete the questionnaire, using a similar two-stage approach.

The resultant estimates were combined with Rubin’s rules, which recognise uncertainty both within and between imputations. 42 All multiple imputation models were implemented in the statistical package Stata/SE version 13.0 (StataCorp LP, College Station, TX, USA).

Statistical analysis: clinical effectiveness

Statistical analyses were conducted according to a pre-specified, published statistical analysis plan written prior to the interim analysis. 43 The final analyses were conducted using Stata/SE version 13.0.

Cost-effectiveness analysis

A full cost-effectiveness analysis was undertaken to assess which route of delivery for early nutritional support in critically ill adults, parenteral compared with enteral, was most cost-effective. This analysis assessed whether or not the additional intervention costs of early nutritional support via the parenteral route compared with the enteral route were justified by any subsequent reductions in morbidity costs and/or improvements in patient outcomes. The cost-effectiveness analysis was reported for three time periods: randomisation to 90 days; randomisation to 1 year; and lifetime. For each time period the cost-effectiveness analysis took a health and personal health services perspective,47 using information on health-related quality of life collected at 90 days and 1-year follow-up, combined with information on vital status, to report QALY. Each QALY was valued using the NICE-recommended threshold of willingness to pay for a QALY gain (£20,000),47 in conjunction with the costs of each strategy to report the incremental net monetary benefits (INBs) of early nutritional support via the parenteral route compared with early nutritional support via the enteral route, overall and for the same pre-specified subgroups as for the evaluation of clinical effectiveness (see Subgroup analyses of the primary outcome, below).

The primary objective of the cost-effectiveness analysis was to compare incremental cost-effectiveness at 1 year of early nutritional support via the parenteral route compared with the enteral route. The secondary objectives were to:

• compare health-related quality of life at 90 days and 1 year between the treatment groups

• compare resource use and costs at 90 days and 1 year between the treatment groups

• estimate the lifetime incremental cost-effectiveness between the treatment groups.

The main assumptions of the cost-effectiveness analysis were subjected to extensive sensitivity analyses.

Participants: sites

Expressions of interests were received from 58 NHS adult, general critical care units in the UK. A total of 34 hospitals in England obtained local NHS permissions and opened to recruitment between 17 June 2011 and 14 October 2013. Twenty-one sites were opened within the first 3 months of the trial opening on 17 June 2011 (Figure 4).

FIGURE 4.

Sites open to recruitment during the trial recruitment period.

The median (IQR) time from local NHS permission to the trial opening at sites (i.e. start of screening) was 63 (23–88) days (Figure 5). Reasons for delays in opening were issues related to confirmation of NHS support costs from the CLRN and delays in local set-up of the trial, for example training staff.

FIGURE 5.

Time (in days) from local NHS permission to start of screening.

Overall, sites participated in the CALORIES trial for a median (IQR) of 31 (13–33) months. Of the 34 sites that opened, five were closed early because of poor recruitment (one unit did not recruit any patients) and one closed due to loss of eligibility to participate in the trial (no longer actively participating in the Case Mix Programme). There were 28 sites that remained open until the end of recruitment in March 2014 (Figure 6).

FIGURE 6.

Duration of participation of sites.

Participants: patients

In total, 11,108 patients were screened between 17 June 2011 and 2 March 2014. Of these, 6195 (55.8%) met one or more exclusion criteria. There were 2513 (22.6%) patients who, although eligible for inclusion in the trial, were not recruited. The most frequently reported reason for not recruiting eligible patients was that the treating clinician had refused to recruit the patient. Other reported reasons included: the patient (or their personal/professional consultee) declined to take part; the patient was unable to provide informed consent and a personal or professional consultee was unavailable or unable to provide agreement; or logistical issues, which were mainly due to no research staff being available, for example the patient was not screened for eligibility prior to commencement of early nutritional support (see Figure 7).

The required sample of 2400 (21.6%) patients was recruited between 22 June 2011 and 2 March 2014, with 1200 randomised to receive early nutritional support via the parenteral route and 1200 to receive early nutritional support via the enteral route (Figures 7 and 8). There was variation across the 33 sites in the rate of recruitment (Figure 9); the overall median (IQR) recruitment rate was 0.6 (0.4–0.8) patients per site per week, with the highest recruitment rate of 1.2 patients per week. Manual randomisation was required for two patients as a result of the telephone randomisation service being briefly unavailable.

FIGURE 7.

Screening, randomisation and follow-up. PEG, percutaneous endoscopic gastrostomy; PEJ, percutaneous endoscopic jejunostomy.

FIGURE 8.

Patient recruitment.

FIGURE 9.

Patient recruitment rate (patients per week).

Patients were generally recruited into the CALORIES trial during week days (Monday to Friday) and during usual office hours (Figures 10 and 11); most of the recruiting sites reported having insufficient resources to enable screening and recruitment at weekends and outside usual office hours.

FIGURE 10.

Randomisation by day of week.

FIGURE 11.

Randomisation by time of day.

For the majority of patients, agreement to participate in the trial was obtained from a personal consultee prior to randomisation (n = 2212, 92.2%). For the remaining patients, agreement was obtained from a professional consultee (6.1%) or informed consent was obtained from the patient prior to randomisation (1.7%) (Table 6). Twelve patients withdrew from the trial, requesting removal of all of their data from the analysis, resulting in data on 2388 patients for initial analysis (n = 1191 parenteral group; n = 1197 enteral group). Five patients were subsequently lost to follow-up before 30 days, resulting in data from 2383 patients for analysis of the primary clinical effectiveness outcome (n = 1188, 99.7% parenteral group; n = 1195, 99.8% enteral group) (see Figure 7). Four patients withdrew from follow-up (n = 3 parenteral group; n = 1 enteral group) and one was lost to follow-up (enteral group) before 90 days. A further two patients withdrew from follow-up (both parenteral group) and three were lost to follow-up (n = 2 parenteral group; n = 1 enteral group) before 1 year, resulting in final follow-up of 2372 patients for 1-year mortality (n = 1181, 99.2% parenteral group; n = 1192, 99.5% enteral group). Follow-up was completed on 23 March 2015.

Adherence to the protocol

Overall, adherence to delivery of nutritional support during the intervention period was high (Table 9). A similar number of patients in each treatment group received nutritional support throughout the intervention period. Any non-adherence to the protocol was reported for 150 (12.6%) patients in the parenteral group and 127 (10.6%) patients in the enteral group (Table 10).

Ninety-seven per cent of patients received nutritional support via the assigned route, with 24 patients assigned to the parenteral route and 26 patients assigned to the enteral route receiving no nutritional support (Table 11). The primary reasons in both groups were due to withdrawal of treatment or death [11 (45.8%) in the parenteral group compared with 11 (42.3%) in the enteral group] and unexpected early extubation [six (25.0%) in the parenteral group vs. seven (26.9%) in the enteral group]. Twelve patients assigned to the parenteral route and four patients assigned to the enteral route received their first nutritional support via the opposite route to that assigned (Table 12). Initiation of nutritional support was delayed beyond 36 hours from original critical care unit admission for 37 patients assigned to the parenteral route and 41 patients assigned to the enteral route (Table 13). In the parenteral group, this was primarily due to delays in prescribing or obtaining PN (n = 12, 32.4%) and problems (i.e. insertion difficulties or delays) with the central line (n = 8, 21.6%). In the enteral group, over half of the delays were due to problems (i.e. insertion difficulties or delays) with the nasogastric tube (n = 24, 58.5%).

Of patients who initially received nutrition via the assigned route, 81 assigned to the parenteral route and 18 assigned to the enteral route subsequently received nutritional support via the alternative route (i.e. ‘crossed over’) during the 120-hour intervention period (Table 14). However, the greatest proportion of these were towards the end of the 120-hour period, with 27 of the crossovers among those assigned to the parenteral route occurring in the final 6 hours of the 120-hour intervention period and 41 in the final 24 hours (Figure 12; see also Table 14). Reasons for these late crossovers were primarily due to an error by the clinical team regarding the precise timing of the end of the 120-hour period (n = 24 in final 24 hours).

FIGURE 12.

Timing of non-adherence according to treatment group. (a) The timing of crossovers in hours from initiation of nutritional support; and (b) days (from initiation of nutritional support) on which patients received no nutritional support for the full day. Seven patients received no nutritional support on more than 1 day.

Patients assigned to the enteral route were more likely to have full days with no nutritional support; this was more likely to occur in the last 48 hours of the intervention period (see Figure 12; Table 15). Among those receiving nutritional support, the mean number of full days with no nutritional support was 0.004 for the parenteral group and 0.044 for the enteral group. Reasons for no nutritional support for a full day in the enteral group were primarily due to the patient vomiting or high volumes of aspirates via the nasogastric tube (n = 12). Nutritional support was stopped temporarily in eight patients for surgery, and a scan or another intervention and was stopped permanently in eight patients because treatment was withdrawn and/or the patient was receiving palliative care.

Delivery of care by treatment group

In both treatment groups, the median time to initiation of nutritional support was within 24 hours of critical care unit admission [median (IQR) 23.5 (16.8–30.0) hours in the parenteral group and 21.8 (15.5–27.8) hours in the enteral group] (Table 16).

The mean total calories received during the 120-hour intervention period was higher in the parenteral group than the enteral group [mean (SD) 88.7 (44.0) kcal/kg in the parenteral group vs. 74.2 (44.2) kcal/kg in the enteral group] (see Table 16). On average, daily caloric intake was higher in patients assigned to the parenteral route [mean (SD) 21.3 (7.7) kcal/kg/day] than in those assigned to the enteral route [mean (SD) 18.5 (7.7) kcal/kg/day] (Table 17; Figure 13). However, in the majority of patients, the targeted delivery of 25 kcal/kg/day was not achieved irrespective of route of delivery (Figure 14).

FIGURE 13.

Daily calories according to treatment group, showing the total calories received per kilogram of actual (or estimated) body weight for each day from days 1–6. Day 1 data are the values from the time of initiation of nutritional support to the end of the day of initiation of nutritional support. The horizontal lines within the boxes indicate medians, the lower and upper ends of the boxes indicate the 25th and 75th percentiles, respectively, and the lower and upper whiskers indicate the 1st and 99th percentiles, respectively.

FIGURE 14.

Percentage of patients meeting daily energy target according to treatment group, showing the percentage of patients who met the daily energy target of 25 kcal/kg/day for each day from days 1–6. Day 1 data are the values from the time of initiation of nutritional support to the end of the day of initiation of nutritional support. The faded bars show the percentage of patients who met the target having adjusted for part-days of feeding (e.g. following the death of a patient).

The mean total protein received during the 120-hour intervention period was similar in the two groups [mean (SD) 2.9 (1.6) g/kg in the parenteral group vs. 2.7 (1.8) g/kg in the enteral group] (see Table 16). The average daily protein received per day was also similar for the two groups [mean (SD) 0.7 (0.3) g/kg/day in the parenteral group vs. 0.6 (0.3) g/kg/day in the enteral group] (see Table 17; Figure 15).

FIGURE 15.

Daily protein according to treatment group. Shows the total protein received for each day from days 1–6. Day 1 data are the values from the time of initiation of nutritional support to the end of the day of initiation of nutritional support. The horizontal lines within the boxes indicate medians, the lower and upper ends of the boxes indicate the 25th and 75th percentiles, respectively, and the lower and upper whiskers indicate the 1st and 99th percentiles, respectively.

For patients assigned to the enteral route, the mean total volume of gastric aspirates during the intervention period was 958 ml (253 ml/day) and one-third received prokinetics (see Tables 16 and 17). These were recorded only for patients receiving EN.

For patients assigned to the parenteral route, the most frequent additive to PN during the intervention period was selenium (22.5%, compared with 4.2% for glutamine and 3.2% for fish oils). These were recorded only for patients receiving PN.

In both groups, a high proportion of patients received vasoactive agents, although this was slightly higher in the enteral group (84.6%) than in the parenteral group (80.9%). In both groups, more than half of the patients received insulin during the intervention period (58.6% in the parenteral group and 56.1% in the enteral group) (see Table 16). However, patients in the parenteral group received a higher total amount of insulin [mean (SD) 153.5 (223.0) IU] than patients in the enteral group [mean (SD) 122.3 (204.1) IU].

Median daily SOFA scores were similar in both groups, and decreased during the intervention period (Figure 16).

FIGURE 16.

Daily SOFA score according to treatment group. Shows the SOFA scores for the 24 hours prior to randomisation (baseline) and from days 1–6. Day 1 data are the values from the time of initiation of nutritional support to the end of the day of initiation of nutritional support. The horizontal lines within the boxes indicate medians, the lower and upper ends of the boxes indicate the 25th and 75th percentiles, respectively, and the lower and upper whiskers indicate the 1st and 99th percentiles, respectively. SOFA scores range from 0 to 24, with higher scores indicating a greater degree of organ failure.

Patients in the enteral group had a higher incidence of diarrhoea (21% vs. 16% in the parenteral group), whereas patients in the parenteral group had a higher incidence of constipation (61% vs. 54% in the enteral group) (see Table 16).

Median time from randomisation to exclusive oral feeding was about 2 weeks in both treatment groups [median (IQR) 14 (5–36) days in the parenteral group and 13 (5–32) days in the enteral group] (see Table 16).

Table 18 details the nutritional support delivered to patients, by treatment group, in the critical care unit after the 120-hour intervention period, that is, from 120 hours following initiation of nutritional support to critical care unit discharge or death. In both groups the most frequently used route for delivery of nutritional support was the enteral route either exclusively (30% for the parenteral group and 43% for the enteral group) or with oral feeding (22% and 35%, respectively) (see Table 18). However, those assigned to the parenteral route were more likely to continue to receive PN, either exclusively or with other routes, beyond 120 hours (37% compared with 7% for the enteral group).

Primary outcome: clinical effectiveness

At 30 days following randomisation, 393 (33.1%) patients assigned to receive nutritional support via the parenteral route had died compared with 409 (34.2%) patients assigned to receive nutritional support via the enteral route, corresponding to an absolute risk reduction of 1.15 percentage points (95% CI −2.65 to 4.94; p = 0.57) and a relative risk of 0.97 (95% CI 0.86 to 1.08). This difference remained non-significant after adjustment for baseline characteristics (odds ratio 0.95, 95% CI 0.79 to 1.13; p = 0.55) (Table 19).

Secondary outcomes: clinical effectiveness

The proportions of patients in the parenteral group who experienced episodes of hypoglycaemia (p = 0.006) and of vomiting (p < 0.001) were significantly lower than for patients in the enteral group. There were no significant differences between the groups in any of the other secondary outcomes, including duration of survival (log-rank test p = 0.056; adjusted Cox proportional hazards regression hazard ratio 0.90, 95% CI 0.80 to 1.00; p = 0.057) (Table 20; Figures 17 and 18).

FIGURE 17.

Kaplan–Meier curves for survival to 90 days following randomisation.

FIGURE 18.

Kaplan–Meier curves for survival to 1 year following randomisation.

Safety monitoring

Fifty-eight patients in both the parenteral group (4.9%) and in the enteral group (4.8%) experienced one or more serious adverse events (Table 21). The most frequently reported serious adverse events were ischaemic bowel [n = 8 (0.7%) in the parenteral group and n = 11 (0.9%) in the enteral group], cardiac arrest [n = 9 (0.8%) in the parenteral group and n = 6 (0.5%) in the enteral group], electrolyte disturbance [n = 8 (0.7%) in the parenteral group and n = 5 (0.4%) in the enteral group], raised liver enzymes [n = 3 (0.3%) in the parenteral group and n = 7 (0.6%) in the enteral group] and gastrointestinal haemorrhage [n = 3 (0.3%) in the parenteral group and n = 8 (0.7%) in the enteral group]. There were eight episodes of hypoglycaemia [n = 5 (0.4%) in the parenteral group and n = 3 (0.3%) in the enteral group] that were reported as a serious adverse event (see Table 21).

There were five serious, unexpected adverse events that were deemed by the site investigator to be potentially related to the study treatment in four patients (one with ischaemic bowel and hypoglycaemia, and one each with upper gastrointestinal haemorrhage and anterolateral myocardial infarction in the parenteral group, and one with lower gastrointestinal haemorrhage in the enteral group).

Subgroup analyses of the primary outcome

There was no statistically significant interaction between the effect of treatment group on 30-day mortality and any of the pre-specified subgroups: age, degree of malnutrition, APACHE II predicted risk of death,34 ICNARC predicted risk of death,35 use of mechanical ventilation, presence of cancer and time from randomisation to start of feeding. The p-values ranged from 0.15 to 0.83 (Figure 19).

FIGURE 19.

Subgroup analyses of the primary outcome (30-day mortality). The p-values are for tests of interaction. The x-axis is presented on a log-scale. The solid line represents no difference between the groups and the dashed line represents the overall effect estimate.

Secondary analyses of the primary outcome

Only five patients were lost to follow-up prior to 30 days and, therefore, making extreme assumptions for the outcomes of these patients made minimal difference to the result (relative risk of 30-day mortality between 0.96 and 0.97).

Adjusting for non-adherence also made minimal difference to the result (relative risk of 30-day mortality 0.96, 95% CI 0.85 to 1.09; p = 0.55).

Cost-effectiveness at 90 days following randomisation

Cost-effectiveness at 90 days

The incremental QALY gain for early nutritional support via the parenteral versus the enteral route was small and with a 95% CI that included zero (Table 27). The average costs were higher for the parenteral group, but this difference was not statistically significant. The INB for the parenteral route compared with the enteral route was negative at −£1263 (95% CI −£2952 to £426) (see Table 27).

TABLE 27 - Cost-effectiveness at 90 days: QALYs, total costs (£) and INB (INB, £)

The QALYs, costs and INB results are all reported after applying multiple imputation to handle missing data.

The INB is calculated according to NICE methods guidance, by multiplying the mean QALY gain (or loss) by £20,000 and subtracting from this the incremental cost.

All values are mean (SD), unless stated otherwise.

End point	Parenteral group (n = 1191)	Enteral group (n = 1197)	Incremental effect (unadjusted), mean (95% CI)	p-value
QALYsa	0.051 (0.048)	0.050 (0.049)	0.002 (−0.002 to 0.006)	0.46
Costs (£)a	24,458 (21,400)	23,164 (20,449)	1293 (−401 to 2988)	0.14
INB (£)a,b			−1263 (−2952 to 426)	0.14

When the uncertainty in the incremental costs and QALYs is represented on the cost-effectiveness plane, the majority of the points are in the quadrant that shows that early nutritional support via the parenteral route has, on average, higher costs and improves QALYs, although the magnitude of these average QALY gains was small (Figure 20). The probability that early nutritional support via the parenteral route is more cost-effective at 90 days than via the enteral route – given the data – is never > 10%, irrespective of how much society is willing to pay for a QALY gain (Figure 21).

FIGURE 20.

Uncertainty in the mean costs (£) and QALY differences and their distribution for early nutritional support via the parenteral vs. the enteral route (within 90 days post-randomisation).

FIGURE 21.

Cost-effectiveness acceptability curve, reporting the probability that early nutritional support via the parenteral route is cost-effective (within 90 days post-randomisation) at alternative willingness to pay for a QALY gain.

The estimated INBs were similar across all the scenarios considered in the sensitivity analyses (Figure 22). For example, the INB remains around −£1300 whether additional staff time is required to deliver nutritional support in the critical care or list prices of feeding products are considered. Similarly, excluding readmissions that were reported from responses to the Health Services Questionnaire – to avoid any risk of double-counting – had only a small impact on the mean INB (−£1147 vs. −£1263).

FIGURE 22.

Sensitivity analyses that report the mean (95% CI) INB (at £20,000 per QALY) within 90 days post-randomisation according to alternative assumptions compared with the base case. The vertical dashed line indicates INBs in the base-case analysis. The solid vertical line indicates no difference in net monetary benefits between comparator groups.

The results of the subgroup analyses are presented in Table 28, and show that the incremental QALYs were similar across all subgroups. Although there were some subgroups of patients for whom the incremental costs of early nutritional support via the parenteral route were negative, and hence the INBs were positive, the statistical uncertainty surrounding these findings was high. For patients with higher APACHE II predicted risk of death (0.52–0.98), early nutritional support via the parenteral route was more costly and the QALY gain was small, leading to significant negative INB. For all other subgroups, as for the overall results, the CIs around the INB included zero.

TABLE 28 - Incremental cost, incremental QALY and INB (at £20,000 per QALY) within 90 days post-randomisation, by pre-specified subgroups

Subgroup		Incremental cost (£) (95% CI)	Incremental QALY (95% CI)	INB (£) (95% CI)
Age (years)	18–53	764 (−2611 to 4140)	0.002 (−0.006 to 0.01)	−729 (−4096 to 2638)
	54–65	2027 (−1453 to 5508)	0.007 (−0.001 to 0.015)	−1893 (−5363 to 1577)
	66–73	1625 (−1814 to 5065)	−0.002 (−0.01 to 0.006)	−1659 (−5090 to 1772)
	74–100	1039 (−2149 to 4227)	0.000 (−0.007 to 0.008)	−1030 (−4209 to 2149)
Degree of malnutrition	None	1171 (−590 to 2933)	0.001 (−0.003 to 0.005)	−1148 (−2904 to 608)
	Moderate/severe	2683 (−3573 to 8939)	0.004 (−0.011 to 0.02)	−2593 (−8830 to 3644)
APACHE II predicted risk	0.01–0.18	−838 (−4253 to 2578)	0.005 (−0.003 to 0.013)	934 (−2469 to 4337)
	0.18–0.34	1669 (−1709 to 5047)	−0.002 (−0.01 to 0.006)	−1706 (−5075 to 1663)
	0.34–0.52	−429 (−3906 to 3047)	0.002 (−0.006 to 0.01)	460 (−3006 to 3926)
	0.52–0.98	4494 (1182 to 7807)	0.001 (−0.006 to 0.009)	−4469 (−7773 to −1165)
ICNARC model predicted risk	0.00–0.22	−1520 (−4899 to 1859)	0.005 (−0.003 to 0.013)	1618 (−1751 to 4987)
	0.22–0.43	2735 (−651 to 6122)	−0.003 (−0.011 to 0.005)	−2798 (−6177 to 581)
	0.43–0.65	1353 (−2024 to 4730)	0.002 (−0.006 to 0.01)	−1311 (−4679 to 2057)
	0.65–0.99	2606 (−757 to 5969)	0.002 (−0.006 to 0.009)	−2572 (−5926 to 782)
Mechanically ventilated	No	2514 (−1666 to 6695)	0.007 (−0.003 to 0.017)	−2372 (−6541 to 1797)
	Yes	1053 (−801 to 2907)	0.000 (−0.004 to 0.005)	−1044 (−2892 to 804)
Presence of cancer	No	1378 (−370 to 3125)	0.001 (−0.003 to 0.005)	−1356 (−3098 to 386)
	Yes	−581 (−7453 to 6292)	0.007 (−0.009 to 0.023)	718 (−6132 to 7568)
Time to start of feeding (hours)	< 24	1482 (−796 to 3759)	0.001 (−0.004 to 0.006)	−1460 (−3731 to 811)
	≥ 24	741 (−1876 to 3358)	0.002 (−0.005 to 0.008)	−704 (−3314 to 1906)
Adherence adjusted		1400 (−435 to 3235)	0.002 (−0.003 to 0.006)	−1367 (−3196 to 462)

Cost-effectiveness at 1 year following randomisation (primary outcome)

Cost-effectiveness at 1 year

The incremental QALY gain for the parenteral group compared with the enteral group was positive, but with a 95% CI that included zero (Table 34). The mean total costs were higher in the parenteral group, with an incremental cost of £1580 (95% CI −£792 to £3951). Hence the INB for the parenteral group compared with the enteral group was negative at −£1320 (95% CI −£3709 to £1069).

TABLE 34 - Cost-effectiveness at 1 year: QALYs, total costs (£) and INB (INB, £)

The QALYs, costs and INB results are all reported after applying multiple imputation to handle missing data.

The INB is calculated according to NICE methods guidance, by multiplying the mean QALY gain (or loss) by £20,000, and subtracting from this the incremental cost.

All numbers are mean (SD), unless stated otherwise.

End point	Parenteral group (N = 1191)	Enteral group (N = 1197)	Incremental effect (unadjusted), mean (95% CI)	p-value
QALYsa	0.348 (0.333)	0.335 (0.332)	0.013 (−0.014 to 0.040)	0.35
Costs (£)a	28,354 (32,144)	26,775 (26,273)	1580 (−792 to 3951)	0.19
INB (£)a,b			−1320 (−3709 to 1069)	0.28

When the uncertainty in the incremental costs and QALYs is represented on the cost-effectiveness plane, the majority of the points are in the quadrant that shows early nutritional support via the parenteral route has, on average, higher costs and improves QALYs, but again the magnitude of the average QALY gains was small (Figure 23). The cost-effectiveness acceptability curve (Figure 24) shows that at 1 year the probability that early nutritional support via the parenteral route is more cost-effective than via the enteral route – given the data – is < 20% at the £20,000 willingness-to-pay threshold stipulated by NICE and does not exceed 50% even at £100,000 per QALY.

FIGURE 23.

Uncertainty in the mean costs (£) and QALY differences and their distribution for early nutritional support via the parenteral route vs. the enteral route (within 1 year post-randomisation).

FIGURE 24.

Cost-effectiveness acceptability curve, reporting the probability that early nutritional support via the parenteral route is cost-effective (within 1 year post-randomisation) at alternative willingness to pay for a QALY gain.

The estimated INBs were similar across all of the scenarios considered in the sensitivity analyses (Figure 25). This shows that the base-case results are robust to alternative assumptions.

FIGURE 25.

Sensitivity analyses that report the mean (95% CI) INB (at £20,000 per QALY) within 1 year post-randomisation according to alternative assumptions compared with the base case. The vertical dashed line indicates INBs in the base-case analysis. The solid vertical line indicates no difference in net monetary benefits between comparator groups.

The estimated INBs were similar across all pre-specified subgroups (Table 35). As for the results at 90 days, although there were some subgroups of patients for whom early nutritional support via the parenteral route was cost-saving and hence their INBs were positive, there was high statistical uncertainty around these findings. For patients with higher APACHE II predicted risk of death (0.52–0.98), early nutritional support via the parenteral route was more costly and the INB was negative, and both of these end points were statistically significant for this subgroup. For all other subgroups, as for the overall results, the CIs around the INBs included zero.

TABLE 35 - Incremental cost, incremental QALY and INB (at £20,000 per QALY) within 1 year post-randomisation, by pre-specified subgroups

Subgroup		Incremental cost (£) (95% CI)	Incremental QALY (95% CI)	INB (£) (95% CI)
Age (years)	18–53	2017 (−2743 to 6777)	0.016 (−0.037 to 0.07)	−1691 (−6509 to 3127)
	54–65	2396 (−2491 to 7283)	0.033 (−0.022 to 0.088)	−1731 (−6671 to 3209)
	66–73	983 (−3843 to 5808)	0.005 (−0.049 to 0.06)	−880 (−5754 to 3994)
	74–100	1283 (−3189 to 5756)	0.005 (−0.045 to 0.056)	−1177 (−5700 to 3346)
Degree of malnutrition	None	1308 (−1157 to 3774)	0.011 (−0.017 to 0.039)	−1086 (−3570 to 1398)
	Moderate/severe	4735 (−4052 to 13521)	0.025 (−0.078 to 0.128)	−4235 (−13103 to 4633)
APACHE II predicted risk	0.01–0.18	−1793 (−6581 to 2995)	0.041 (−0.013 to 0.094)	2611 (−2210 to 7432)
	0.18–0.34	1776 (−2959 to 6512)	−0.01 (−0.064 to 0.043)	−1985 (−6770 to 2800)
	0.34–0.52	−442 (−5308 to 4423)	0.006 (−0.048 to 0.06)	564 (−4333 to 5461)
	0.52–0.98	6380 (1745 to 11014)	0.013 (−0.04 to 0.065)	−6129 (−10810 to −1448)
ICNARC model predicted risk	0.00–0.22	−2692 (−7441 to 2058)	0.034 (−0.018 to 0.086)	3370 (−1430 to 8170)
	0.22–0.43	4059 (−701 to 8818)	−0.01 (−0.063 to 0.043)	−4260 (−9078 to 558)
	0.43–0.65	612 (−4155 to 5379)	0.009 (−0.043 to 0.061)	−428 (−5238 to 4382)
	0.65–0.99	4272 (−443 to 8988)	0.015 (−0.037 to 0.066)	−3977 (−8736 to 782)
Mechanically ventilated	No	3114 (−2746 to 8973)	0.056 (−0.012 to 0.123)	−1998 (−7910 to 3914)
	Yes	1280 (−1320 to 3880)	0.005 (−0.025 to 0.034)	−1187 (−3806 to 1432)
Presence of cancer	No	1743 (−702 to 4189)	0.009 (−0.02 to 0.037)	−1573 (−4037 to 891)
	Yes	−1850 (−11479 to 7779)	0.065 (−0.045 to 0.174)	3142 (−6565 to 12,849)
Time to start of feeding	< 24 hours	2766 (−443 to 5974)	0.015 (−0.022 to 0.051)	−2473 (−5704 to 758)
	≥ 24 hours	−90 (−3778 to 3598)	0.01 (−0.033 to 0.052)	282 (−3441 to 4005)
Adherence adjusted		1710 (−858 to 4277)	0.014 (−0.015 to 0.044)	−1428 (−4014 to 1158)

Lifetime incremental cost-effectiveness

Long-term survival

The Kaplan–Meier survival curves show that when the time horizon was extended beyond 1 year, for those with survival data available, the probability of survival was higher for the parenteral group than the enteral group (Figure 26). The survival difference between the treatment groups was maintained over time, even though a large proportion of cases were censored beyond year 2 following randomisation.

FIGURE 26.

Kaplan–Meier survival curves.

To calculate QALYs over 20 years, the long-term survival for each patient was estimated by combining the observed survival for each patient up to 1 year with their predicted survival for years 2–20. Alternative parametric extrapolation approaches were compared with predict the longer-term survival of patients recruited to the CALORIES trial. Figure 27 considers alternative parametric extrapolations for CALORIES trial patients, using the observed survival data after day 30. The survival data were pooled across the treatment groups, given that there was no statistically significant evidence of treatment effect on survival at 1 year and beyond. The parametric models predict excess mortality in patients recruited to the CALORIES trial compared with the age–gender-matched general population. Of the alternative survival functions, the log-normal distribution appears to fit the observed data best, in that it reports the lowest Akaike information criterion (AIC) and Bayesian information criterion (BIC) (Table 36). It also offers the most plausible projections of future survival (see Figure 27 and Table 36), for which the excess mortality of the CALORIES trial patients is maintained over the time horizon of the study. In the base case, death rates were applied according to the most plausible parametric model (i.e. log-normal) for years 2–20. The logistic model also provides a reasonable fit to the observed data, but the excess mortality is higher than that of the base-case parametric model. Therefore, a sensitivity analysis extrapolating survival according to the logistic function was run.

FIGURE 27.

Comparison of alternative parametric survival functions applied to CALORIES data after day 30.

TABLE 36 - Fit of alternative parametric survival functions applied to CALORIES data after day 30

Distribution	AIC	BIC
Gompertz	2930.062	2962.260
Log-normal	2862.885	2895.083
Logistic	2867.942	2900.140
Weibull	2870.448	2902.647
Exponential	3003.800	3030.632

Lifetime incremental cost-effectiveness

Table 37 presents the resultant lifetime QALYs, costs and INB according to the base case assumptions. Overall, at the NICE-stipulated threshold of £20,000 per QALY, the INB was positive (£440) but with a wide 95% CI that included zero.

TABLE 37 - Lifetime cost-effectiveness: QALYs, total costs (£) and INB (INB, £)

The QALYs, costs and INB results are all reported after applying multiple imputation to handle missing data.

The INB is calculated according to NICE methods guidance, by multiplying the mean QALY gain (or loss) by £20,000, and subtracting from this the incremental cost.

All numbers are mean (SD), unless stated otherwise.

End point	Parenteral group (n = 1191)	Enteral group (n = 1197)	Incremental effect (unadjusted), mean (95% CI)	p-value
QALYsa	3.996 (3.432)	3.849 (3.448)	0.147 (−0.129 to 0.423)	0.30
Costs (£)a	53,100 (42,282)	50,595 (37,968)	2505 (−733 to 5744)	0.13
INB (£)a,b			440 (−3586 to 4466)	0.83

Figure 28 presents the uncertainties in costs and QALYs extrapolated to the lifetime. There are considerable uncertainties around both costs and QALYs, with the majority of the points in the quadrant which shows that early nutritional support via the parenteral route has, on average, higher costs and improves QALYs. The cost-effectiveness acceptability curve shows that that the probability of early nutritional support via the parenteral route being most cost-effective is around 60% at the NICE-stipulated threshold of £20,000 per QALY, increasing to around 80% at higher willingness to pay (Figure 29).

FIGURE 28.

Uncertainty in the mean costs (£) and QALY differences and their distribution for early nutritional support via the parenteral vs. the enteral route (at lifetime).

FIGURE 29.

Cost-effectiveness acceptability curve, reporting the probability that early nutritional support via the parenteral route is cost-effective (at lifetime) at alternative willingness to pay for a QALY gain.

The sensitivity analyses on the lifetime results suggest that these findings are robust to alternative assumptions, including those applied to extrapolation of long-term survival and to quality of life for survivors (Figure 30). For example, alternative assumptions for survival extrapolation had small effect on the mean INB. Similarly, a large decrement compared with a smaller decrement in quality of life had only marginal impact on the mean INB.

FIGURE 30.

Sensitivity analyses that report the mean (95% CI) INB (at £20,000 per QALY) at lifetime according to alternative assumptions compared with the base case. The vertical dashed line indicates INBs in the base-case analysis. The solid vertical line indicates no difference in net monetary benefits between comparator groups.

The results of the subgroup analyses presented in Table 38 show that there were some differences in the direction of mean incremental effects, but high statistical uncertainty surrounds these findings. For each subgroup, as for the overall results, there was high statistical uncertainty surrounding INBs and all 95% CIs included zero.

TABLE 38 - Lifetime incremental cost, lifetime incremental QALY and lifetime INB (at £20,000 per QALY), by pre-specified subgroups

Subgroup		Incremental cost (£) (95% CI)	Incremental QALY (95% CI)	INB (£) (95% CI)
Age (years)	18–53	4246 (−2125 to 10,617)	0.342 (−0.191 to 0.875)	2596 (−5317 to 10,509)
	54–65	4558 (−1990 to 11,106)	0.35 (−0.2 to 0.899)	2434 (−5720 to 10,588)
	66–73	1192 (−5285 to 7668)	0.034 (−0.51 to 0.578)	−514 (−8568 to 7540)
	74–100	1057 (−4942 to 7056)	−0.03 (−0.533 to 0.474)	−1649 (−9104 to 5806)
Degree of malnutrition	None	2094 (−1269 to 5458)	0.125 (−0.162 to 0.411)	400 (−3781 to 4581)
	Moderate/severe	6654 (−5318 to 18,626)	0.306 (−0.723 to 1.335)	−529 (−15,589 to 14,531)
APACHE II predicted risk	0.01–0.18	304 (−6116 to 6724)	0.347 (−0.187 to 0.881)	6636 (−1259 to 14,531)
	0.18–0.34	1958 (−4391 to 8307)	0.014 (−0.516 to 0.544)	−1679 (−9521 to 6163)
	0.34–0.52	−1810 (−8362 to 4742)	−0.198 (−0.743 to 0.348)	−2142 (−10,160 to 5876)
	0.52–0.98	8776 (2562 to 14,991)	0.364 (−0.155 to 0.884)	−1488 (−9170 to 6194)
ICNARC model predicted risk	0.00–0.22	−1038 (−7359 to 5282)	0.275 (−0.251 to 0.8)	6529 (−1310 to 14,368)
	0.22–0.43	5481 (−867 to 11,829)	0.197 (−0.329 to 0.723)	−1539 (−9359 to 6281)
	0.43–0.65	−246 (−6592 to 6101)	−0.119 (−0.643 to 0.404)	−2141 (−9927 to 5645)
	0.65–0.99	5438 (−853 to 11,730)	0.185 (−0.334 to 0.705)	−1730 (−9447 to 5987)
Mechanically ventilated	No	6324 (−1658 to 14,305)	0.527 (−0.153 to 1.207)	4217 (−5726 to 14,160)
	Yes	1762 (−1783 to 5308)	0.074 (−0.229 to 0.376)	−292 (−4705 to 4121)
Presence of cancer	No	2421 (−908 to 5750)	0.106 (−0.177 to 0.39)	−298 (−4439 to 3843)
	Yes	1016 (−12,078 to 14,109)	0.486 (−0.628 to 1.6)	8700 (−7597 to 24,997)
Time to start of feeding	< 24 hours	4014 (−345 to 8373)	0.196 (−0.175 to 0.568)	−85 (−5524 to 5354)
	≥ 24 hours	394 (−4614 to 5401)	0.079 (−0.348 to 0.505)	1181 (−5070 to 7432)
Adherence adjusted		2711 (−794 to 6217)	0.159 (−0.14 to 0.458)	476 (−3881 to 4833)

Principal findings

Among adults with an unplanned critical care unit admission for whom early nutritional support could be provided through either the parenteral or the enteral route, there was no significant difference in mortality at 30 days according to the route of delivery. In addition, there was no significant interaction on the basis of age, degree of malnutrition, severity of illness or timing of the initiation of nutritional support. The enteral route was associated with significantly more episodes of hypoglycaemia and vomiting, but there were no significant differences between treatment groups in the duration of organ support, infectious complications, critical care unit or hospital length of stay, or duration of survival up to 1 year. The energy target of 25 kcal/kg/day was not reached in a majority of patients in each treatment group.

Providing nutritional support to critically ill adult patients via the parenteral route compared with the enteral route is unlikely to be cost-effective. At the primary end point for the cost-effectiveness analysis at 1 year post-randomisation, on average, early nutritional support via the parenteral route had higher intervention and morbidity costs, similar QALYs and a negative INB than the enteral route. Cost-effectiveness results for the pre-specified subgroups were similar to the overall results and the sensitivity analyses indicated that the conclusions were robust to alternative assumptions to those made in the base-case analysis. The lifetime analysis indicated that early nutritional support via the parenteral route had higher mean lifetime QALY at higher additional mean costs, leading to a positive INB but with a wide 95% CI that included zero.

Interpretation

The CALORIES trial was conducted in a sample of adult general critical care units in the NHS in England that had pre-existing, established protocols for nutritional support, prevention of infection and for glycaemic control – reflecting good mainstream practice. The characteristics of participating units were broadly representative (with a slight preponderance of larger units located in university hospitals unlikely to jeopardise the generalisability of our findings) and, as a pragmatic effectiveness study, probably represents the reality of current, NHS nutritional practice in critical care.

There is debate not only about the route, but also about the timing, dose, duration, delivery (continuous vs. intermittent) and type of nutritional support for critically ill patients. The aim of the CALORIES trial was to address solely the question about the optimal route. Our pragmatic trial had two major findings. First, there was no significant difference in outcomes between the two groups, other than the perhaps unsurprising increase in the incidence of vomiting and hypoglycaemia in patients receiving early nutritional support via the enteral route. Specifically, we observed neither the trend towards decreased mortality nor the increase in infectious complications previously reported for patients fed via the parenteral route. 2–4 It is possible that the lack of an infectious burden from feeding via the parenteral route is because of improvements in central venous catheter management (CALORIES data indicate a low incidence of catheter-related and bloodstream infections);60 delivery and composition of feed; and avoidance of overfeeding and hyperglycaemia.

Second, there was only a very small difference in the energy delivered. In both groups, a majority of patients did not reach caloric targets. Although enteral feeding is commonly associated with a failure to reach nutritional targets,7,8 there is a widespread assumption that the parenteral route should be more reliable in guaranteeing delivery. 61,62 Although we do not know exactly why target was not achieved in those fed via the parenteral route, we do know that participation in the CALORIES trial required units to deliver nutritional support via the parenteral route to more patients than would be their usual practice. Other possible factors include the lack of availability of feed out-of-hours (nights/weekends); the use of commercially available products with fixed energy content rather than individualised feeds (with selection of fixed energy products to under- rather than overprovide energy); interruptions in delivery to allow other critical care-related procedures to occur; and transfer out of the critical care unit for other procedures. Another contributory factor, derived from adherence data from the CALORIES trial, indicates that there was a reluctance to continue delivery via the parenteral route towards the end of the 5-day (120 hours) intervention period if this necessitated insertion of a new central venous catheter and/or commencement of a new bag of feed. Finally, there may have been clinical preference for a lower dose. The similar energy intake between the groups, however, reinforced the evaluation and interpretation of our findings on the route of delivery, unconfounded by dose.

Although there was a low probability of early nutritional support via the parenteral route being cost-effective at the primary end point of 1 year, extrapolation to the lifetime resulted in positive INB. At 1 year, the parenteral route group had higher mean survival than the enteral route group, but survival difference between the groups was not statistically significant. The lifetime analysis allowed for the non-significant gains observed in survival at 1 year but did not assume that any gains in mortality were maintained after 1 year. The projected lifetime results indicated QALY gain and higher net monetary benefits for the parenteral group compared with the enteral route group. However, considerable uncertainty surrounded the lifetime cost-effectiveness results. The results of the subgroup analysis suggested that the point estimate of the INB was positive for some subgroups, negative for others, but that the CIs around each of these estimates were wide and included zero. In interpreting these findings, it should be recognised that this study was not powered to detect whether there were subgroup by treatment interactions for either the clinical effectiveness or cost-effectiveness end points, and hence the subgroup results should be regarded as exploratory.

Strengths and limitations

The CALORIES trial is the largest trial addressing the question of the optimal route of nutritional support in critical care. In comparison, of trials included in the Canadian Clinical Practice Guidelines Committee’s meta-analysis of the use of EN versus PN, no trial published prior to the year 2000 contained more than 100 critically ill patients, and those published since 2000 contribute a total of 207 critically ill patients. 63 The CALORIES trial was rigorously conducted, with randomised treatment groups that were well balanced at baseline and with early initiation of nutritional support, as intended. Protocol compliance was high and loss to follow-up was extremely low.

Our understanding of the consequences of critical illness is much greater than when we designed this trial back in 2007. The primary outcome measure of mortality, although objective and accurate, does not recognise the other health states, in particular the consequences of muscle wasting and fatigue, experienced by many survivors of critical illness. It may also have not been sufficiently sensitive to detect meaningful differences between the groups.

Blinding of nutritional support was deemed to be impractical and, although the primary outcome was objective, some of the secondary outcomes, although defined and objectively assessed, may have been more vulnerable to observer bias. Given the very large number of participating critical care units and investigators, it seems improbable that any resulting bias could have been systematic. A relatively large number of secondary outcomes was evaluated and no formal statistical approach was taken to control for the multiple analyses; caution should therefore be taken in interpreting statistically significant results on the secondary outcomes.

Although the trial was not designed specifically to recruit malnourished patients, the small number of these patients recruited limits the relevant subgroup analysis, thus we cannot rule out a difference in outcome according to route for this group of patients. The results of the CALORIES trial should not be generalised to other types of units, other patient groups (from those studied) or other timing, dose (target) or duration of nutritional support. In addition, a large proportion of eligible but not randomised patients (28%) were excluded by the clinician, which may limit the generalisability of the results.

The CALORIES trial included the detailed measurement of resource use within a prospectively designed cost-effectiveness analysis with the collection of resource use data from index and readmissions beyond the trial intervention period. Three sources (trial case report forms, linked data from the Case Mix Programme and responses to Health Service Questionnaires) provided detailed resource use measurement for those events that were anticipated to be the key drivers of the incremental costs of early nutritional support via the parenteral and enteral routes. Costs of the interventions were calculated to represent routine NHS critical care practice and extensive sensitivity analyses were performed. Unlike previous cost-effectiveness analyses, our study has made a direct comparison of early nutritional support via the parenteral route compared with the enteral route, based on data from a large trial with projected lifetime cost-effectiveness results using appropriate methods. The study has assessed quality of life at two time points allowing comparison of changes over time and comparison with the age–gender-matched general population. Quality of life was measured with the EQ-5D-5L; this version of the instrument was anticipated to be sensitive to differences in health status between the treatment groups. Nutritional quality of life was measured by Satisfaction with Food-related Life Questionnaire; it ended up being poorly completed and thus limited the reliability of the results. To address missing data, we undertook a recommended approach for multiple imputation and imputed missing values, conditional on all the information observed.

Inevitably, assumptions – in particular, about mortality, quality of life and costs – were required to be made beyond the observed data. The cost-effectiveness analysis made maximum use of available data from the CALORIES trial and followed a recommended approach to inform these assumptions. These, and other requisite structural assumptions, were made transparent and were subjected to extensive sensitivity analyses. The cost-effectiveness analysis presented results for the same pre-specified subgroups as for the clinical effectiveness analysis.

Results in context

How do our findings compare with those of other recent trials on nutritional support in the critically ill?

In the Early Parenteral Nutrition Completing Enteral Nutrition in Adult Critically Ill Patients (EPaNIC) trial,61 which was conducted in two hospitals (involving seven different ICUs) with patients recruited to receive early or late parenteral supplementation of enteral feeding (if the enteral route alone was not meeting their nutritional target). The investigators found an association between supplemental PN delivered within 48 hours after admission (compared with supplemental PN delayed until after 8 days) and an increased number of new infectious episodes and days of mechanical ventilation. These differences were found both for the large subgroup of cardiac surgical patients and for other critically ill patients. However, the need of many patients for nutritional support, the high target energy intake and the practice of using tight glycaemic control have all been questioned, and make the generalisability of these findings to NHS critical care potentially challenging. Post hoc analysis suggested a dose–response relationship between an increased amount of parenteral supplementation and an increased rate of infectious episodes. 64 Despite important differences between the CALORIES and the EPaNIC trials (research question evaluated, patients studied, nutrition and other care practices), our results potentially support the hypothesis that among patients receiving early PN/supplementation, the dose administered may be more associated with harm than the route of delivery.

In a trial conducted at two ICUs, Heidegger et al. 62 found no difference in the rate of infection between day 8 and day 28 among patients receiving individually optimised PN to supplement inadequate enteral intake on day 4 and patients receiving EN only. In a trial conducted at 31 ICUs, Doig et al. 65 studied patients with relative contraindications to early enteral feeding and found no differences in 60-day mortality or the incidence of infection but fewer days of mechanical ventilation in patients receiving early PN than those with standard care. However, in the standard care group, 27% received early PN and 41% received no nutritional support.

There are two major and contradictory perspectives when it comes to how much to feed critically ill patients in the early phase. One perspective maintains that overfeeding is potentially harmful,61,66 a second that underfeeding is potentially harmful67–69 and a third that there is no difference between underfeeding and standard feeding. 70–72 Still others argue that any effect of nutritional support is likely to be seen only in selected patients who are at greater nutritional risk as a result either of pre-existing malnutrition, obesity or of the nature of their presenting illness. In the CALORIES trial, in both groups, the amount of nutrition delivered was below target but similar to that seen in previous studies in which nutritional targets were also commonly not met. 67,73,74 This suggests that there are substantial practical and organisational obstacles for both routes of feeding. Other research, more recently, has suggested that it is adequacy of protein intake, rather than simply energy intake, which requires to be supported in critical illness. 70

Previous economic analyses report cost savings with the use of the enteral route, rather than the parenteral route, in critically ill patients. 3,75,76 However, these results need to be interpreted with caution, as no incremental cost-effectiveness results were provided. A few economic analyses performed only cost-minimisation analysis because there were no differences in outcome between the parenteral and enteral routes. These studies have suggested that the parenteral route may significantly reduce total costs of hospital care. 65,77 A few other studies have performed economic analysis of nutritional support in critically ill patients but the treatment comparators were different from those in the CALORIES trial. For example, use of early PN compared with late PN in critically ill patients was associated with higher costs and no additional clinical benefit. 78 Early nutrition compared with standard enteral nutrition was found cost-effective for patients admitted to ICUs. 79

Compared with previous economic analyses, our analysis undertook an integrated, full economic evaluation to provide a direct comparison of early nutritional support via the parenteral route compared with the enteral route and extrapolated to lifetime cost-effectiveness results. A key advantage of the integrated economic evaluation, undertaken as part of the CALORIES trial, is that individual-level data on quality of life and resource use were collected prospectively. The quality-of-life data were collected at 90 days and at 1 year post-randomisation with the EQ-5D-5L version. 38 Hence the cost-effectiveness analysis was able to incorporate any quality-of-life differences between the treatment groups into the final measures of cost-effectiveness. The quality-of-life results also showed that, for both treatment groups and time points, patients’ average quality of life (which was between 0.65 and 0.68) was substantially lower than that for the age–gender-matched general population (approximately 0.81)59 and similar to previous estimates for general ICU survivors. 80 In the CALORIES trial, at 1 year post-randomisation, about 30–40% of responders reported ‘severe’ or ‘extreme’ problems with mobility and/or undertaking usual activities indicating substantial ongoing morbidity for this patient group.

The findings of our study support previous analyses that nutritional support via the parenteral route is more costly. The results of Doig et al.,65 that early PN is cost saving, are in contrast with both our integrated economic analysis and another recent costing analysis, by Vanderheyden et al. ,78 reporting increased treatment costs attributable to PN. However, clinical indication and population groups are different across these studies. The CALORIES trial compared costs and QALYs in critically ill adults for whom either route was indicated. Doig et al. 65 addressed the financial consequences of administering PN to patients who were unable to receive early enteral nutrition because of short-term relative contraindications and Vanderheyden et al. 78 assessed costs of early PN compared with late PN to critically ill patients who were able to receive enteral nutrition.

In drawing any comparisons between nutrition studies, it must always be noted that the CALORIES trial asked a different research question in a different population of critically ill patients to other studies. the CALORIES trial does, however, suggest that modern, early nutritional support via the parenteral route, as typically utilised in critical care units in the NHS, is neither more harmful nor more beneficial than via the enteral route and is unlikely to be cost-effective.

Implications for health care

Providing nutritional support to critically ill patients (who are typically unable to eat and therefore require artificial feeding) is an accepted fundamental element of providing critical care.

Recommendations for research

Research staff at participating sites

We acknowledged that there have been many other individuals who made a contribution within the participating units. It is impossible to thank everyone personally; however, we would like to thank the following research staff:

Addenbrooke’s Hospital (Jacobus Preller, Petra Polgarova); Blackpool Victoria Hospital (Jason Cupitt, Sally Baddeley); Bradford Royal Infirmary (Stephen Fletcher, Martin Northey, Linda Kenny); Bristol Royal Infirmary (Jeremy Bewley, Kathryn Pollock, Katie Sweet); Dorset County Hospital (Andy Ball, Sarah Moreton, Stephanie Jones); Furness General Hospital (Andrzej Szymczakowski, Carol McArthur, Kimberley Pallister); Kettering General Hospital (Philip Watt, Parizade Raymode, Nigel Dunk); King’s College Hospital (Philip Hopkins, Daniel Hadfield, Fiona Wade-Smith); Medway Maritime Hospital (Paul Hayden, Lucy Cooper, Claire Pegg); Musgrove Park Hospital (Richard Innes, Pippa Richards, Angela Locke, Dawn Bayford); New Cross Hospital (Shameer Gopal, Jagtar Singh Pooni, Hazel Spencer); Norfolk and Norwich University Hospital (Steve Hutchinson, Melissa Rosbergen, Georgina Glister); Northern General Hospital (Sarah Irving, Rachel Walker, Angela Pinder); Papworth Hospital (Stephen Webb, Fiona Bottrill); Poole Hospital (Henrik Reschreiter, Julie Camsooksai, Helena Barcraft-Barnes); Queen Alexandra Hospital (David Pogson, Steve Rose, Nicola Lamb, Johanna Mouland); Royal Free Hospital (Daniel Martin, Paula Meale, Sarah James); Royal Hampshire County Hospital (Arthur Goldsmith, Dawn Trodd, Jane Martin); Royal Shrewsbury Hospital (Robin Hollands, Mandy Carnahan); Royal Surrey County Hospital (Ben Creagh-Brown, Justin Kirk-Bayley, Peter Carvalho); Southampton General Hospital (Susan Tanser, Clare Bolger, Kim Golder, Karen Salmon); Southend University Hospital (David Higgins, Sarah Martin); St Mary’s Hospital, London (Richard Leonard, Verena Hauer, Adaeze Ochelli-Okpue); St Richard’s Hospital (Mike Margarson, Yolanda Baird, Justin Dickens); St Thomas’ Hospital (Richard Beale, Katie Lei, John Smith); Stafford Hospital (Moses Chikungwa, Garud Chandan, Clare Jackson); The Ipswich Hospital (Robert Lewis, Stephanie Bell, Heather Blaylock); The Princess Alexandra Hospital (Dev Dutta, Jackie Power, Carol Keel); The Queen Elizabeth Hospital, King’s Lynn (Parvez Moondi, Kate Wong, Katharine Draper); The Royal Blackburn Hospital (Anton Krige, Lynne Bullock, Donna Harrison-Briggs, Martin Bland); University College Hospital (David Brealey, Jung Ryu, Georgia Bercades); University Hospital Lewisham (Marthin Mostert, Clair Harris, David Grannell); and University Hospital of North Staffordshire (Bryan Carr, Ruth Salt, Claire Matthews).

Contributions of authors

Dr Sheila E Harvey (CTU Manager and Senior Research Fellow) contributed to the design of the trial and the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Francesca Parrott (Statistician) contributed to the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr David A Harrison (Senior Statistician and Honorary Senior Lecturer in Medical Statistics) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr M Zia Sadique (Lecturer in Health Economics) contributed to the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Professor Richard D Grieve (Professor of Health Economics Methodology) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Ruth R Canter (Research Assistant) assisted in the management of the trial, contributed to the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Blair KP McLennan (Trial Manager) managed the trial, contributed to the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Jermaine CK Tan (Trials Data Manager) contributed to the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Ms Danielle E Bear (Principal Critical Care Dietitian) contributed to the design of the trial and the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Ms Ella Segaran (Advanced Dietitian for Critical Care) contributed to the design of the trial and the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr Richard Beale (Clinical Director/Consultant, Intensive Care Unit) contributed to the design of the trial and the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Professor Geoff Bellingan (Divisional Clinical Director/Consultant, Critical Care and Theatres) contributed to the design of the trial and the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr Richard Leonard (Clinical Director/Consultant, Intensive Care Unit) contributed to the design of the trial and the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Professor Michael G Mythen (Professor of Anaesthesia & Critical Care) contributed to the design of the trial and the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Professor Kathryn M Rowan (Director of Scientific & Strategic Development/CTU Director and Honorary Professor) conceived and designed the trial, contributed to acquisition, analysis and interpretation of the data, and drafted and critically revised the manuscript.

Publications

Harvey SE, Parrott F, Harrison DA, Mythen M, Rowan KM. The CALORIES trial: statistical analysis plan. Crit Care Resusc 2014;16:248–54.

Harvey SE, Parrott F, Harrison DA, et al. Trial of the route of early nutritional support in critically ill adults. N Engl J Med 2014;371:1673–84.

Trial Management Team

Kimberley Anderson (previous Research Assistant); Ruth Canter (Research Assistant); Louise Clement (previous Trial Manager); Sarah Corlett (previous Trial Administrator); Dr David Harrison (Senior Statistician); Dr Sheila Harvey (CTU Manager/Senior Research Fellow); Blair McLennan (Trial Manager); Hannah Muskett (previous Assistant Trial Manager); Francesca Parrott (Trial Statistician); Krishna Patel (previous Trial Statistician); Professor Kathryn Rowan (Chief Investigator); Dr Rachael Scott (previous Senior Trial Manager); and Jermaine Tan (Data Manager).

Trial Management Group

Professor Richard Beale (Co-Investigator); Danielle Bear (Trial Dietitian); Geoff Bellingan (Co-Investigator); Dr David Harrison (Senior Statistician); Dr Sheila Harvey (CTU Manager/Senior Research Fellow); Dr Richard Leonard (Co-Investigator); Professor Michael Mythen (Lead Clinical Investigator); Professor Kathryn Rowan (Chief Investigator); and Ella Segaran (Trial Dietitian).

Trial Steering Committee

Professor Michael Stroud (independent, chairperson); Professor Peter Emery (independent); Professor Alastair Forbes (Co-Investigator); Peter Gibb (independent, patient representative); Dr George Grimble (Co-Investigator, observer); Carys Jones (independent); Professor Hugh Montgomery (Co-Investigator); Janet Myers (previous independent, patient representative); Dr Dewi Williams (independent); Professor Michael Mythen (Lead Clinical Investigator); and Professor Kathryn Rowan (Chief Investigator).

Data Monitoring and Ethics Committee

Dr Elizabeth Allen (chairperson), Professor Peter Andrews and Professor Steve Webb.

Data sharing statement

Data can be obtained from the corresponding author.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health.

Acute Physiology and Chronic Health Evaluation version II

The APACHE II Acute Physiology score consists of weightings for 12 physiological parameters to give a total score ranging from 0 to 60, with higher scores indicating greater severity of illness. 34 The 12 physiological parameters are as follows:

• temperature

• mean arterial pressure

• heart rate

• respiratory rate

• alveolar–arterial gradient (if FiO₂ ≥ 0.5) or PaO₂ (if FiO₂ < 0.5)

• arterial pH (or serum bicarbonate if no arterial blood gas recorded)

• serum sodium

• serum potassium

• serum creatinine (with double weighting for acute renal failure)

• haematocrit (estimated from haemoglobin)

• white blood cell count, and

• GCS score (assumed to be normal for patients sedated or paralysed).

The APACHE II Score comprises the Acute Physiology score plus additional weightings for age and severe comorbidities in the past medical history to give a total score ranging from 0 to 71. Severe comorbidities must have been present and documented in the past medical history within the 6 months prior to presentation at hospital and are defined as follows:

• Severe liver condition – presence of portal hypertension, biopsy-proven cirrhosis or hepatic encephalopathy.

• Severe cardiovascular condition – presence of fatigue, claudication, dyspnoea or angina at rest (New York Heart Association Functional Class IV).

• Severe respiratory condition – presence of permanent shortness of breath with light activity as a result of pulmonary disease, or on home ventilation.

• Severe renal condition – receiving chronic renal replacement therapy (haemodialysis, haemofiltration or peritoneal dialysis) for irreversible end-stage renal disease.

• Immunological condition – receiving chemotherapy, radiotherapy or daily high-dose steroid treatment (0.3 mg/kg, or greater, prednisolone or equivalent) for 6 months, or diagnosis of human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS), lymphoma, acute or chronic myelogenous/lymphocytic leukaemia, multiple myeloma or active metastatic disease.

The APACHE II predicted risk of death combines the APACHE II Score with additional weightings for admission following emergency surgery and for diagnostic categories from the primary reason for admission to the critical care unit to estimate the predicted risk of death before ultimate discharge from an acute hospital.

The APACHE II severity of illness scores and predicted risk of mortality were calculated from raw data using standardised computer algorithms. Severity of illness scores were based on the most extreme (highest or lowest) values of physiological parameters recorded during the first 24 hours following admission to the critical care unit. Coefficients for the APACHE II risk prediction model were taken from the most recent (2013) UK recalibration of the model, based on 242,450 admissions to 207 UK ICUs participating in the Case Mix Programme between 1 January 2011 and 31 December 2012.

Intensive Care National Audit & Research Centre model

The ICNARC Physiology Score consists of objectively defined weightings for 12 physiological parameters to give a total score ranging from 0 to 100, with higher scores indicating greater severity of illness. 35 The 12 physiological parameters are as follows:

• heart rate

• systolic blood pressure

• temperature

• respiratory rate

• PaO₂/FiO₂ ratio (weighted differently, depending on whether the patient was ventilated at any time during the first 24 hours in the unit, or the entire stay if < 24 hours)

• arterial pH

• serum urea

• serum creatinine

• serum sodium

• urine output

• white blood cell count, and

• GCS score (plus additional weightings for patients sedated or paralysed and sedated for the whole of the first 24 hours in the unit, or the entire stay if < 24 hours).

The ICNARC model predicted risk of death combines the ICNARC Physiology Score with additional weightings for age, cardiopulmonary resuscitation within 24 hours prior to admission to the critical care unit, location prior to admission to the critical care unit, urgency of surgery (for admissions from theatre), primary reason for admission to the critical care unit, and interactions between the Physiology Score and primary reason for admission to estimate the predicted risk of death before ultimate discharge from an acute hospital.

The ICNARC Physiology Score and predicted risk of mortality were calculated from raw data using standardised computer algorithms. The Physiology Score was based on the most extreme (highest or lowest) values of physiological parameters recorded during the first 24 hours following admission to the critical care unit. Coefficients for the ICNARC risk prediction model were taken from the most recent (2013) UK recalibration of the model based on 242,450 admissions to 207 UK ICUs participating in the Case Mix Programme between 1 January 2011 and 31 December 2012.

Sequential Organ Failure Assessment

The SOFA score consists of weightings for six organ systems to give a total score ranging from 0 to 24, with higher scores indicating a greater degree of organ failure. 33 The organ failure assessments are as follows:

• respiratory dysfunction, based on lowest PaO₂/FiO₂

• cardiovascular dysfunction, based on vasopressor use and lowest mean arterial pressure

• renal dysfunction, based on highest creatinine

• neurological dysfunction, based on lowest (or last pre-sedation) GCS score

• hepatic dysfunction, based on highest bilirubin, and

• coagulation dysfunction, based on lowest platelet count.

The SOFA score was calculated from raw physiology and treatment data from the 24 hours prior to randomisation.

Definitions

Duration of organ support in the critical care unit was defined as the number of days alive and free from support of each of the following organ systems, as defined by the UK Department of Health CCMDS during the first 30 days following randomisation. 36 Patients who died within the first 30 days were assigned zero days alive and free from organ support. Organ support definitions were as follows:

• Advanced respiratory – indicated by one or more of invasive mechanical ventilatory support through a translaryngeal tube or tracheostomy; bilevel positive airway pressure through a translaryngeal tube or tracheostomy; continuous positive airway pressure through a translaryngeal tube; or extracorporeal respiratory support.

• Advanced cardiovascular – indicated by one or more of receipt of multiple intravenous and/or rhythm controlling drugs (of which at least one must be vasoactive) when used simultaneously to support or control arterial pressure, cardiac output or organ/tissue perfusion; continuous observation of cardiac output and derived indices; an intra-aortic balloon pump or other assist device; or temporary cardiac pacemaker.

• Renal – indicated by receipt of acute renal replacement therapy (e.g. haemodialysis, hemofiltration, etc.); or receipt of renal replacement therapy for chronic renal failure when other acute organ support is received.

• Neurological – indicated by one or more of central nervous system depression that was sufficient to prejudice the airway and protective reflexes (except when caused by sedation prescribed to facilitate mechanical ventilation or by poisoning, e.g. deliberate or accidental self-administered overdose, alcohol, drugs, etc.); receipt of invasive neurological monitoring or treatment (e.g. intracranial pressure monitoring, jugular bulb sampling, external ventricular drain, etc.); receipt of continuous intravenous medication to control seizures and/or for continuous cerebral monitoring; or receipt of therapeutic hypothermia using cooling protocols or devices.

• Gastrointestinal – indicated by receipt of PN or EN (i.e. any method of feeding other than normal oral intake).