Medically assisted nutrition for palliative care patients
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
The objectives of this review are to determine the effect of medically assisted nutrition in palliative care patients on their quality and length of life.
Background
Many palliative care patients have a reduced oral intake during their illness. The cause of this varies, but may be part of a physical obstruction, anorexia / cachexia syndrome, generalised weakness, bowel obstruction, loss of desire to drink or no specific cause may be identified. The most common time for this decreased oral intake is during the terminal phase, when the patient becomes less conscious and therefore less able to receive nutrition orally (Morita 1998).
Management of this condition includes discussion with the patient, family and staff involved and either no medical intervention or the provision of nutrition with medical assistance. The aim of this intervention can be to prolong the length of life of a patient, improve their quality of life, or both. These benefits may come via the reversal of the physiological factors associated with the patient's decline. Balanced against these potential benefits are adverse events that can be associated with any intervention (infection, bleeding, pain etc) (Bozzetti 1996).
Medically assisted nutrition can be performed via a tube inserted into any part of the gastrointestinal system (enteral) or via a tube inserted into the venous system (parenteral). There is some controversy and varying views on the ethics of medically assisted nutrition (Casarett 2005). This review will concentrate on assessing the benefit of provision of nutrition with medical assistance versus the harm caused by such intervention in palliative care patients. It is only with this information that clinicians and patients can make informed decisions about whether this type of intervention is beneficial or harmful to an individual patient.
There will be a separate review conducted looking at the provision of medically assisted hydration for palliative care patients (Good 2006).
Objectives
The objectives of this review are to determine the effect of medically assisted nutrition in palliative care patients on their quality and length of life.
Methods
Criteria for considering studies for this review
Types of studies
All relevant randomised controlled studies (RCTs) or prospective controlled studies (if no RCTs are found).
Types of participants
Types of participants will be palliative care patients who receive medically assisted nutrition. Participants in the review will be those patients that are receiving palliative care (WHO 2005). These patients will include (but not be limited to) incurable cancer, dementia, neurodegenerative diseases (e.g. Motor Neuron Disease), Human Immunodeficiency Virus, Chronic Airways Limitation and Chronic Heart Failure.
Patients who are having medically assisted nutrition as part of a perioperative, chemotherapy or radiotherapy regime, or because of chemotherapy or radiotherapy adverse effects will be excluded.
Types of interventions
Medically assisted administration of nutrition :
-
parenteral nutrition ‐ administration of nutritional liquid via a central or peripheral venous catheter, that does not enter the gastrointestinal system;
-
enteral nutrition ‐ administration of nutritional liquid through a tube via the gastrointestinal system (nasogastric tube, jejunostomy, gastrostomy).
Comparisons:
-
placebo,
-
no Intervention,
-
usual treatment or supportive care.
Types of outcome measures
Primary
1. Quality of life on any measure (including symptom assessment scales)
Secondary
1. Survival
2. Adverse Events
Search methods for identification of studies
A. Electronic Databases
The following electronic databases will be searched using a search strategy developed for MEDLINE, but modified appropriately for each database.
1.The Cochrane Library: Cochrane Pain, Palliative & Supportive Care Register, Cochrane Controlled Trials Register, Cochrane Database of Systematic Reviews, Cochrane Database of Reviews of Effectiveness.
2. MEDLINE (1966 to present)
3. EMBASE (1980 to present)
4. LILACS (1992 to present)
5. SIGLE (1995 to present)
6. CINAHL
7. CANCERLIT
8. Caresearch ‐ database listing conference proceedings and grey literature
9. Dissertation abstracts
10. SCIENCE CITATION INDEX
B. Reference Lists
The reference lists of all eligible trials, key textbooks, and previous systematic reviews will be searched for additional studies.
C. Language
The search will attempt to identify all relevant studies irrespective of language. Non‐English papers will be assessed and, if necessary, translated, with the assistance of a native speaker.
The subject search will use a combination of controlled vocabulary and free text terms based on the following search strategy for searching MEDLINE:
#1 MeSH descriptor PALLIATIVE CARE explode all trees
#2 palliat* in All Text
#3 MeSH descriptor TERMINALLY ILL this term only
#4 MeSH descriptor TERMINAL CARE explode all trees
#5 (terminal* in All Text near/6 care* in All Text)
#6 ( (terminal* in All Text near/6 ill* in All Text) or terminal‐stage* in All Text or dying in All Text or (close in All Text near/6 death in All Text) )
#7 (terminal* in All Text near/6 diseas* in All Text)
#8 (end in All Text near/3 life in All Text)
#9 hospice* in All Text
#10 (end‐stage next disease* in All Text or end next stage next disease* in All Text or end‐stage next illness in All Text or end next stage next illness in All Text or end‐stage next care in All Text or end next stage next care in All Text)
#11 incurable next illness* in All Text
#12 incurable next disease* in All Text
#13 (advanced next directive* in All Text or living next will* in All Text or do‐not‐resuscitate next order* in All Text)
#14 (end‐stage next disease* in All Text or end next stage next disease* in All Text or end‐stage next illness in All Text or end next stage next illness in All Text or end‐stage next care in All Text or end next stage next care in All Text)
#15 (advanced in All Text near/6 disease* in All Text)
#16 (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15)
#17 MeSH descriptor NUTRITION explode all trees
#18 MeSH descriptor NUTRITION ASSESSMENT explode all trees
#19 MeSH descriptor NUTRITION THERAPY explode all trees
#20 MeSH descriptor FEEDING METHODS explode all trees
#21 (feed in All Text or feeding in All Text or fed* in All Text or food* in All Text)
#22 MeSH descriptor FOOD explode all trees
#23 diet* in All Text
#24 nutrition* in Record Title
#25 nutrition* in Abstract
#26 (#17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25)
#27 (#16 and #26)
This search strategy will be adapted for other databases searched.
Data collection and analysis
Studies identified by the search strategy will have the title and abstract (where possible) assessed by the lead review author (PG) to identify potentially relevant articles. These articles will then be retrieved in full. The full text of potentially relevant articles will be assessed independently by two review authors (PG & MM or PG & PR) with regards to inclusion criteria and quality. Where there is disagreement, this will be resolved via discussion and involvement of a third review author (JC).
Quality
All trials will have their methodological quality assessed. There will be two scales used.
1. Randomised controlled trials will be assessed via the Oxford Quality Scale devised by Jadad et al (Jadad 1996).
This scale uses the following questions to rate the likelihood of bias (the higher the score the less likelihood of bias, scale of zero to five).
a. Was the study described as randomised (one = yes; zero = no)?;
b. Was the study described as double‐blind (one = yes; zero = no)?;
c. Was there a description of withdrawals and dropouts (one = yes; zero = no)?;
d. Was the method of randomisation well described and appropriate (one = yes; zero = no)?;
e. Was the method of double‐blinding well described and appropriate (one = yes; zero = no)?;
f. Deduct one point each if methods for randomisation and blinding were inappropriate.
Scoring system: maximum score = five; minimum score = zero.
2. The quality of non RCTs will be assessed using a scale devised by Rinck et al (Rinck 1997).
Data extraction
The following information will be obtained for each study:
-
study methods (study design, allocation, blinding, setting, inclusion criteria);
-
participants (sample size, exclusions / inclusions, number, disease, duration of trial, withdrawals and dropouts, site ‐ e.g. hospital, hospice, home);
-
intervention (type, route of delivery, control used);
-
outcome (quality of life, symptom measures, survival, time from death intervention was initiated);
-
adverse Effects.
The extraction will occur independently by two review authors (PG & MM or PG & PR).
Data analysis
The overall effectiveness of medically assisted nutrition in palliative care patients will be assessed and also specific sub‐group analysis (where possible) will be undertaken by:
-
study design:
data from randomised controlled trials and prospective controlled studies will be evaluated separately
-
participants:
‐ cancer
‐ non‐cancer
‐ dementia
‐ neurodegenerative diseases
-
intervention:
‐ medically assisted nutrition ‐ parenteral, enteral nutrition
-
study quality
-
timing of intervention (in relation to death)
-
site
Statistical analysis
Where RCTs are clinically and statistically homogenous, and if there are sufficient studies that use comparable outcome measures, meta‐analyses will be undertaken (using RevMan Analyses in RevMan 4.2). Data will be pooled using the random effect model but the fixed effect model will also be analysed to ensure robustness of the model chosen and susceptibility to outliers. If appropriate data are available, continuous outcomes will be analysed using weighted mean differences (WMDs) with 95% confidence intervals (CIs). Relative risks (RRs) with 95% confidence intervals will be calculated for dichotomous data. Number needed to harm (NNH) with corresponding 95% CIs will be calculated for adverse effects. If data cannot be pooled, we will write a description of the main findings from the non‐randomised studies.
