Anda di halaman 1dari 9

European Journal of Clinical Nutrition (2003) 57, 895903

& 2003 Nature Publishing Group All rights reserved 0954-3007/03 $25.00
www.nature.com/ejcn

ORIGINAL COMMUNICATION
A systematic review of protein and energy
supplementation for hip fracture aftercare
in older people
A Avenell1* and HHG Handoll2
1
Health Services Research Unit, University of Aberdeen, Aberdeen, UK; and 2School of Health and Social Care,
University of Teesside, based at the Department of Orthopaedic Surgery, Royal Infirmary of Edinburgh,
Little France, Edinburgh, UK

Objectives: To evaluate whether protein and energy supplementation influences recovery after hip fracture.
Design: Systematic review of randomised and quasi-randomised trials in people aged 65 y and over.
Data sources: We searched seven electronic databases from 1966 to April 2002, four journals and reference lists of relevant
articles. We contacted trial investigators and experts for details of other trials.
Main outcome measures: Mortality, complications and unfavourable outcome (mortality or survivors with complications) were
the primary outcomes. We also sought data on length of hospital stay, functional status after hip fracture, quality of life and
compliance with supplementation.
Results: In total, 12 randomised trials involving 898 participants were included. Nine trials evaluated protein and energy
supplementation (five oral and four nasogastric feeding), and a further three trials tested oral protein supplementation. Potential
biases resulting from inadequate allocation concealment and lack of assessor blinding and intention-to-treat analysis, as well as
the limited outcome data, mean that the results must be interpreted with caution.
Pooled data from eight of the nine trials evaluating protein and energy supplements showed no evidence for an effect on
mortality (relative risk 0.92, 95% CI 0.561.50). Limited data from only three trials showed that oral protein and energy
supplements may reduce unfavourable outcome (relative risk 0.52, 95% CI 0.320.84).
Conclusion: Based on limited evidence, oral protein and energy supplementation after hip fracture may reduce unfavourable
outcome. Further evidence from good-quality randomised trials is required to inform clinical practice.
Sponsorship: Medical Research Council, Chief Scientist Office of the Scottish Executive Health Department.
European Journal of Clinical Nutrition (2003) 57, 895903. doi:10.1038/sj.ejcn.1601623
Keywords: hip fracture; nutritional support; aged; meta-analysis

*Correspondence: A Avenell, Health Services Research Unit, University of


Aberdeen, Foresterhill Lea, Foresterhill, Aberdeen AB25 2ZD, UK.
E-mail: a.avenell@abdn.ac.uk
Guarantors: A Avenell and H Handoll.
Contributors: AA initiated the review, wrote the first draft of the
protocol, undertook the subject-specific literature search, contacted
trialists and wrote the first draft of the review. Both authors assessed
and extracted data from trials and devised the analyses. HH provided
methodological support at all stages and critically rewrote the review
and subsequent drafts.
Potential conflict of interest: none known.
Supported by the Medical Research Council (AA); Chief Scientist
Office of the Scottish Executive Health Department (AA, HHGH). The
Health Services Research Unit is funded by the Chief Scientist Office
of the Scottish Executive Health Department; however, the views
expressed are those of the authors.
Received 27 May 2002; revised 22 July 2002;
accepted 13 August 2002

Introduction
Mortality 1 y after hip fracture varies from 14 to 37% (Lyons,
1997). Acute hospital costs after hip fracture are substantial,
and costs of rehabilitation and care in the community may
be greater; respectively, estimated at d4800 and d7500 at
1995/6 prices in the UK (Dolan & Torgerson, 1998). Figures
from Belgium covering a similar period showed average costs
of an acute hospital stay at A8667, with extra costs compared
with non-hip fracture controls of A6636 in the year after
hospitalisation (Autier et al, 2000).
People with hip fractures, who are often under-nourished
at the time of the fracture (Mansell et al, 1990; Lumbers et al,
1996; Maffulli et al, 1999), have poor food intakes in hospital
(Bastow et al, 1983a; Jallut et al, 1990; Patterson et al, 1992;

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

896
Lumbers et al, 1999). Under-nutrition leads to mental
apathy, reduced muscle power, and impaired cardiac function and immunity (Keys et al, 1950; Lesourd, 1997); these
increase the tendency to develop postoperative complications and hinder recovery. Prolonged rehabilitation and
increased risk of death after hip fracture are associated with
under-nutrition (Bastow et al, 1983a; Lumbers et al, 1996).
We evaluated whether protein and energy supplementation of people with hip fracture affects the outcome, by
examining the evidence from randomised controlled trials.

Where possible, the numbers of patients with unfavourable


outcome have been presented; this is defined as the sum of
the patients who died plus survivors with complications.
However, we also accepted the slightly different definition of
unfavourable outcome (mortality or survivors with a major
complication or two or more minor complications) used in
three trials (Delmi et al, 1990; Tkatch et al, 1992; Hankins,
1996). In these trials bedsores, pneumonia, pyelonephritis,
severe anaemia, deep vein thrombosis, acute renal failure,
pulmonary embolism and heart failure were defined as
major complications.

Methods
Search strategy
We identified relevant randomised and quasi-randomised
trials in any language (see Table 1 for eligibility details) by
electronic searching of seven databases, hand searching of
four nutrition journals, checking of reference lists of
reported trials and review articles, and contacting trialists
and other experts. Our most recent search was concluded in
April 2002. Full details of our search strategy are available
(Avenell et al, 2001).

Assessment of eligibility and trial quality


Both reviewers independently assessed reports for predefined
eligibility and methodological quality (see Table 1). Differences in opinion were resolved by discussion. We included
published and unpublished trials, whether available in full or
abstract only. Authors of articles were contacted for further
details of their trials.

Outcomes assessed
We sought data on all-cause mortality, numbers of participants with complications (as defined by the investigators),
and lengths of acute hospital and rehabilitation unit stay.

Data extraction and synthesis


Data and trial information were independently extracted by
each of us and, upon resolving any discrepancies, the final
data set was entered into Review Manager version 4.1.
(Oxford, England: The Cochrane Collaboration, 2000). For
each study, relative risks and 99% confidence limits were
calculated for dichotomous outcomes. Where appropriate,
the data from comparable groups of trials were combined
using the fixed effect model, and results presented with 95%
confidence limits. In the meta-analyses, the weight given to
each study was the inverse of its variance, ie more precise
estimates (from larger studies with more events) were given
more weight. Heterogeneity between trials was tested using a
standard w2 test. Subgroup and sensitivity analyses, based on
the feeding route (oral, enteral or parenteral), pre-existing
nutritional status and aspects of methodology, were planned
from the outset and undertaken where appropriate. This
review is derived from a regularly updated review on all
forms of nutritional supplementation after hip fracture
published in The Cochrane Library (Avenell & Handoll,
2001).
Mortality results have been presented using denominators
based on the numbers of participants at randomisation
(intention-to-treat-analysis).

Table 1 Eligibility and quality criteria


Eligibility criteria
Randomised or quasi-randomised controlled trial
Patients over 65 y recovering from hip fracture
Intervention provided in any setting from hospital or at home
Intervention group received protein and/or energy supplement compared with no treatment or placebo; provided by oral, nasogastric or parenteral
route; started within 1 month of hip fracture and continued for up to 1 y
Outcome data provided on at least one of the following: all-cause mortality, participants with complications, length of hospital or rehabilitation unit
stay, postoperative functional status, participants perceived quality of life, putative side effects, participant compliance with intervention
Quality criteria
Treatment adequately concealed prior to allocation
Outcomes of patients who withdrew described and included in the analysis (intention-to-treat analysis)
Outcome assessors blinded to treatment status
Treatment and control group comparable at entry
Care programmes, other than the trial options, identical
Inclusion and exclusion criteria clearly defined
Interventions clearly defined (including estimates of nutritional value)
Active and appropriate follow-up
Duration of surveillance clinically appropriate

European Journal of Clinical Nutrition

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

897

Results
We found 12 eligible studies, involving 898 participants
(Bastow et al, 1983b; Stableforth, 1986; Delmi et al, 1990;
Brown & Seabrook, 1992; Gallagher et al, 1992; Tkatch et al,
1992; Madigan, 1994; Hankins, 1996; Hartgrink et al, 1998;
Schurch et al, 1998; Sullivan et al, 1998; Espaulella et al,
2000). Details of the trials are provided in Table 2. Nine
tested a combination of protein and energy supplementation
(the route was oral in five and by nasogastric tube in four).
Three trials tested oral protein supplementation.
Where reported, the mean age of participants was usually
over 80 y. Many of the studies excluded patients with
medical conditions (dementia, stroke, endocrine disease,
cancer, kidney or liver failure). Six trials failed to provide any
nutritional assessment of trial participants by anthropometry, for example, weight or mid-arm circumference (Delmi
et al, 1990; Gallagher et al, 1992; Tkatch et al, 1992; Hartgrink
et al, 1998; Sullivan et al, 1998; Espaulella et al, 2000).
Four studies specifically examined nutritional supplementation in malnourished participants, each study assessing
nutritional status differently; three used anthropometric
criteria (Bastow et al, 1983b; Brown & Seabrook, 1992;
Hankins, 1996;) and one used serum albumin values
(Gallagher et al, 1992).
The quality of reported trial methodology was disappointing. Concealment of allocation was adequate in only three
trials (Hankins, 1996; Sullivan et al, 1998; Espaulella et al,
2000). Intention-to-treat analysis was carried out in four
studies (Bastow et al, 1983b; Brown & Seabrook, 1992;
Hankins, 1996; Sullivan et al, 1998). In three studies,
participants were excluded after randomisation because of
poor compliance with dietary supplementation (Tkatch et al,
1992; Madigan, 1994; Espaulella et al, 2000). Data on some of
these participants were obtained for two studies (Madigan,
1994; Espaulella et al, 2000).
Only two studies (Brown & Seabrook, 1992; Espaulella et al,
2000) reported blinding of outcome assessors. While being
difficult for nasogastric feeding, blinded assessment should
be possible for other interventions, and after discharge from
hospital. Blinding of both participants and treatment
providers was reported in only one study (Espaulella et al,
2000).
Inclusion and exclusion criteria were well defined in all
trials. Eight studies (Delmi et al, 1990; Tkatch et al, 1992;
Madigan, 1994; Hankins, 1996; Hartgrink et al, 1998;
Schurch et al, 1998; Sullivan et al, 1998; Espaulella et al,
2000) gave clear details of the exact nutritional content, and
of the protocol for application, of the supplement. Comparable care programmes, other than the trial interventions,
were confirmed in three trials (Stableforth, 1986; Hartgrink
et al, 1998; Espaulella et al, 2000).
Follow-up of trial participants was generally active and
appropriate in terms of outcome assessment. However, only
five studies followed participants up for 6 months or more
(Delmi et al, 1990; Tkatch et al, 1992; Schurch et al, 1998;
Sullivan et al, 1998; Espaulella et al, 2000), and no study

reported patients perceived quality of life after discharge.


Economic outcomes were also not reported.
Postoperative complications were reported as a wide
variety of individual conditions (including diarrhoea, nausea, vomiting, aspiration pneumonia, gastrointestinal ulcer,
pressure sore, face flushing, chest infection, urinary tract
infection, deep venous thrombosis, pulmonary embolism,
cardiac failure, anaemia, confusion and acute renal failure)
and generic conditions (gastrointestinal, surgical, infection,
postoperative, life-threatening). The definition of complications was consistent in three trials (Delmi et al, 1990; Tkatch
et al, 1992; Hankins, 1996), but varied in definition and
scope in the rest.
Lengths of hospital stay in acute and rehabilitation
hospitals were often reported, but have not been presented
in this paper. Such data may be misleading for two main
reasons. Firstly, this outcome, while important from the
costs perspective, is a poor indicator of patient health;
administrative procedures are often more influential. Secondly, even when means and standard deviations (s.d.) for
these outcomes have been reported, it is unlikely that
lengths of stay were normally distributed. A comprehensive
account of length of stay results is available elsewhere
(Avenell & Handoll, 2001). Suffice to record here that we
found no consistent pattern in the results that could be
relied upon to draw conclusions of relative effect.

Protein and energy supplementation


Protein and energy supplementation was supplied orally in
five studies (Stableforth, 1986; Delmi et al, 1990; Brown &
Seabrook, 1992; Madigan, 1994; Hankins, 1996) of which
two (Brown & Seabrook, 1992; Hankins, 1996) targeted
malnourished patients. Four studies evaluated nasogastric
supplementation (Bastow et al, 1983b; Gallagher et al, 1992;
Hartgrink et al, 1998; Sullivan et al, 1998) of which two
(Bastow et al, 1983b; Gallagher et al, 1992) targeted
malnourished patients. No data were available for pooling
from one trial (Gallagher et al, 1992) published as an abstract
only.
Overall mortality (Figure 1) was similar in the intervention
and control groups (relative risk 0.92, 95% CI 0.561.50).
There was marginally significant heterogeneity (Po0.1).
There was no evidence of an effect for mortality when split
by feeding route (Figure 1). There was however considerable
heterogeneity (P 0.04) for the mortality results from the
three nasogastric feeding trials. All seven deaths in one trial
(Hartgrink et al, 1998) occurred in the intervention group
during the 2-week period of observation. This could have
been because of chance, as the deaths were not obviously
related to tube feeding or aspiration pneumonia, a complication of tube feeding. Conversely, the five deaths in the
control group in another trial (Sullivan et al, 1998) might in
part reflect the greater frailty of this group at recruitment.
When trials were subgrouped according to nutritional
status, mortality tended to be reduced in the supplemented
European Journal of Clinical Nutrition

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

898
Table 2 Characteristics of included studies

Trial ID

Patients
randomised /
assessed for Intervention Control
mortality
groupa
groupa

Bastow
1983b

122/122

Brown
1992

10/10

Delmi
1990

59/59

Espaulella
2000

171/171

Gallagher
1992

97/not
available

Hankins
1996

32/31

Oral feed,
protein and
energy

Hartgrink
1998

140/140

Madigan
1994

34/30

Schurch
1998

82/82

Nasogastric
feed, protein
and energy
Oral feed,
protein and
energy
Oral feed,
protein and
energy
Oral feed,
protein and
energy
Nasogastric
feed, protein
and energy
Oral feed,
protein and
energy

Stableforth 61/58
1986
Sullivan
1998

18/18

Tkatch
1992

72/62

Nasogastric
feed, protein
and energy
Oral feed,
protein and
energy
Oral feed,
protein and
energy
Oral feed,
protein and
energy
Nasogastric
feed, protein

Review outcomes reported in trials


(Note: Data/results were
unavailable or incomplete for
some of the outcomes listed below)

Duration of
follow-up

Methodological
rating for concealment
of randomisation

Normal
ward diet

Until death or
discharge

Mortality, length of hospital stay, postoperative


functional status, dietary intake, compliance

Normal
ward diet

At least
21 days

Mortality, length of hospital stay,


postoperative functional status, dietary intake

Normal
ward diet

6 months

Mortality, patients with complications,


unfavourable outcome, length of stay in hospital and
rehabilitation, dietary intake, compliance

Oral feed,
energy, no
protein
Normal
ward diet
and energy
Normal
ward diet

6 months

Mortality, unfavourable outcome, length of hospital stay,


postoperative functional status, compliance

At least
21 days

Length of stay in rehabilitation,


postoperative functional status

2 months

Mortality, patients with complications,


unfavourable outcome, length of hospital and
rehabilitation stay, postoperative functional status,
dietary intake, compliance
Mortality, length of hospital stay,
dietary intake, compliance

Normal
ward diet

2 weeks

Normal
ward diet

3 months post- Mortality, length of hospital stay, postoperative


hospital
functional status, dietary intake, compliance
discharge
Oral feed, 12 months
Mortality, length of hospital and rehabilitation
energy, no
stay, postoperative functional status, compliance
protein
Normal
4 weeks
Mortality, patients with complications,
ward diet
unfavourable outcome
Normal
ward diet

6 months

Oral feed, 7 months


energy, no
protein

Mortality, patients with complications, length


of hospital stay, postoperative functional status,
dietary intake
Mortality, unfavourable outcome, length of
hospital and rehabilitation stay, compliance

Methodological rating for concealment of randomisation: A, method did not allow disclosure of assignment; B, small, but possible chance of disclosure of
assignment, or states random but no description given; C, quasi-randomised (alternate allocation to groups).
a
Where nasogastric or oral supplements were given, patients were always allowed the normal ward diet.

groups of those trials that targeted malnourished participants (relative risk 0.61, 95% CI 0.281.34) but not in the
other trials (relative risk 1.20, 95% CI 0.632.28). However, a
test of interaction showed that the difference in the results of
the two groups of trials was not statistically significant
(P 0.2).
Pooled data from four trials for patients with complications tended to favour supplementation (relative risk of 0.68,
95% CI 0.431.06) (Figure 2). However, there was significant
heterogeneity (P 0.05), which reflected the different results
of the only nasogastric trial (Sullivan et al, 1998). Combined
results (Figure 2) from the three oral feeding studies
(Stableforth, 1986; Delmi et al, 1990; Hankins, 1996;)
European Journal of Clinical Nutrition

showed a reduction, of borderline statistical significance, in


participants with complications at final follow-up in the
supplemented group (relative risk 0.50, 95% CI 0.251.00).
The results for unfavourable outcome (mortality or
complications) were only available for three oral supplementation trials (Stableforth, 1986; Delmi et al, 1990;
Hankins, 1996). Pooled data from these trials gave a
statistically significant result (Figure 3), in favour of the
supplemented group, at final follow-up (relative risk 0.52,
95% CI 0.320.84). However, an exploratory analysis based
on numbers randomised (in which it was assumed that all
excluded participants in the supplemented group had
complications at follow-up) showed the care that should be

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

899

Figure 1 Protein and energy supplementation (oral and nasogastric routes): mortality; 99% CI are presented for individual trials and 95% CI for
summary estimates.

taken when viewing results from small trials. Both the results
for complications (relative risk 0.70, 95% CI 0.371.34) and
unfavourable outcome (relative risk 0.67, 95% CI 0.421.05)
were no longer statistically significant.
One oral feeding study (Hankins, 1996) found no statistically significant effect of the supplement at 2 months on the
Barthel Index of functional ability. The greater tendency to
return to premorbid mobility of supplementation group
participants in another oral feeding study (Madigan, 1994)
may reflect that more supplemented participants were sent
to a rehabilitation hospital. Physiotherapy goals were
achieved more quickly in the intervention groups in one
nasogastric feeding study (Bastow et al, 1983b). A second
nasogastric feeding study (Sullivan et al, 1998) showed no
statistically significant difference between intervention and
control groups for activities of daily living at discharge.
No problems were reported for compliance in one oral
feeding study (Delmi et al, 1990); whereas 35% of patients
did not complete the course in another oral study (Hankins,
1996). Only 26% of the intervention group tolerated feeding
for the full 2 weeks in one nasogastric feeding trial (Hartgrink
et al, 1998), whereas 78% tolerated nasogastric feeding until
discharge from the ward in another (Bastow et al, 1983b).
Although 18 participants of the intervention group developed diarrhoea in the trial by Bastow et al (1983b), this was
ascribed mainly to antibiotics; no information was provided

on gastrointestinal complications for the control group.


Sullivan et al (1998) reported that there was no feed-induced
diarrhoea from nasogastric feeding, but three participants in
the intervention group had bloating in the early morning.
Volitional food intake was not significantly affected by
supplementation in three oral feeding studies (Delmi et al,
1990; Madigan, 1994; Hankins, 1996) and one nasogastric
feeding study (Sullivan et al, 1998). Another trial (Bastow
et al, 1983b) found that nasogastric feeding significantly
suppressed oral intake in thin, but not very thin patients.

Effect of protein-only supplement


Three studies (Tkatch et al, 1992; Schurch et al, 1998;
Espaulella et al, 2000) investigated whether approximately
20 g of protein in a supplement influenced the outcome from
hip fracture. The studies failed to carry out intention-to-treat
analyses for all outcomes. No significant effect on mortality
(Figure 4) could be demonstrated (relative risk 1.38, 95% CI
0.822.34).
An unfavourable outcome was significantly reduced by
protein supplementation in two trials (Figure 5) (relative risk
0.78, 95% CI 0.650.95). However, the results for one study
(Espaulella et al, 2000) should be viewed in the context of the
greater number of deaths in the protein supplementation
group.
European Journal of Clinical Nutrition

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

900

Figure 2 Protein and energy supplementation (oral and nasogastric routes): patients with complications at end of the study; 99% CI are
presented for individual trials and 95% CI for summary estimates.

Figure 3 Protein and energy supplementation (oral and nasogastric routes): unfavourable outcome; 99% CI are presented for individual trials
and 95% CI for summary estimates.

European Journal of Clinical Nutrition

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

901
No significant differences between intervention and
control groups, at 6 months follow-up, were reported in
one study (Espaulella et al, 2000) for mobility or Barthel
Index scores, and in another study (Schurch et al, 1998) for
activities of daily living.

Discussion
This review found limited evidence from 12 small trials
testing the effectiveness of nutritional supplementation after
hip fracture. Flawed study design, including the exclusion of
participants for poor compliance and short duration of
follow-up with insufficient ascertainment of important
outcomes, also compromises the quality of the available
evidence. While we acknowledge the particular difficulties in
conducting randomised controlled trials in frail, elderly
patients, it remains important to undertake intention-totreat analysis to examine the effectiveness of supplementation in the real world.
Combining data for protein and energy supplementation studies reveals no clear benefit in terms of mortality
or patients with complications. There is no firm evidence
that malnourished patients are more likely to benefit
from protein and energy supplementation than those
who are not malnourished. This could result from the

variation in definition of malnutrition and insufficient


numbers.
Oral protein and energy supplementation appears to
reduce the number of patients with complications and
unfavourable outcomes. However, this is based on three
small studies, and the largest study (Delmi et al, 1990) did
not account for all the participants randomised.
No clear effect of nasogastric feeding is evident. The high
mortality in the intervention group in one trial (Hartgrink
et al, 1998) is unexplained.
There is weak evidence that including protein in the
supplement improves rehabilitation. However, the higher
mortality in the supplemented group in one study (Espaulella et al, 2000) is unexplained, and the three studies failed
to account for all patients.
Inability to tolerate nasogastric tubes and problems with
palatability of oral feeds were common problems. Since
malnutrition produces mental apathy (Keys et al, 1950), this
may further reduce supplement intake. Ensuring increased
nutritional intake has major implications for hospital service
provision, and has ethical implications when the patient is
unwilling to feed or to tolerate nasogastric feeding.
Nasogastric feeding allowed the provision of higher energy
intakes in these studies (3.96.3 MJ/day, or 9001500 kcal/
day). Oral supplements provided under 1.7 MJ/day (400 kcal)

Figure 4

Oral protein supplementation: mortality; 99% CI are presented for individual trials and 95% CI for summary estimates.

Figure 5

Oral protein supplementation: unfavourable outcome; 99% CI are presented for individual trials and 95% CI for summary estimates.

European Journal of Clinical Nutrition

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

902
daily. Nasogastric feeding, which potentially has more risk of
complications, is likely to be targeted at those requiring
higher levels of supplementation. Attempts to overcome the
poor compliance with oral supplements, include specially
fortified hospital meals, the provision of frequent small
snacks, and small volume oral feeds prescribed with drug
rounds (Gall et al, 1998; Stephen et al, 1998; Potter et al,
2001). The provision of health-care assistants, specially
trained to help all aspects of patients feeding, is being
evaluated in the care of acutely ill older patients (Hickson
et al, 1999).
In conclusion, there is some evidence of the effectiveness
of nutritional supplementation for oral protein and energy
feeds, but the evidence is weak. The benefits of nasogastric
feeding are less certain, and it should probably be reserved
for the very malnourished, with intakes unresponsive to oral
supplementation.
Large, well-designed trials are needed of protein and
energy supplementation after hip fracture. Such trials should
examine the possible effect modification because of nutritional status, and consider the quality of life and cost issues
over at least 1 y after hip fracture.

Acknowledgements
We are grateful for the helpful comments from the following
during production of the Cochrane review: Dr Stuart Bruce,
Professor William Gillespie, Dr Ronald Koretz, Professor
Rajan Madhok, Professor Gordon Murray, Mr Martyn
Parker, Professor Gwyn Seymour and Professor Marc
Swiontkowski. We thank Professor Simon Allison, Ms
Melinda Bopp, Dr Katrina Brown, Professor Ian Cameron,
Ms Heidi Guyer, Dr Henk Hartgrink, Dr Ronald Koretz and Dr
Dennis Sullivan for providing further information about
trials. We thank Dr Ronald Koretz for help with hand
searching the Journal of Parenteral and Enteral Nutrition. We
particularly thank Professor Adrian Grant for his
constructive comments. We thank the anonymous referees
for their comments.

References
Autier P, Haentjens P, Bentin J, Baillon JM, Grivegnee AR, Closon MC
& Boonen S for the Belgian Hip Fracture Study Group (2000):
Osteoporosis Int. 11, 373380.
Avenell A, Handoll HHG & Grant AM (2001): Lessons for search
strategies from a systematic review, in The Cochrane Library, of
nutritional supplementation trials in patients after hip fracture.
Am. J. Clin. Nutr. 73, 505510.
Avenell A & Handoll HHG (2001): Nutritional Supplementation for Hip
Fracture Aftercare in the Elderly (Cochrane Review). The Cochrane
Library, Issue 3, 2001. Oxford: Update Software.
Bastow MD, Rawlings J & Allison SP (1983a): Undernutrition,
hypothermia, and injury in elderly women with fractured
femur: an injury response to altered metabolism? Lancet 1,
143146.
Bastow MD, Rawlings J & Allison SP (1983b): Benefits of supplementary tube feeding after fractured neck of femur: a randomized
controlled trial. BMJ 287, 15891592.

European Journal of Clinical Nutrition

Brown KM & Seabrook NA (1992): Nutritional influences on recovery


and length of hospital stay in elderly women following femoral
fracture. Proc. Nutr. Soc. 51, 132A.
Delmi M, Rapin C-H, Bengoa J-M, Delmas PD, Vasey H & Bonjour J-P
(1990): Dietary supplementation in elderly patients with fractured
neck of the femur. Lancet 335, 10131016.
Dolan P & Torgerson DJ (1998): The cost of treating osteoporotic
fractures in the United Kingdom female population. Osteoporosis
Int. 8: 611617.
Espaulella J, Guyer H, Diaz-Escriu F, Mellado-Navas JA, Castells M &
Pladevall M (2002): Nutritional supplementation of elderly hip
fracture patients. A randomized, double-blind, placebo-controlled
trial. Age Ageing 29, 425431.
Gall MJ, Grimble GK, Reeve NJ & Thomas SJ (1998): Effect
of providing fortified meals and between-meal snacks on
energy and protein intake of hospital patients. Clin. Nutr. 17,
259264.
Gallagher J, Schermbeck J, Dixon L & Labbe-Bell M (1992): Aggressive
early management of malnutrition in hip fracture patients. J.
Parenter. Enteral. Nutr. 16, 19S.
Hankins C (1996): Dietary Supplementation with Sustagen in Elderly
Patients with Fractured Neck of Femur (MSc dissertation). Sydney,
Australia: Sydney University.
Hartgrink HH, Wille J, Konig P, Hermans J & Breslau PJ (1998):
Pressure sores and tube feeding in patients with a fracture of the
hip: a randomized clinical trial. Clin. Nutr. 17, 287292.
Hickson M, Nicholl C, Bulpitt C, Fry M, Frost G & Davies L (1999):
The design of the Feeding Support TrialFdoes intensive feeding
support improve nutritional status and outcome in acutely ill
older in-patients? J. Hum. Nutr. Diet. 12, 5359.
Jallut D, Tappy L, Kohut M, Bloesch D, Munger R, Schutz Y, Chiolero
R, Felber JP, Livio JJ & Jequier E (1990): Energy balance in elderly
patients after surgery for a femoral neck fracture. J. Parenter.
Enteral. Nutr. 14, 563568.
Keys A, Brozek J, Henschel A, Mickelsen O & Taylor HL (1950): The
Biology of Human Starvation. Minneapolis, USA: University of
Minnesota.
Lesourd BM (1997): Nutrition and immunity in the elderly:
modification of immune responses with nutritional treatments.
Am. J. Clin. Nutr. 66, 478S484S.
Lumbers M, Murphy MC, Bowers V, Panayidou A, Older MW & New
SA (1999): Catering for elderly hospital patients: a study of food
choice in hip-fracture patients and the selection of suitable foods
for supplementation. Proc. Nutr Soc. 58, 148A.
Lumbers M, Driver LT, Howland RJ, Older MW & Williams CM
(1996): Nutritional status and clinical outcome in elderly female
surgical orthopaedic patients. Clin. Nutr. 15, 101107.
Lyons AR (1997): Clinical outcomes and treatment of hip fractures.
Am. J. Med. 103, 51S64S.
Madigan C (1994): Benefits of Dietary Supplementation in Elderly
Patients with Fractured Neck of Femur (MSc dissertation). Sydney,
Australia: Sydney University.
Maffulli N, Dougall TW, Brown MT & Golden MH (1999): Nutritional
differences in patients with proximal femoral fractures. Age Ageing.
28, 458462.
Mansell PI, Rawlings J, Allison SP, Bendall MJ, Pearson M, Bassey EJ &
Bastow M (1990): Low anthropometric indices in elderly females
with fractured neck of femur. Clin. Nutr. 9, 190194.
Patterson BM, Cornell CN, Carbone B, Levine B & Chapman D
(1992): Protein depletion and metabolic stress in elderly
patients who have a fracture of the hip. J. Bone Joint Surg. Am.
74, 251260.
Potter JM, Roberts MA, McColl JH & Reilly JJ (2001): Protein energy
supplements in unwell elderly patientsFa randomised controlled
trial. J. Parenter. Enteral. Nutr. 25, 323329.
Schurch MA, Rizzoli R, Slosman D, Vadas L, Vergnaud P & Bonjour JP
(1998): Protein supplements increase serum insulin-like
growth factor-I levels and attenuate proximal femur bone
loss in patients with recent hip fracture. A randomized,

Protein and energy supplementation for hip fracture


A Avenell and HHG Handoll

903
double-blind, placebo-controlled trial. Ann. Intern. Med. 128,
801809.
Stableforth PG (1986): Supplement feeds and nitrogen and
calorie balance following femoral neck fracture. Br. J. Surg. 73,
651655.
Stephen AD, Beigg CL, Elliott ET, MacDonald IA & Allison SP (1998):
A recipe for improving the nutritional intake in elderly hospital
patients. Proc. Nutr. Soc. 57, 89A.

Sullivan DH, Nelson CL, Bopp MM, Puskarich-May CL &


Walls RC (1998): Nightly enteral nutrition support of
elderly fracture patients: a phase I trial. J. Am. Coll. Nutr. 17,
155161.
Tkatch L, Rapin CH, Rizzoli R, Slosman D, Nydegger V, Vasey H &
Bonjour JP (1992): Benefits of oral protein supplementation in
elderly patients with fracture of the proximal femur. J. Am. Coll.
Nutr. 11, 519525.

European Journal of Clinical Nutrition