Anaesthesia For Hip Fracture Surgery in Adults Cochrane 2004

Anaesthesia for hip fracture surgery in adults (Review)
Parker MJ, Handoll HHG, Griffiths R, Urwin SC
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2004, Issue 3
http://www.thecochranelibrary.com
Anaesthesia for hip fracture surgery in adults (Review) 1

Copyright © 2004 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
TABLE OF CONTENTS
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
SYNOPSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
CRITERIA FOR CONSIDERING STUDIES FOR THIS REVIEW . . . . . . . . . . . . . . . . . . 3
SEARCH STRATEGY FOR IDENTIFICATION OF STUDIES . . . . . . . . . . . . . . . . . . . . 4
METHODS OF THE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
DESCRIPTION OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
METHODOLOGICAL QUALITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
REVIEWERS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
NOTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
POTENTIAL CONFLICT OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Characteristics of included studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Characteristics of excluded studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
GRAPHS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Comparison 01. Regional (spinal or epidural) versus general anaesthesia . . . . . . . . . . . . . . . . 27
Comparison 02. Spinal and ’light’ general anaesthetic versus general anaesthetic . . . . . . . . . . . . . 28
Comparison 03. Regional (spinal or epidural) versus lumbar plexus nerve blocks . . . . . . . . . . . . . 28
Comparison 04. Intravenous ketamine versus general anaesthesia . . . . . . . . . . . . . . . . . . 28
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
COVER SHEET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
GRAPHS AND OTHER TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Comparison 04. 01 Mortality - 1 month . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Comparison 04. 02 Mortality - 1 month (random effects model) . . . . . . . . . . . . . . . . . . 31
Comparison 04. 03 Mortality - 3 months . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Comparison 04. 06 Mortality - early and up to 1 month . . . . . . . . . . . . . . . . . . . . . 33
Comparison 04. 07 Length of operation (mins) . . . . . . . . . . . . . . . . . . . . . . . . 34
Comparison 04. 08 Operative hypotension . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Comparison 04. 09 Operative hypotension (random effects model) . . . . . . . . . . . . . . . . . 35
Comparison 04. 10 Operative blood loss (mls) . . . . . . . . . . . . . . . . . . . . . . . . . 35
Comparison 04. 11 Patients receiving blood transfusion . . . . . . . . . . . . . . . . . . . . . 36
Comparison 04. 12 Transfusion requirements (mls) . . . . . . . . . . . . . . . . . . . . . . . 36
Comparison 04. 13 Post-operative hypoxia . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Comparison 04. 14 Length of hospital stay . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Comparison 04. 15 Pneumonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Comparison 04. 16 Myocardial infarction . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Comparison 04. 17 Cerebrovascular accident . . . . . . . . . . . . . . . . . . . . . . . . . 40
Comparison 04. 18 Congestive cardiac failure . . . . . . . . . . . . . . . . . . . . . . . . . 41
Comparison 04. 19 Renal failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Comparison 04. 20 Acute confusional state . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Comparison 04. 21 Urine retention . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Comparison 04. 22 Vomiting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Comparison 04. 23 Deep vein thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Anaesthesia for hip fracture surgery in adults (Review) i
Comparison 04. 24 Pulmonary embolism . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Comparison 04. 25 Pulmonary embolism (random effects model) . . . . . . . . . . . . . . . . . . 46
Comparison 04. 26 Pulmonary embolism (fatal and non fatal) . . . . . . . . . . . . . . . . . . . 47
Comparison 04. 01 Mortality - 1 month . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Comparison 04. 02 Length of operation . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Comparison 04. 03 Pneumonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Comparison 04. 04 Confusional state . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Comparison 04. 05 Deep vein thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Comparison 04. 01 Incomplete or unsatisfactory analgesia . . . . . . . . . . . . . . . . . . . . 50
Comparison 04. 02 Operative hypotension . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Comparison 04. 03 Mean fall in arterial blood pressure (mmHg) . . . . . . . . . . . . . . . . . . 51
Comparison 04. 04 Mean dose of ephedrine used (mg) . . . . . . . . . . . . . . . . . . . . . . 51
Comparison 04. 05 Adverse effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Comparison 04. 06 Post-operative confusion . . . . . . . . . . . . . . . . . . . . . . . . . 52
Comparison 04. 01 Mortality - during hospital stay . . . . . . . . . . . . . . . . . . . . . . . 53
Comparison 04. 02 Myocardial infarction . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Comparison 04. 03 Congestive cardiac failure . . . . . . . . . . . . . . . . . . . . . . . . . 54
Comparison 04. 04 Pulmonary embolism . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Comparison 04. 05 Length of hospital stay (discharge home) . . . . . . . . . . . . . . . . . . . . 54
Anaesthesia for hip fracture surgery in adults (Review) ii

Anaesthesia for hip fracture surgery in adults (Review)
Parker MJ, Handoll HHG, Griffiths R, Urwin SC
This record should be cited as:

Parker MJ, Handoll HHG, Griffiths R, Urwin SC. Anaesthesia for hip fracture surgery in adults. The Cochrane Database of Systematic
Reviews , Issue . Art. No.: CD000521. DOI: 10.1002/14651858.CD000521.
This version first published online: 23 October 2001 in Issue , .

Date of most recent substantive amendment: 04 July 2001
ABSTRACT
Background
The majority of hip fracture patients are treated surgically, requiring anaesthesia.
Objectives
To compare different types of anaesthesia for surgical repair of hip fractures (proximal femoral fractures) in adults.
Search strategy
We searched the Cochrane Musculoskeletal Injuries Group specialised register (December 2000), MEDLINE (1996 to December
Week 4 2000) and reference lists of relevant articles.
Selection criteria
Randomised and quasi-randomised trials comparing different methods of anaesthesia for hip fracture surgery in skeletally mature
persons. The primary focus of this review was the comparison of regional (spinal or epidural) anaesthesia versus general anaesthesia; this
has been expanded to include other comparisons. The use of nerve blocks pre-operatively or in conjunction with general anaesthesia is
evaluated in another review. The primary outcome was mortality.
Data collection and analysis
Two reviewers independently assessed trial quality, using a nine item scale, and extracted data. Results were pooled wherever appropriate
and possible.
Main results
Seventeen trials, involving 2305 patients, comparing regional anaesthesia with general anaesthesia were included. All trials had method-
ological flaws. Pooled results from eight trials showed regional anaesthesia to be associated with a decreased mortality at one month
(53/781(6.8%) versus 78/826(9.4%)); this was of borderline statistical significance (relative risk (RR) 0.72, 95% confidence interval
(CI) 0.51 to 1.00). The results from six trials for three month mortality were not statistically significant, although the confidence
interval does not exclude the possibility of a clinically relevant reduction (86/726 (11.8%) versus 98/765 (12.8%), RR 0.92, 95% CI
0.71 to 1.21). The reduced numbers of patients at one year, coming exclusively from two studies, preclude any useful conclusions for
long term mortality (80/354 (22.6%) versus 78/372 (21.0%), RR 1.07, 95% CI 0.82 to 1.41).
Regional anaesthesia was associated with a tendency to a longer operation (weighted mean difference 4.8 minutes, 95% CI 1.1 to
8.6 minutes), and a reduced risk of deep venous thrombosis (39/129 (30%) versus 61/130 (47%); RR 0.64, 95% CI 0.48 to 0.86),
although this conclusion is insecure due to possible selection bias in the subgroups in whom this outcome was measured. No other
statistically significant differences in outcome were identified.
There was insufficient evidence to draw any conclusions from a further four included trials, involving a total of 179 patients, which
compared other methods of anaesthesia (a ’light’ general with spinal anaesthesia; intravenous ketamine; nerve blocks).
Reviewers’ conclusions
Regional anaesthesia and general anaesthesia appear to produce comparable results for most of the outcomes studied. Regional anaesthesia
may reduce short-term mortality but no conclusions can be drawn for longer term mortality.
SYNOPSIS
Regional anaesthesia may reduce the number of early deaths after hip fracture surgery but more evidence is needed to establish if it is
better than general anaesthesia
The majority of people with hip fracture are treated surgically. Anaesthesia is used to prevent pain during the operation. There are a
number of different types of anaesthesia and the most common are ’general’ and ’spinal’. General anaesthesia, which usually involves
a loss of consciousness, typically includes inhalation of gases. Spinal (regional) anaesthesia involves an injection into the space around
the spinal cord, to prevent pain in the involved limb. Although there were fewer early deaths (within one month) in people given spinal
anaesthesia, there was not enough clear evidence to tell if regional anaesthesia was superior to general anaesthesia. The effectiveness of
other methods of anaesthesia could not be determined.
BACKGROUND an airway, intra-operative hypotension, aspiration of gastric con-

tents, post-operative nausea, respiratory depression and damage
The scope of this review, originally published in Issue 4, 1999, has to the teeth or upper airways.
been expanded in this update, published in Issue 4, 2001, to cover Regional (also termed spinal) anaesthesia for hip fracture surgery
other methods of anaesthesia. However, the main focus remains refers to the injection of a local anaesthetic into the epidural or
the comparison of regional versus general anaesthesia. subarachnoid space at the lumbar spine. In some cases the pa-
The term proximal femoral fracture, or ’hip fracture’, refers to a tient also receives sedatives whilst the block is inserted and possi-
fracture of the femur in the area of bone immediately distal to the bly during the surgery itself. The main complication of a regional
articular cartilage of the hip, to a level of about five centimetres technique is intra-operative hypotension, which may lead to cere-
below the lower border of the lesser trochanter. The majority of brovascular or myocardial ischaemia or infarction. Other prob-
these fractures occur in an elderly population with an average age lems may be an inadequate regional block, the rare complications
of around 80 years. Females predominate over males by about of damage to local structures and headache secondary to leakage
four to one (Parker 1993) and the injury is usually the result of a of cerebrospinal fluid from the dural puncture site. Specific advan-
simple fall. Whilst the hip fracture is usually the only injury, the tages of regional anaesthesia may be a reduction in the incidence
patients frequently have many other medical problems associated of thrombotic episodes and a reduced operative blood loss. These
with aging. may be a consequence of an increased peripheral limb blood flow
in combination with reduced venous tone. Alternatively they may
An estimated 1.7 million hip fractures occurred worldwide in the arise from an alteration of blood viscosity and coagulability, as a
year 1990 (WHO study group 1994). The number of hip fracture result of changes in the metabolic and neurohumoral responses to
patients continues to rise, due to a combination of an increasingly surgery (Modig 1983).
elderly population and an increase in the age specific incidence. A
prediction for global numbers of 6.26 million hip fractures by the Other forms of anaesthesia used for hip fracture surgery are the
year 2050 has been made (Melton 1993). The majority of these insertion of local nerve blocks around the hip. These may be sup-
fractures are treated surgically; thus hip fracture surgery represents plemented with sedatives, analgesics or other parental drugs. A
one of the most common emergency orthopaedic procedures. Sur- lumber plexus block refers to injection of a local anaesthetic agent
gical treatment may be either fixation of the fracture or replace- into the area of the lumbar plexus close to the transverse process
ment of the femoral head with an arthroplasty. Internal fixation of the forth lumbar vertebrae (Winnie 1974). Only the plexus on
involves using screws or pins, either alone or in combination with the side of the fracture needs to be blocked, which may reduce
a side plate applied to the femur, or by the use of an intramedullary the incidence of complications such as operative hypotension. A
nail with a cross screw inserted into the femoral head. Arthroplasty sacral plexus block refers to the injection of a local anaesthetic
involves excision of the fractured area of bone and replacement agent in the area around the sacral nerves (Mansour 1993). The
with a partial or total hip replacement, which may be cemented use of nerve blocks pre-operatively or in conjunction with gen-
in place. eral anaesthesia is considered in another Cochrane review (Parker
2001).
General anaesthesia refers to the use of a variety of intravenous and
An alternative type of anaesthetic involves the use of intravenous
or inhalation drugs to render the patient unconscious. The pa-
ketamine on its own. Ketamine renders the patient unconscious,
tient may breathe spontaneously or require mechanical ventilation
thereby acting as a general anaesthetic, and has analgesic effects.
following the administration of neuromuscular blocking agents.
Potential complications of general anaesthesia include adverse re- No consensus exists as to which is the best method of anaesthesia.
actions to the drugs used, difficulty in maintaining or establishing Currently the choice of anaesthesia used for hip fracture surgery
is determined by the personal preference of the anaesthetist con- Types of intervention
cerned, following assessment of the patient’s medical state. A gen- 1. Regional anaesthesia (if necessary supplemented by sedatives)
eral review of anaesthesia for hip fracture surgery (Covert 1989) achieved by injection of local anaesthetic into the epidural or sub-
summarised the possible advantages of different anaesthetic meth- arachnoid spaces. This type of anaesthesia is also referred to as
ods using information from eight of the randomised trials on this ’spinal’ or ’epidural’.
subject. In a meta-analysis, using Bayesian methods, of 11 ran-
domised trials of regional versus general anaesthesia for surgical 2. General anaesthesia using intravenous or inhalation agents to
repair of hip fractures, Sorensen 1992 concluded that the supe- render the patient unconscious. Unless otherwise stated, general
riority of one method over the other was unproven. Not all cur- anaesthesia refers to general anaesthesia using inhalation agents in
rently available randomised trials were included and, moreover, this review.
some trial data from two studies were duplicated in the analysis. 3. Intravenous ketamine.
A more recent meta-analysis of randomised trials for all types of
surgery has demonstrated a reduction of early post-operative mor- 4. Local nerve blocks (if necessary supplemented by sedatives)
tality and morbidity with epidural or spinal anaesthesia (Rodgers when used as the primary method of anaesthesia.
2000).
Trials testing other methods of anaesthesia as the primary method
of anaesthesia were considered for inclusion. Trials comparing the
use of local nerve blocks in conjunction with general anaesthesia
OBJECTIVES
and the use of nerve blocks pre-operatively, are evaluated in another
Cochrane review (Parker 2001). Also not considered in this review
To determine the optimum anaesthetic technique for hip frac-
were trials comparing different types of drugs or techniques of
ture surgery. Different types of anaesthesia, namely regional (ei-
individual methods of anaesthesia.
ther spinal or epidural), inhalation general anaesthesia, local nerve
blocks and intravenous ketamine anaesthesia were compared. Vari- Types of outcome measures
ations in anaesthetic drug dosage and delivery or supplementary The primary outcome measure was mortality (at 1 month, 3
regional blocks were not considered within this review. months, 6 months and 1 year). In addition, data were sought from
The following null hypotheses were tested within the trials in- each study for outcomes in the following categories.
cluded so far in this review:
a) Peri-operative outcomes:
1. There is no difference in outcome between regional anaesthesia - length of operation (in minutes)
(spinal or epidural) and general anaesthesia. - hypotension (intra-operative or immediately post-operative)
- operative blood loss (in millilitres)
2. There is no difference in outcome between regional anaesthesia - transfusion requirements/fall in haemoglobin
(spinal or epidural) supplemented with a ’light’ general anaesthetic - need for supplementary drugs to complete anaesthetic (new in
and general anaesthesia alone. second update)
3. There is no difference in outcome between regional anaesthesia - changes in body temperature
(spinal or epidural) and regional nerve blocks alone. - pre and post-operative arterial blood gases
- changes in catecholamines and other stress response chemicals
4. There is no difference in outcome between anaesthesia using during and after surgery
ketamine (with or without a benzodiazepine) and inhalation gen- - intra-operative cardiac arrhythmias
eral anaesthesia. - time to mobilisation
- length of hospital stay (in days)
CRITERIA FOR CONSIDERING b) Complications specific to the method of treatment:
STUDIES FOR THIS REVIEW - aspiration pneumonia
- post-dural puncture headache
Types of studies - damage to the upper airways or mouth from general anaesthesia
- secondary intervention required for anaesthetic complications
All randomised controlled trials comparing different methods of - any other adverse effects as detailed in each study (new in second
anaesthesia were included. Quasi-randomised trials (for example, update)
alternation), and trials in which the treatment allocation was in-
adequately concealed, were considered for inclusion. c) General post-operative complications:
(unless otherwise specified, the definition for these complications
Types of participants will be as detailed in each study, or by post-mortem)
Skeletally mature patients undergoing hip fracture surgery. - pneumonia
- myocardial infarction for its quality of methodology. Any differences were resolved by
- cerebrovascular accident discussion between the reviewers.
- congestive cardiac failure
The main assessment was by the quality of concealment of
- renal failure
allocation which was scored either A, B or C according to the
- cardiac arrhythmias
criteria in the Cochrane Reviewers’ Handbook (Clarke 2000), or
- acute confusional state
3, 2, 1 or 0 as described below (item 1). A further eight aspects of
- urine retention (requiring catheterisation)
methodology were used, giving a maximum score for each study
- post-operative nausea and /or vomiting
of 11.
- deep vein thrombosis (diagnosis confirmed by post-mortem,
venography, isotope scanning, ultrasound or phlethysmography, 1. Trials with clear concealment of allocation (e.g. numbered
whether this was performed routinely or only as clinically indi- sealed opaque envelopes drawn consecutively) were coded as A and
cated) scored 3. Those in which there was a possible chance of disclosure
- pulmonary embolism (diagnosed by isotope scanning, angiog- of assignment were coded as B and scored 2. Those in which
raphy or post-mortem) allocation concealment was not stated, or unclear, were coded as
B and scored 1. Those where allocation concealment was clearly
d) Final outcome measures:
not concealed, such as trials using quasi-randomisation (e.g. even
- mortality (primary outcome)
or odd date of birth), were coded as C and scored 0.
- change in mental function
- functional status 2. Were the inclusion and exclusion criteria clearly defined? Score
- return of patient to their pre-fracture place of residence 1 if text states type of patients included and those excluded;
otherwise score 0.
3. Were the outcomes of patients who withdrew or were excluded
SEARCH STRATEGY FOR after allocation described and included in an intention to treat
IDENTIFICATION OF STUDIES analysis? This particularly applies to patients allocated to regional
anaesthesia where it was not achieved due to technical difficulties.
See: search strategy Score 1 if these patients were either detailed separately or included
We searched the Cochrane Musculoskeletal Injuries Group in the analysis group to which they were allocated, or if text states
specialised register (December 2000), MEDLINE (1996 to that no withdrawals occurred; otherwise score 0.
December Week 4 2000) and reference lists of relevant articles. 4. Were the treatment and control groups adequately described at
In MEDLINE (OVID WEB) the following search strategy was entry? Score 1 if a minimum of four admission details were given
combined with the first two levels of the optimal trial search (e.g. age, sex, mobility, fracture type, function score, ASA grade,
strategy (Clarke 2000). mental test score); otherwise score 0.
1. exp Hip Fractures/ 5. Were the care programmes other than trial options identical?
2.((hip$ or femur$ or femoral$ or trochant$ or pertrochant$ Score 1 if text states they were; otherwise score 0.
or intertrochant$ or subtrochant$ or intracapsular$ or
extracapsular$) adj4 fracture$).tw. 6. Were the outcome measures clearly defined in the text? Score 1
3. or/1-3 if yes; otherwise score 0.
4. exp Anesthesia/ 7. Were the outcome assessors blind to treatment group? Score 1
5. ((an?esthet$ or an?esthesia) adj4 (regional$ or local$ or general if yes; otherwise score 0.
or spinal or epidural)).tw.
6. or/4-5 8. Was the timing of outcome measures appropriate? This was
7. and/3,6 considered to be a minimum of three months follow-up for all
surviving patients. Score 1 if yes; otherwise score 0.
Articles of all languages were considered and translated if
necessary. 9. Was loss to follow-up reported and if so were less than five per
cent of patients lost to follow-up? Score 1 if yes; otherwise score
0.
METHODS OF THE REVIEW Heterogeneity between comparable trials was tested using a
standard chi-squared test. In accordance with the revised
Data for the outcome measures listed above were independently statistical policy of the Cochrane Musculoskeletal Injuries Group,
extracted by two reviewers, and checked by at least one of the other announced in March 2000, relative risks and 95 per cent
two reviewers. In addition each trial was assessed without masking confidence limits have been calculated instead of Peto odds ratios
and 95 per cent confidence limits for dichotomous outcomes. 1980) of 40 patients, which compared a ’light’ general anaesthetic
Mean differences and 95 per cent confidence limits have been in conjunction with spinal anaesthesia versus general anaesthesia,
calculated for continuous outcomes. Results of comparable groups is considered separately. A further group of 20 patients in this
of trials were pooled using fixed and random effects models and study were allocated to receive a psoas nerve block in conjunction
95 per cent confidence limits. Both Peto odds ratio and relative with general anaesthesia, which is outside the scope of this review
risk plots were viewed and a note was taken of where there was but included in another Cochrane review on localised nerve blocks
statistically significant heterogeneity (P< 0.1) using either method. (Parker 2001). Two newly included studies compared spinal anaes-
The results for the random effects model are presented when thesia with nerve blocks (de Visme 2000; Eyrolle 1998). The re-
there is significant heterogeneity in the results of individual trials. maining trial (Spreadbury 1980) compared ketamine anaesthesia
Any tests of interaction, calculated to determine if the results for with inhalation general anaesthesia in 60 patients.
subgroups were significantly different, are based on odds ratio
results. Further details of the individual trials are given in the Character-
istics of Included Studies table.
DESCRIPTION OF STUDIES Additional information on trial methodology and results would be

welcomed from the authors of any of the studies, or from authors
Three new studies (de Visme 2000; Eyrolle 1998; Ungemach of trials which have not been identified.
1993) were included in this second update. In the first update, the
study of Juelsgaard 1998 was included.
METHODOLOGICAL QUALITY
In all, 37 studies were identified of which 21 trials were included in
this review, 15 were excluded and one remains in Studies Awaiting
Treatment allocation was considered to be definitely concealed
Assessment. Of the 15 excluded studies, two were not randomised
(Cochrane code A) in only one study (McKenzie 1984), which
trials; eight involved comparisons outside the scope of this review;
used sealed envelopes and random numbers. Allocation conceal-
two (Tonczar 1981; Wickstrom 1982) involved neuroleptic gen-
ment was possible (Cochrane code B) in a further five studies
eral anaesthesia which was considered to be no longer appropri-
(Brown 1994; Couderc 1977; de Visme 2000; Maurette 1988;
ate for hip fracture surgery; one (Darling 1994) only reported
Racle 1986) which gave incomplete details of their methods of
one outcome, the rate of clearance of injected indocyanine green,
randomisation, as well as the 14 studies which did not provide
which was considered not to have direct clinical relevance; one
any details. Allocation was not concealed in the only overtly quasi-
(El-Zahaar 1995) involving a mixed population of orthopaedic
randomised trial (Adams 1990) which allocated treatment by the
patients did not provide separate data for hip fracture patients;
date of operation.
and one (Dyson 1988) with a factorial design which focused on a
comparison outside the review scope, did not provide any results The methodology scores using the scoring system described earlier
for the spinal versus general anaesthesia comparison. Further de- were:
tails of these are given in the Characteristics of Excluded Studies
table. The one trial (Wajima 1995) in Studies Awaiting Assess- REGIONAL VERSUS GENERAL ANAESTHESIA
ment awaits translation from Japanese. A request has been sent to 1 2 3 4 5 6 7 8 9 Total (maximum 11)
the contact author for further details of the study. ————————————-
0 0 0 1 0 0 0 0 1 2 Adams 1990
The 21 included trials involved a total of 2484 predominantly fe-
1 1 0 1 1 1 1 1 0 8 Berggren 1987
male and elderly hip fracture patients. Translations were obtained
1 1 0 1 1 1 1 1 0 7 Bigler 1985
for three trial reports in French and one in German. Seventeen tri-
1 1 0 1 1 1 0 0 1 6 Bredahl 1991
als were published as full reports in peer-reviewed journals; the four
1 1 0 0 0 1 1 0 0 4 Brichant 1995
exceptions (Brichant 1995; Eyrolle 1998; Tasker 1983; Ungemach
2 1 0 1 0 1 0 0 1 6 Brown 1994
1993) being only available as conference abstracts. Two trial re-
1 1 0 1 1 1 0 0 1 6 Davis 1981
ports were available for Davis 1981, one of which focused on a
2 1 0 1 0 1 0 1 0 6 Davis 1987
sub-group of patients monitored for deep vein thrombosis. Four
1 1 0 1 1 1 1 0 0 6 Juelsgaard 1998
references, one again which focused on a sub-group of patients
2 1 0 1 0 1 0 0 1 6 Maurette 1988
monitored for deep vein thrombosis, were available for McKen-
1 0 0 1 0 1 0 0 1 4 McLaren 1978
zie 1984. Though these at first appeared to be reports of separate
3 0 0 0 0 1 0 1 1 6 McKenzie 1984
trials, further details supplied by another trialist indicated that all
2 1 0 0 1 1 0 1 1 7 Racle 1986
the references applied to one study.
1 0 0 0 0 1 0 0 0 2 Tasker 1983
Seventeen included trials involving 2305 patients compared spinal 1 0 0 0 0 0 0 0 0 1 Ungemach 1993
or epidural anaesthesia with general anaesthesia. One study (White 1 1 0 1 0 1 1 1 1 7 Valentin 1986
’LIGHT’ GENERAL ANAESTHESIA COMBINED WITH thrombosis, oxygen saturation and psychological evaluation re-
SPINAL ANAESTHESIA VERSUS GENERAL ANAESTHE- spectively. Where possible, data for mortality up to one, three,
SIA six and twelve months were deduced or extracted from study re-
1 2 3 4 5 6 7 8 9 Total ports, and pooled, for these four pre-specified time periods. Data
————————————- for three months and beyond were extracted from graphs for two
1 1 0 1 0 1 0 0 0 4 White 1980 studies (Davis 1987; Valentin 1986). Additional mortality data
were obtained for McKenzie 1984 from another trialist. Mortal-
REGIONAL (SPINAL) ANAESTHESIA VERSUS LOCAL ity data for undefined follow-up periods, or for under one month
NERVE BLOCKS were provided by four studies (Adams 1990; Bigler 1985; Tasker
1 2 3 4 5 6 7 8 9 Total 1983; Ungemach 1987). The data for the first two studies, which
————————————- were for early deaths during hospital stay, and those for Ungemach
2 1 0 1 1 1 0 0 1 7 de Visme 2000 1987, which were at two weeks, were pooled with those for one
1 0 0 0 0 0 0 0 1 2 Eyrolle 1998 month in an extra analysis. Tasker 1983 reported, without pro-
KETAMINE VERSUS GENERAL ANAESTHESIA viding denominators, that the difference in mortality was not sta-
1 2 3 4 5 6 7 8 9 Total tistically different between the two groups (4 versus 6).
————————————- Results for all these studies are shown in the analyses tables. The
1 1 0 1 1 0 0 0 1 5 Spreadbury 1980 reduced mortality for regional anaesthesia at one month (53/781
Two items meriting specific comment are items 3 (intention to (6.8%) versus 78/826 (9.4%)) was of borderline statistical sig-
treat) and 7 (assessor blinding). No trial satisfied the criteria for nificance when evaluated using the fixed effects method (relative
the first item, either because no information was available for pa- risk (RR) 0.72, 95% confidence interval (CI) 0.51 to 1.00), but
tients withdrawn from the study or because those who had been not statistically significant when using the random effects model
withdrawn or excluded were not included in the baseline or out- (RR 0.73, 95% CI 0.47 to 1.12). There was a similar pattern
come analyses, or because an intention to treat analysis was not when the results from the three studies (Adams 1990; Bigler 1985;
done. The extent of assessor blinding was usually limited to select Ungemach 1987), which provided data on deaths during hospi-
outcomes in most of the trials scoring on this item. tal stay or under one month, were pooled with the data for one
month mortality (see analysis). The difference in mortality be-
tween the two groups was smaller and not statistically significant
at subsequent follow-up times. The number of trials and associ-
RESULTS
ated data for pooling shrank at each time interval, with only the
two largest trials (McKenzie 1984; Valentin 1986) contributing
The outcome measures listed earlier were extracted for each study
to the 12 month analysis. Mortality at three months appeared
and, where appropriate data were available, summarised in the
marginally less in the regional anaesthesia group (86/726 (11.8%)
analysis tables. The results are presented using the fixed effects
versus 98/765 (12.8%), RR 0.92, 95% CI 0.71 to 1.21), but
model except where there is statistically significant heterogeneity
slightly greater at six months (103/613 (16.8%) versus 115/651
between study results (P < 0.1) where the random effects model
(16.1%), RR 1.04, 95% CI 0.81 to 1.33) and 12 months (80/354
is applied. Since the primary outcome for this review, as stated in
(22.6%) versus 78/372 (21.0%), RR 1.07, 95% CI 0.82 to 1.41).
the protocol, is mortality, this is considered first. Other outcomes
are presented in the categories listed under Types of outcome mea- Other outcomes:
sures; these include surrogate or intermediate outcomes, such as
peri-operative hypotension, body temperature and arterial blood a) Peri-operative outcomes
gases. Although such outcomes may be predictive of important Length of operation
clinical outcomes, the relationship is usually not an exact one and Most studies that recorded this outcome reported a statistically
some conditions, such as operative hypotension, may be remedied non-significant increase in the time taken to complete the opera-
to reduce the risk of a serious clinical event occurring. Thus the tion for regional anaesthesia (Adams 1990; Berggren 1987; Bigler
results of such outcomes are not accurate guides of ’hard’ clinical 1985; Maurette 1988; McKenzie 1984; Racle 1986). One study
outcomes and may be misleading. had a non-significant increase for general anaesthesia (Bredahl
REGIONAL VERSUS GENERAL ANAESTHESIA 1991) and three studies found no difference between the two
groups (Davis 1981; Juelsgaard 1998; White 1980). Pooling of
Mortality data from six studies showed a statistically significant increase of
around five minutes for regional anaesthesia (weighted mean dif-
Mortality was reported in most studies, except for four short-
ference 4.8 minutes, 95% CI 1.1 to 8.6 minutes).
term studies (Bredahl 1991; Brichant 1995; Brown 1994; Mau-
rette 1988), whose primary foci were body temperature, deep vein Hypotension
The definition of hypotension, when stated, was a greater than: haemoglobin to be greater in the regional anaesthesia group (22%
30 per cent reduction in systolic blood pressure (Berggren 1987), versus 19%, not significant).
33 per cent fall (Juelsgaard 1998), 40mmHg fall (Couderc 1977),
and 20 per cent fall from the baseline in three studies (Davis 1987; Pre- and post-operative arterial blood gases
Maurette 1988; Racle 1986). The reports of six studies (Berggren 1987; Brown 1994; Couderc
1977; Davis 1981; McLaren 1978; McKenzie 1984) contained
Two studies (Adams 1990; Davis 1981) stated, without data for data for blood gases taken either pre-operatively, operatively or
pooling, that the drop in systolic blood pressure was significantly post-operatively. Berggren 1987 reported numbers of patients with
greater in the regional anaesthesia group. Bigler 1985 reported post-operative arterial oxygen tension of less than 60mmHg and
no significant difference in the maximum drop of systolic blood these are presented in an analysis table (10/28 (36%) versus 14/29
pressure (48 versus 51mmHg). Pooling of data from eight studies (48%), RR 0.74, 95% CI 0.40 to 1.38). Brown 1994, in a study
(Berggren 1987; Brown 1994; Couderc 1977; Davis 1987; Juels- of post-operative oxygen saturation in 20 patients, found signif-
gaard 1998; Maurette 1988; McLaren 1978; Racle 1986) showed icantly lower oxygen saturation for the group who received gen-
hypotension to be more common after regional anaesthesia. This eral anaesthesia. Davis 1981 reported that the general anaesthesia
difference was statistically significant when viewed using the Peto group showed a post-operative fall in oxygen saturation in the early
fixed effects method (158/441 (35.8%) versus 125/461 (27.1%), post-operative period, which was not seen after regional anaesthe-
RR 1.31, 95% CI 1.09 to 1.58), but not when adopting the ran- sia. By the first post-operative day there was no significant differ-
dom effects model (RR 1.18, 95% CI 0.87 to 1.60) which is more ence between the two groups. McKenzie 1984 reported a signifi-
probably more appropriate given the significant heterogeneity of cant decrease in the oxygen saturation at one hour post-operatively
trial results (chi-square = 11.28, P = 0.08). An exploration of the in those who received general anaesthesia compared with those
effect of removing each of the trials in turn from the analysis re- who received regional anaesthesia. In contrast, two studies (Coud-
vealed that the removal of the data from Couderc 1977 produced erc 1977; McLaren 1978) reported no difference in the mean arte-
the most homogeneous result (chi-square = 0.32). Although this rial oxygen or carbon dioxide tensions for the two types of anaes-
may reflect the different definition of hypotension in this trial, thesia.
there are too many other reasons to be certain that this is the case.
Length of hospital stay
Operative blood loss Most studies reporting this found no difference in the length of
Pooled data for three studies (Bredahl 1991; Davis 1981; McKen- hospital stay. Juelsgaard 1998 observed that the results for hospital
zie 1984) show a statistically non-significant increase in operative stay were affected by a lack of rehabilitation facilities. Adams 1990
blood loss for general anaesthesia (weighted mean difference 81ml, reported 21 days for regional versus 20 days for general anaesthesia.
95% CI -53 to 216ml). Five other studies contained insufficient Berggren 1987 stated there was no difference in length of hospital
data to enable pooling. Adams 1990 and Juelsgaard 1998 reported stay between the two groups. Davis 1987 reported an average of
a non-significant increase in blood loss for regional anaesthesia; 16 days for both groups, and Racle 1986, 20 days for both groups.
McLaren 1978 reported no significant difference; Ungemach 1987 Valentin 1986 reported a median stay of 10 days for regional
reported no difference; and Valentin 1986 reported a significantly anaesthesia and 11 days for general anaesthesia. Finally, McKenzie
increased blood loss in the general anaesthesia group. 1984 recorded a mean of 38 days for regional anaesthesia against
43 days for general anaesthesia. Summation of the two studies
Transfusion requirements which quoted standard deviations (McKenzie 1984; Racle 1986),
Seven studies gave data for blood transfusion, which are presented shown in the analysis tables, demonstrated no difference in the
as either the numbers of patients who were transfused in three stud- length of hospital stay between groups (weighted mean difference
ies (Adams 1990; Bigler 1985; Davis 1981), or the mean volume of -0.2 days, 95% CI -5.2 to 4.8 days).
blood transfused (transfusion requirement) (Couderc 1977; Juels-
gaard 1998; Maurette 1988; Racle 1986). Similar proportions of Other peri-operative outcomes
patients received transfusion in each group in the first three studies Other peri-operative outcomes recorded were changes in body
(63/108 (58.3%) versus 68/120 (56.6%)). In contrast the trans- temperature (Bredahl 1991), serum catecholamine and endocrine
fusion requirements were greater in the regional anaesthesia group levels (Adams 1990; Tasker 1983), ECG changes (Juelsgaard 1998)
but there was significant heterogeneity (chi square = 12.63, P < and time to ambulation (Bigler 1985; Valentin 1986). Ungemach
0.01) in the trial results and the pooled result was not statisti- 1993 used a scoring system which included level of consciousness,
cally significant (weighted mean difference 141ml, 95% CI -40 respiration, circulation, blood loss and laboratory tests.
to 322ml). Juelsgaard 1998 reported statistically non-significantly
lower mean values of blood volume transfused over the opera- Bredahl 1991, who recorded body temperatures of 30 patients,
tive and peri-operative period for the regional anaesthesia group concluded that temperature changes during the peri-operative pe-
(237ml versus 257ml). Bigler 1985 reported the mean falls in riod were unrelated to the type of anaesthesia.
Adams 1990 reported raised serum adrenaline and noradrenaline those listed as complications, or not wholly associated with deaths,
levels at the end of the operation for a sub-group of 32 patients, the in trial reports.
rise in levels being greater in those patients who received a general
anaesthetic. Tasker 1983, in a study of 100 patients, reported a Pneumonia
significantly greater increase in plasma noradrenaline and cortisol Pneumonia or ’chest infection’ was reported in nine studies (Adams
levels after general anaesthesia in comparison with regional anaes- 1990; Berggren 1987; Bigler 1985; Davis 1981; Davis 1987; Juels-
thesia. There was no report of intra-operative cardiac arrhythmias. gaard 1998; McKenzie 1984; McLaren 1978; Racle 1986). Pool-
ing of the results indicates no clear difference between the two
Juelsgaard 1998 reported a significant increase in the overall num- anaesthetic methods (29/554 (5.2%) versus 31/581 (5.3%), RR
ber of ST segment depressions for those in the spinal anaesthesia 0.99, 95% CI 0.62 to 1.59).
group (125 versus 16 events).
Myocardial infarction
Bigler 1985 reported a significant reduction in the mean time from This complication was reported in six studies (Couderc 1977;
surgery to ambulation of 3.3 days after regional anaesthesia versus Davis 1981; Davis 1987; Juelsgaard 1998; McKenzie 1984; Racle
5.1 days after general anaesthesia. Valentin 1986 however reported 1986). Summation of the results from five trials showed a non
no difference in the time to ambulation for patients in the two statistically significant reduction in myocardial infarction in the
groups. regional anaesthesia group (5/446 (1.1%) versus 8/471 (1.7%),
Post-operative scores in Ungemach 1993 were reported as ’better’ RR 0.70, 95% CI 0.26 to 1.85).
in the spinal group, but it was not clear by how much and how Cerebrovascular accident
this was manifested. This complication was reported in seven studies (Berggren 1987;
b) Complications specific to the method of treatment Bigler 1985; Couderc 1977; Davis 1981; Davis 1987; McKenzie
1984; Racle 1986). Pooling of results demonstrated a tendency to
Davis 1981 was the only study to report on aspiration pneumo- a lower risk with general anaesthesia but the difference was not
nia, with two cases in the general anaesthesia group. These have statistically significant (10/529 (1.9%) versus 6/556 (1.1%), RR
been included under the complication of pneumonia. A persistent 1.51, 95% CI 0.64 to 3.57).
headache, lasting three days, in one person in the spinal anaesthe-
sia group was noted in Bigler 1985. McLaren 1978 reported that Congestive cardiac failure
there were no post-anaesthetic headaches. There was no mention This complication was reported in seven studies (Adams 1990;
within the included studies of other complications such as damage Berggren 1987; Bigler 1985; Davis 1981; Davis 1987; Juelsgaard
to the upper airways or mouth from general anaesthesia. 1998; Racle 1986). Pooling of data gave similar results for both
groups (12/454 (2.6%) versus 12/477 (2.5%), RR 1.05, 95% CI
Failure of spinal anaesthesia, usually resulting in the secondary
0.49 to 2.23).
use of general anaesthesia, was reported in both studies conducted
by Davis et al (Davis 1981; Davis 1987). Spinal anaesthesia, of- Renal failure
ten performed by junior staff, was unsuccessful in eight out of 72 Renal failure was reported in four studies (Adams 1990; Davis
patients (11.1%) in Davis 1981 and in 30 out of 259 patients 1981, Davis 1987; Racle 1986). Summation of results in the anal-
(11.6%) in Davis 1987. Davis 1987 also referred to a 10% failure ysis table demonstrated no difference between anaesthetic tech-
rate in the study of Valentin 1986. The treatment of these spinal niques (2/382 (0.5%) versus 3/414 (0.7%), RR 0.86, 95% CI
anaesthesia failures in the analyses presented by these three tri- 0.22 to 3.41).
als has further implications regarding intention to treat analysis.
For instance, it may be that the excluded patients had different Post operative cardiac arrhythmia
characteristics and outcomes than those patients in which spinal More abnormal cardiac rhythms were detected in the general
anaesthesia was successful. The eight patients in Davis 1981 were anaesthesia group in Couderc 1977. However, Couderc 1977
incorrectly analysed in the general anaesthesia group, whereas the reported that there was no difference in the overall electrocar-
30 patients in Davis 1987 were analysed in the spinal anaesthesia diographic results; these included results for other peri-operative
group, and lastly, Valentin 1986 chose to exclude them from the changes in the cardiogram.
analysis. Acute confusional state
c) General post-operative complications This complication was reported in three small studies (Berggren
1987; Bigler 1985; Racle 1986). Summation of the limited results
Data for most of the life threatening complications such as pneu- showed a non statistically significant reduction in the regional
monia, myocardial infarction, cerebral vascular accident, conges- anaesthesia group (10/83 (12.0%) versus 19/84 (22.6%), RR 0.53,
tive cardiac failure and pulmonary embolism were only available 95% CI 0.27 to 1.07).
as causes for deaths in many of the trial reports. To reflect this,
the data from fatal events have been sub-grouped separately from Urine retention
Pooling of the data from the two studies (Berggren 1987; Bigler fatal pulmonary embolism, shows a contrasting and unexplained
1985) reporting this complication showed similar results for the picture for these two outcomes (test for interaction, based on Peto
two anaesthetic techniques (10/48 (20.8%) versus 10/49 (20.4%), odds ratio results: P = 0.004), where there is less fatal but more
RR 1.02, 95% CI 0.47 to 2.23). non-fatal pulmonary embolism in the regional anaesthesia group.
Post-operative vomiting Composite outcome

Pooling of the data from the two studies (Bigler 1985; McLaren Ungemach 1993 used a scoring system which included compli-
1978) reporting this complication again showed similar results for cations such as heart failure, thrombosis and apoplexy, as well as
the two anaesthetic techniques (2/46 (4.3%) versus 3/49 (6.1%), cardiopulmonary evaluation and laboratory tests. No difference
RR 0.70, 95% CI 0.12 to 3.94). between the two groups was found in the scores at two weeks.
Deep vein thrombosis d) Final outcome measures

Deep vein thrombosis was the primary outcome for one study Mortality has already been considered above.
(Brichant 1995), and for two subgroups of patients from a further
two studies (Davis 1981; McKenzie 1984). Awareness of the risk Changes in mental function
of deep vein thrombosis was evident in several other studies who Two studies (Bigler 1985; Maurette 1988) reported on long term
did not report this outcome, with various prophylactic interven- changes in mental function. Bigler 1985 reported that there was
tions being deployed: Dextran 70 (Berggren 1987); early mobilisa- no persistent impairment in mental function, and no significant
tion (Bigler 1985); anti-vitamin K and early mobilisation (Coud- differences between the two groups in the mental scores achieved at
erc 1977); heparin and active movement (Racle 1986) and anti- three months. Maurette 1988 performed psychological evaluations
embolic stockings (Valentin 1986). Patients in Brichant 1995 also on 33 patients and found no significant difference relating to the
received thromboembolic prophylaxis with low molecular weight type of anaesthesia.
heparin and anti-embolism stockings. Venography screening was
Functional outcome
used to detect deep vein thrombosis in two studies (Brichant 1995;
No study reported on the difference in functional outcomes be-
McKenzie 1984) and fibrinogen scanning in Davis 1987. Pooled
tween groups. Only McKenzie 1984 provided limited data on the
data, grouped by method of diagnosis, include two deaths whose
location of patients at 12 months, but not for the return of patients
underlying cause was deep vein thrombosis from McLaren 1978.
to their previous residence.
Significantly fewer thromboses were detected in patients in the re-
gional anaesthesia group (39/129 (30%) versus 61/130(47%); RR ’LIGHT’ GENERAL ANAESTHESIA COMBINED WITH
0.64, 95% CI 0.48 to 0.86). Though the difference in incidence SPINAL ANAESTHESIA VERSUS GENERAL ANAESTHE-
rates was consistent between trials, whether measured by venog- SIA
raphy, fibrinogen update or at post-mortem, these results have to
The only study (White 1980) in this category involved only 20
be viewed with caution since these were the results of subgroups
patients in each group. No patients died within the one month
of patients for whom data from venography or fibrinogen were
follow-up period of the study. The mean length of operation was
available. In turn, the patients specially monitored for deep vein
58 minutes in both groups. There was no significant difference
thrombosis were also subgroups of the trial populations in two
in the mean post-operative blood oxygen or carbon dioxide levels
studies (Davis 1981; McKenzie 1984).
between the two groups. Complications reported were pneumonia
Pulmonary embolism (4 versus 5 cases), confusional states (3 in each group), deep vein
Pulmonary embolism was reported in ten studies (Adams 1990; thrombosis (1 in the general anaesthesia group) and post-operative
Berggren 1987; Bigler 1985; Brichant 1995; Couderc 1977; Davis vomiting (1 in each group). Results for most of these outcomes
1981; Davis 1987; McKenzie 1984; McLaren 1978; Racle 1986) are presented in the analysis tables.
but mostly as a reason for death rather than through active moni-
REGIONAL (SPINAL) ANAESTHESIA VERSUS LOCAL
toring for non-fatal pulmonary embolism. Pooling the results from
NERVE BLOCKS
nine studies using Peto odds ratios showed statistically significant
heterogeneity (chi-square = 14.85, P = 0.06). Summation of re- Two studies, involving 79 patients, were included. One study (Ey-
sults from nine studies using the random effects model to allow rolle 1998) compared spinal anaesthesia with a lumbar plexus
for this heterogeneity showed little difference in overall incidence block in 50 patients; supplementary intravenous propofol seda-
of pulmonary embolism in the two groups (8/575 (1.4%) versus tion was performed when necessary. The other study (de Visme
10/609 (1.6%), RR 0.98, 95% CI 0.37 to 2.64). The source of 2000) compared spinal anaesthesia with a lumbar plexus block in
heterogeneity resides mainly in the significantly different results conjunction with a sacral plexus block and iliac crest block (for
in trials presenting solely results for fatal pulmonary embolism, lateral cutaneous nerve of the thigh). Intravenous alfentanil or
and those presenting results for non-fatal pulmonary embolism. sedatives were also used if necessary. Both studies only reported on
A second analysis, which presents these grouped by fatal and non- outcome during the peri-operative period and did not report on
post-operative complications or mortality. Results where available The mean length of hospital stay for the 39 patients who returned
and appropriate are given in the analysis tables. home was 36 days for the ketamine group against 24 days for the
general anaesthesia group. This difference is statistically signifi-
In Eyrolle 1998, the need for propofol supplementation, of dosage cant and is related to the higher incidence of unsatisfactory surgi-
greater than 1mg/kg/hr, was significantly less common in the cal results in the ketamine group (see below). Although the gen-
spinal group (5/25 versus 19/25). No cases of incomplete or unsat- eral anaesthesia group mobilised more quickly than the ketamine
isfactory anaesthesia in the spinal group were reported in de Visme group, Spreadbury 1980 reported that the differences were not
2000 as opposed to four cases of incomplete anaesthesia and one statistically significant. The proportions of patients who returned
case, requiring repeated sedation that was judged as unsatisfactory, home were similar (19/30 versus 20/30).
in the nerve block group (0/14 versus 5/15). Overall, the need for
supplementary sedation was significantly less in the spinal group Spreadbury 1980 also reported that the numbers of patients who
(5/39 versus 24/40; RR 0.23, 95% CI 0.10 to 0.50). experienced dreams and hallucinations were similar for the two
groups (4 versus 5 patients). They stated however that the dreams
A fall in mean arterial blood pressure of more than 20 per cent were more likely to be unpleasant after general anaesthesia. Spread-
occurred in significantly more patients in the spinal group (18/25 bury 1980 also reported the incidence of unsatisfactory surgical
versus 3/25; RR 6.0, 95% CI 2.02 to 17.83) in Eyrolle 1998. The results, either due to later dislocation of the prosthesis or an un-
mean fall in arterial blood pressure was also significantly greater stable fixation, which subsequently required bed rest or traction.
in the spinal group in de Visme 2000 (mean difference 16mmHg, There were 7/30 (23%) such cases for the ketamine group against
95% CI -1.3 to 30.7mmHg). In both trials, significantly higher 3/30 (10%) for general anaesthesia.
doses of ephedrine were used to stabilise blood pressure in the
spinal group (weighted mean difference 5.96mg, 95% CI 4.46 to
7.45mg).
DISCUSSION
Pain as measured by the visual analogue scale (VAS) was stated
as showing no difference between groups in Eyrolle 1998. Eleven REGIONAL VERSUS GENERAL ANAESTHESIA
patients failed to complete VAS in de Visme 2000, who considered
Many of the studies within this review involved small numbers of
that VAS rating for pain was unsatisfactory when there were cases
patients and reported only a few outcome measures. The trial re-
of “sensorial” deficiency.
ports of all studies indicated a poor level of methodological rigour,
Insertion difficulty was significantly more common in the spinal in particular regarding concealment of allocation, assessor blind-
group in Eyrolle 1998 (10/25 cases versus 3/25). In contrast, the ing and intention to treat analysis. Despite these limitations, there
mean time to administer the spinal was reported as being statisti- is a reasonable agreement between trials for many of the outcome
cally significantly lower in the spinal group in de Visme 2000 (12 measures reported, particularly for mortality. It remains possible
versus 18 minutes; reported p = 0.013). that some of the differences in outcome within the studies could
be related to the differences in the experience, and competence,
Adverse effects, including five cases of urinary retention, were more of the anaesthetists. Inexperience with the anaesthetic techniques
common in the spinal group in Eyrolle 1998 (6/25 versus 1/25; RR could be inferred in some studies. For example, there was a high
6.00, 95% CI 0.78 to 46.29). No adverse effects of the techniques failure rate of spinal anaesthesia, often performed by junior staff,
were reported by de Visme 2000. of over 11 per cent in both Davis 1981 and Davis 1987. However,
Post-operatively, similar numbers of patients had impaired cog- there was no evidence that the seniority of the anaesthetists ap-
nitive function in de Visme 2000 (5/14 versus 6/15); this was plying the different methods of anaesthesia differed in any given
reflected in the comparable mini-mental test scores (mean 15.5 trial.
versus 14.5). Hip fractures occur predominantly in the frail elderly who have
KETAMINE VERSUS GENERAL ANAESTHESIA multiple other medical conditions. The high mortality within this
group of patients often results from these other medical conditions
The only study included in this category (Spreadbury 1980) in- rather than being a direct consequence of the hip fracture and its
volved 60 female patients. The limited results available are sum- treatment. Regional anaesthesia may reduce short-term mortality,
marised in the analysis tables. Data were presented for early deaths yet this finding is borderline in that it is statistically significant
(within 14 days) and late deaths (time unspecified, in hospital). when using the fixed effects model but not with the random effects
These showed no difference in the overall mortality during hospi- model. The three month mortality results retain a potential for
tal stay (9/30 (30%) versus 9/30 (30%)). Data presented for the a reduction in mortality in the regional anaesthesia group; these
complications of myocardial infarction (1 case), congestive car- are consistent with up to a 30 per cent reduction (95% CI 0.71
diac failure (2 cases) and pulmonary embolism (3 cases) were all to 1.21). There is no evidence of substantial differences between
derived from causes of death for the seven early deaths. regional and general anaesthesia in terms of long-term mortality,
although the small numbers of patients with long-term follow- insufficient data to confirm this in this review and the care that
up, available from two trials of poor methodology, means that needs to be taken in the interpretation of surrogate outcomes, such
we cannot exclude clinically relevant differences. It would not be as hypotension, has already been mentioned (see start of Results).
unexpected that changing one aspect of hip fracture treatment
Juelsgaard 1998 specifically targeted patients with known coronary
(the type of anaesthesia) did not substantially affect long-term
artery disease. Whilst appropriate, the numbers of patients in the
mortality. Potentially, regional anaesthesia could enable the group
trial were too small to determine which type of anaesthesia is best
of very frail elderly to survive the initial surgery, only for death to
for this specific patient population.
ensue later from other medical complications.
Because of the low incidence of many of the complications fol- No attempt was made at a cost evaluation within any of the trials.
lowing surgery, no individual study had numbers large enough to Regional anaesthesia is cheaper with respect to drug costs incurred
during the administration of the anaesthetic, but the time taken
determine if any difference exists. As much of the data for many of
for a regional anaesthesia is slightly longer.
these complications was for fatal complications, these results are
far from complete. Some possible, although unconfirmed, trends In their comprehensive review of regional anaesthesia, Rodgers et
for regional anaesthesia were for less myocardial infarction, more at (Rodgers 2000) found that post-operative mortality up to 30
cerebrovascular accidents and less fatal pulmonary embolism but days was significantly reduced for all types of surgery (general,
more non-fatal pulmonary embolism. Although there were fewer orthopaedic, urological and vascular) and concluded that their
cases of acute confusion when regional anaesthesia was used, more findings supported a “more widespread use of neuraxial blockade
evidence is required to draw valid conclusions. [spinal / epidural anaesthesia]”. It was notable that over half the
Pooled results for deep vein thrombosis showed a statistically sig- trials with at least 10 deaths per trial involved patients with hip
nificant reduction in the incidence of deep vein thrombosis in the fracture; thus enhancing the contribution of the findings of these
regional anaesthesia group. This should not be considered con- trials to the overall result. Rodgers 2000 considered that a lack
clusive as the data were from subgroups of patients who had been of statistical power in individual trials and meta-analyses could
’selected’ by their compliance with a method of diagnosis, and be the principal reason for a conclusion that “neuraxial blockade
thus the effect, and certainly the effect size, may have been dis- had no important effect on mortality”. In fact, our conclusions
torted. The effects of thromboembolic prophylaxis may also af- are phrased in a more tentative way than Rodgers et al imply and,
fect the incidence of thromboembolic complications. The routine although there is a lack of statistical power in our review, we also
use of thromboembolic prophylaxis was mentioned in six stud- consider that there is an important lack of longer term outcome
ies (Berggren 1987; Bigler 1985; Brichant 1995; Couderc 1977; data. Like Rodgers 2000 we consider further research is warranted.
Racle 1986; Valentin 1986). It is also possible that thromboem- ’LIGHT’ GENERAL ANAESTHESIA COMBINED WITH
bolic prophylaxis may have been withheld in those receiving re- SPINAL ANAESTHESIA VERSUS GENERAL ANAESTHE-
gional anaesthesia in some studies. The results do suggest a trend SIA
towards a reduced risk of thromboembolic complications with re-
gional anaesthesia but, because of the small number of trials that The sole study to address this question (White 1980) involved
reported this outcome and the heterogeneity of results, firm con- only 20 patients in each group. There was no difference between
clusions cannot be made for this outcome. techniques for any of the outcome measures reported. Because of
the small numbers of patients involved, no conclusions about the
As would be expected from clinical practice, operations with re- lack of difference between the two techniques can be made.
gional anaesthesia were found to take approximately 5 to 10 min-
utes longer than general anaesthesia. This may be due to the time REGIONAL (SPINAL) ANAESTHESIA VERSUS LOCAL
taken to administer the regional anaesthesia and then the time NERVE BLOCKS
taken for the analgesic effect to occur. Regional anaesthesia results The two included trials (de Visme 2000; Eyrolle 1998) involved
in vasodilatation of the lower limbs and this results in an increased only 79 patients in total. In addition there was incomplete report-
tendency to operative hypotension, as demonstrated by the results.
ing of outcomes and no follow-up of patients. The limited results
In addition, the increased blood flow to the lower limbs with alter-
available suggest that the local nerve blocks are associated with a
ations in coagulability and viscosity of the blood, may be the rea- reduced risk of operative hypotension but have a greater risk of
son for the reduced incidence of venous thrombosis. It is possible incomplete or unsatisfactory analgesia. Because of the limited in-
that the benefits of the reduced thromboembolic complications formation, no conclusions can be made on the use of nerve blocks
may be negated if thromboembolic prophylaxis is used. compared with spinal anaesthesia.
There was a tendency for more hypotension with regional anaes-
KETAMINE VERSUS GENERAL ANAESTHESIA
thesia. This may result in a predisposition to an increased inci-
dence of cerebrovascular complications as hypotension is one of The sole trial (Spreadbury 1980) comparing ketamine with gen-
the aetiological factors for this complication. However, there are eral anaesthesia involved only 60 patients. The only key difference
was a reduction in the 14-day mortality for ketamine, which re- This review was first updated in Issue 4, 2000. The trial search was
lated to a reduction in the risk of early fatal thromboembolic com- updated to August 1999 and one small trial (Juelsgaard 1999) was
plications. However, this difference in mortality did not persist, included. A consumer synopsis was added and relative risks instead
and the mortality during hospital stay was equal in both groups. of Peto odds ratios were presented for dichotomous outcomes.
The numbers of patients were too small to show if the increase in There were no significant changes to the conclusions of the review.
’unsatisfactory surgical results’ in the ketamine group was a signif-
The second update appeared in Issue 4, 2001. This included one
icant factor of ketamine use.
trial (Ungemach 1993) comparing general versus spinal anaesthe-
sia, and two trials (Eyrolle 1998; de Visme 2000) which compared
spinal anaesthesia with lumbar plexus blocks. There were no sig-
REVIEWERS’ CONCLUSIONS
nificant changes to the conclusions of the review.
Implications for practice
POTENTIAL CONFLICT OF
Both regional and general anaesthesia produce comparable results
INTEREST
and therefore the anaesthetists should choose which technique is
most appropriate for each individual patient. Regional anaesthesia
None known.
may be preferable for those patients at high risk for thromboem-
bolic complications.
Due to the limited data available, it is not possible to determine the ACKNOWLEDGEMENTS
roles of nerve blocks, ketamine or spinal anaesthesia with ’light’
general anaesthesia for hip fracture anaesthesia. We would like to thank the following for useful comments from
editorial review of the original review: Gordon Drummond (De-
Implications for research partment of Anaesthetics, University of Edinburgh), William
Well designed randomised trials, with active follow-up of at least Gillespie, Rajan Madhok, Gordon Murray, Tom Pedersen (De-
six months, of regional versus general anaesthesia involving large partment of Anaesthesiology, Copenhagen University Hospital)
numbers of patients and which record, at minimum, the primary and Marc Swiontkowski. We thank William Gillespie, Leeann
clinical outcomes of death, post-operative complications, and long Morton and Lesley Gillespie for their help with the first update.
term outcomes, would help clarify the relative merits of regional For this update, we are indebted to Lesley Gillespie, William Gille-
and general anaesthesia. Large trials with sub-group analysis may spie, Peter Herbison, Leeann Morton, Tom Pedersen, Janet Wale
be able to determine if patients with specific medical conditions and Tony Wildsmith for their assistance and helpful feedback at
(such as cardiac disease, previous stroke) are better managed with editorial review.
one of these two forms of anaesthesia.
SOURCES OF SUPPORT
NOTES
External sources of support
This review and first update was published under the title: “Gen- • Chief Scientist Office, Department of Health, The Scottish
eral versus spinal/epidural anaesthesia for surgery for hip fractures Office UK
in adults”. The title was changed in the second update to reflect
an expansion in the scope of the review to include comparisons of Internal sources of support
all forms of anaesthesia. • No sources of support supplied

REFERENCES
References to studies included in this review elderly. British Journal of Anaesthesia 1987;59:1080-8. [MedLine:
Adams 1990 {published data only} 1988024611].
Adams HA, Wolf C, Michaelis G, Hempelmann G. Postoper- de Visme 2000 {published data only}
ative course and endocrine stress response of geriatric patients de Visme V, Picard F, Le Jouan R, Legrand A, Savry C, Morin V.
with fractured neck of femur [Postoperativer verlauf und en- Combined lumbar and sacral plexus block compared with plain bupi-
dokrine streb-reaktion geriatrischer patienten mit huftnahen frak- vacaine spinal anesthesia for hip fractures in the elderly. Regional
turen; prospektiv-randomisierte studie zum vergleich von spinalanas- Anesthesia and Pain Medicine 2000;25(2):158-62.
thesin und halothan-intubatinosnarkosen]. Anasthesie, Intensivthera-
Eyrolle 1998 {published data only}
pie, Notfallmedizin 1990;25:263-70. [MedLine: 1991023367].
Eyrolle L, Zetlaoui P, Belbachir A, Rosencher N, Conseiller C. Re-
Berggren 1987 {published data only} gional anaesthesia for femoral neck fracture surgery: comparison of
Berggren D, Gustafson Y, Eriksson B, Bucht G, Hansson L-H, Reiz lumbar plexus block and spinal anaesthesia [Abstract]. British Journal
S, Winblad B. Postoperative confusion after anesthesia in elderly of Anaesthesia 1998;80 Suppl 1:112.
patients with femoral neck fractures. Anesthesia and Analgesia 1987;
Juelsgaard 1998 {published data only}
66:497-504. [MedLine: 1987211148].
Juelsgaard P, Sand NPR, Felsby S, Dalsgaard J, Jakobsen KB, Brink
Bigler 1985 {published data only} O, et al. Perioperative myocardial ischaemia in patients undergoing
Bigler D, Adelhoj B, Petring OU, Pederson NO, Busch P, Kalhke surgery for fractured hip randomized to incremental spinal, single-
P. Mental function and morbidity after acute hip surgery during dose spinal or general anaesthesia. European Journal of Anaesthesiology
spinal and general anaesthesia. Anaesthesia 1985;40:672-6. [Med- 1998;15(6):656-63.
Line: 1985277121].
Maurette 1988 {published data only}
Bredahl 1991 {published data only} Maurette P, Castagnera L, Vivier C, Erny P. Comparative repercus-
Bredahl C, Hindsholm KB, Frandsen PC. Changes in body heat dur- sions of general and spinal anesthesia on psychological functions of
ing hip fracture surgery: a comparison of spinal analgesia and gen- the aged subject [Repercussions comparees de l’anesthesie generale et
eral anaesthesia. Acta Anaesthesiologica Scandinavica 1991;35:548- de la rachianesthesie sur les fonctions psychiques du sujet age]. An-
52. [MedLine: 1991377412]. nales Francaises d’ Anesthesie et de Reanimation 1988;7:305-8. [Med-
Brichant 1995 {published data only} Line: 89075140].
Brichant JF, Blom-Peters L, Buffels R, Lamy M. Central neural block- McKenzie 1984 {published and unpublished data}
age failed to decrease deep venous thrombosis in patients undergoing McKenzie PJ, Wishard HY. Anaesthesia for fractured neck of femur
hip surgery and receiving low molecular weight heparin. [Abstract]. (letter). BMJ 1981;282:399-400. [MedLine: 81111251].
British Journal of Anaesthesia 1995;74 Suppl 1:75.
McKenzie PJ, Wishart HY, Dewar KMS, Gray I, Smith G. Compar-
Brown 1994 {published data only}
ison of the effects of spinal anaesthesia and general anaesthesia on
Brown AG, Visram AR, Jones RDM, Irwins MG, Bacon-Shone
postoperative oxygenation and perioperative mortality. British Jour-
J. Preoperative and postoperative oxygen saturation in the el-
nal of Anaesthesia 1980;52:49-53. [MedLine: 80198011].
derly following spinal or general anaesthesia - an audit of current
practice. Anaesthesia and Intensive Care 1994;22:150-4. [MedLine: McKenzie PJ, Wishart HY, Gray I, Smith G. Effects of anaesthetic
1994270545]. technique on deep vein thrombosis: a comparison of subarachnoid
Couderc 1977 {published data only} and general anaesthesia. British Journal of Anaesthesia 1985;57:853-
Couderc E, Mauge F, Duvaldestin P, Desmonts J-M. [Compara- 7. [MedLine: 1985280155].
tive results of general and peridural anesthesia for hip surgery in ∗
McKenzie PJ, Wishart HY, Smith G. Long-term outcome after re-
the very old patient]. [French] [Resultats comparatifs de l’anesthesie pair of fractured neck of femur; comparison of subarachnoid and gen-
generale et peridurale chez le grand vieillard dans la chirurgie de eral anaesthesia. British Journal of Anaesthesia 1984;56:581-4. [Med-
la hanche]. Anesthesie, Analgesie, Reanimation 1977;34(5):987-98. Line: 1984203273].
[MedLine: 78185115].
McLaren 1978 {published data only}
Davis 1981 {published data only}
McLaren AD, Stockwell MC, Reid VT. Anaesthetic techniques for
∗
Davis FM, Laurenson VG. Spinal anaesthesia or general anaesthesia
surgical correction of fractured neck of femur: a comparative study
for emergency hip surgery in elderly patients. Anaesthesia and Inten-
of spinal and general anaesthesia in the elderly. Anaesthesia 1978;33:
sive Care 1981;9:352-8. [MedLine: 1982089249].
10-4. [MedLine: 1978121768].
Davis FM, Quince M, Laurenson VG. Deep vein thrombosis and Racle 1986 {published data only}
anaesthetic technique in emergency hip surgery. BMJ 1980;281: Racle JP, Benkhadra A, Poy JY, Gleizal B, Gaudray A. Compara-
1528-9. tive study of general and spinal anesthesia in elderly women in hip
Davis 1987 {published data only} surgery [Etude comparative de l’anesthesie generale et de la rachi-
Davis FM, Woolner DF, Frampton C, Wilkinson A, Grant A, Har- anestesie chez la femme agee dans la chirurgie de la hanche]. Annales
rison RT, et al. Prospective, multi-centre trial of mortality follow- Francaises d’ Anesthesie et de Reanimation 1986;5:24-30. [MedLine:
ing general or spinal anaesthesia for hip fracture surgery in the 1986213298].
Spreadbury 1980 {published data only} Favarel 1996
Spreadbury TH. Anaesthetic techniques for surgical correction of Favarel Garrigues JF, Sztark F, Petitjean ME, Thicoipe M, Lassie P,
fractured neck of femur: a comparative study of ketamine and re- Dabadie P. Hemodynamic effects of spinal anaesthesia in the elderly:
laxant anaesthesia in elderly women. Anaesthesia 1980;35:208-14. single dose versus titration through a catheter. Anesthesia and Anal-
[MedLine: 1980218116]. gesia 1996;82:312-6.
Tasker 1983 {published data only} Maurette 1993
Tasker TPB, Raitt DG, Kohn RLJ, Vater M, Crawshaw C. Subarach- Maurette P, Bonada G, Djiane V, Erny P. A comparsion between
noid block or general anaesthesia?: a study of the stress response dur- lidocaine alone and lidocaine with meperidine for continous spinal
ing and after surgery for prosthetic replacement of fractured neck of anesthesia. Regional Anesthesia 1993;18:290-5.
femur. [Abstract]. Journal of Bone and Joint Surgery. British Volume Owen 1982
1983;65:660. Owen H, Hutton P. Doxapram and the fractured femur. Anaesthesia
Ungemach 1993 {published data only} 1982;37:301-4.
Ungemach JW, Andres FJ, Eggert E, Schoder K. The role of anaes- Sinclair 1997
thesia in geriatric patients with hip fractures: A prospective study. Sinclair S, James S, Singer M. Intraoperative intravascular volume op-
European Journal of Anaesthesiology. 1993;10(5):380. timisation and length of hospital stay after repair of proximal femoral
Valentin 1986 {published data only} fracture: randomised controlled trial. BMJ 1997;315:909-12.
Valentin N, Lomholt B, Jensen JS, Hejgaard N, Kreiner S. Spinal Sutcliffe 1994
or general anaesthesia for surgery of the fractured hip? A prospective Sutcliffe AJ, Parker MJ. Mortality after spinal and general anaesthesia
study of mortality in 578 patients. British Journal of Anaesthesia 1986; for surgical fixation of hip fractures. Anaesthesia 1994;49:237-40.
58:284-91. [MedLine: 1986131270].
Tonczar 1981
White 1980 {published data only} Tonczar L, Hammerle AF. The impairment of stress parameters by hip
White IW, Chappell WA. Anaesthesia for surgical correction of frac- joint close operations and the influence of anaesthesia. Preliminary
tured femoral neck: a comparison of three techniques. Anaesthesia results of a prospective study (author’s translation) [Auswirkungen
1980;35:1107-10. [MedLine: 81083959]. huftgelenksnaher operationen auf das verhalten von stressparametern
und ihre beeinflussung durch anasthesie. Vorlaufige ergebnisse einer
References to studies excluded from this review prospektiven studie]. Unfallchirurgie 1981;7(3):138-41.
Barna 1981
Ungemach 1987
Barna B. Comparison of spinal and general anesthesia in the surgical
Ungemach JW. Inhalation anesthesia or “balanced anesthesia”? A
treatment of hip fractures [A spinalis es az altalanos anaesthesia ossze-
comparative perioperative study in geriatric patients [Inhalation-
hasonlitasa csipotaji toresek multejeinek erzesteleniteseben]. Orvosi
sanaesthesie oder “ balancierte anaesthesie ”?: Eine vergleichende pe-
Hetilap 1981;122:1135-8.
rioperative studie geriatrischer patienten]. Anaesthesist 1987;36:288-
Coleman 1988 91.
Coleman SA, Boyce WJ, Cosh PH, McKenzie PJ. Outcome after gen- Van Gessel 1989
eral anaesthesia for repair of fractured neck of femur: a randomised Van Gessel EF, Forster A, Gamulin Z. Surgical repair of hip fractures
trial of spontaneous v. controlled ventilation. British Journal of Anaes- using continous spinal anaesthesia: comparison of hypobaric solu-
thesia 1988;60:43-7. tions of tetracaine and bupivaciane. Anesthesia and Analgesia 1989;
Critchley 1995 68:276-81.
Critchley LA, Stuart JC, Conway F, Short TG. Hypotension during Wickstrom 1982
subarachnoid anaesthesia: haemodyamic effects of ephedrine. British Wickstrom I, Holmberg I, Stefansson T. Survival of female geriatric
Journal of Anaesthesia 1995;74:373-8. patients after hip fracture surgery. A comparison of 5 anesthetic meth-
ods. Acta Anaesthesiologica Scandinavica 1982;26:607-14.
Darling 1994
Darling JR, Murray JM, Hainsworth AM, Trinick TR. The effect of References to studies awaiting assessment
isoflurane or spinal anesthesia on Indocyanine green disappearance
rate in the elderly. Anesthesia and Analgesia 1994;78:706-9. Wajima 1995
Wajima Z, Kurosawa H, Inoue T, Yoshikawa T, Ishikawa G, Shitara
Dyson 1988 T, et al. Changes in dementia rating scale scores of elderly patients
Dyson A, Henderson AM, Chamley D, Campbell ID. An assessment with femoral neck fracture during perioperative period [Original in
of postoperative oxygen therapy in patients with fractured neck of Japanese]. Masui 1995;44(11):1489-97.
femur. Anaesthesia and Intensive Care 1988;16:405-10.
El-Zahaar 1995 Additional references
El-Zahaar MS, Al-Kawally HM, Said AS. A double-blind random- Clarke 2000
ized study of the effects of torniquet use and type of anesthetic tech- Clarke M, Oxman AD, editors. Assessment of study quality.
niques on the incidence of deep vein thrombosis (DVT) in orthope- Cochrane Reviewers’ Handbook 4.1 [updated June 2000]; Section
dic surgery. Journal of Neurological & Orthopaedic Medicine & Surgery 6. In: Review Manager (RevMan) [Computer program] Version 4.1.
1995;16(2):70-4. Oxford, England: The Cochrane Collaboration 2000.
Covert 1989 Rodgers 2000
Covert CR, Fox GS. Anaesthesia for hip surgery in the elderly. Cana- Rodgers A, Walker N, Schug S, McKee A, Kehlet H, van Zundert
dian Journal of Anaesthesia 1989;36:311-9. A, et al. Reduction of postoperative mortality and morbidity with
Dickersin 1994 epidural or spinal anaesthesia: results from overview of randomised
Dickersin K, Schere R, Lefebvre C. Identifying relevant studies for trials. British Medical Journal 2000;321:1493-7.
systematic reviews. BMJ 1994;309:1286-91. Sorensen 1992
Mansour 1993 Sorenson RM, Pace NL. Anesthetic techniques during surgical repair
Mansour NY. Reevaluating the sciatic nerve block: another landmark of femoral neck fractures. A meta-analysis. Anesthesiology 1992;77:
for consideration. Regional Anesthesia 1993;18:322-3. 1095-104.
Melton 1993 WHO study group 1994

Melton LJ III. Hip fractures: a worldwide problem today and tomor- WHO study group. Assessment of fracture risk and its application
row. Bone 1993;14 Suppl 1:S1-8. to screening for postmenopausal osteoporosis. WHO; 1994 WHO
technical report series no.: 843
Modig 1983
Modig J, Borg T, Bagge L, Saldeen T. Role of extradural and of general Winnie 1974
anaesthesia in fibrinolysis and coagulation after total hip replacement. Winnie AP, Ramamurthy S, Durrani Z, Radonjic R. Plexus blocks
British Journal of Anaesthesia 1983;55:625. for lower extremity surgery. Anesthesiology Reviews 1974;1:11-6.
Parker 1993 References to other published versions of this review

Parker MJ, Pryor GA. Hip fracture management. Oxford: Blackwell
Urwin 2000
Scientific Publications, 1993.
Urwin SC, Parker MJ, Griffiths R. General versus regional anaesthesia
Parker 2001 for hip fracture surgery: a meta-analysis of randomized trials. British
Parker MJ, Griffiths R, Appadu BN. Nerve blocks (subcostal, lat- Journal of Anaesthesia 2000;84(4):450-455.
eral cutaneous, femoral, triple, psoas) for hip fractures (Cochrane
Review). In: The Cochrane Library, 3, 2001.Oxford: Update
Software.10.1002/14651858.CD001159 ∗
Indicates the major publication for the study
TABLES
Characteristics of included studies
Study Adams 1990

Methods Quasi-randomised trial: by the date of operation
Methodological score: 2
Participants Orthopaedic hospital in Gieben, Germany
56 patients with a proximal femoral fracture.
Mean age 79/81 years (range 63-91).
Male: 18%
Number lost to follow-up: not stated
Interventions Spinal anaesthesia using 0.5% bupivacaine and 4% mepivacaine
versus
General anaesthesia using thiopentone, halothane, nitrous oxide/oxygen, vencuronium, succinycholine, at-
ropine
Outcomes Length of follow-up: period of hospital stay
Mortality - during hospital stay
Length of operation
Hypotension
Operative blood loss
Transfusion requirements
Blood levels of catecholamines, ADH and adrenalin (see notes)
Characteristics of included studies (Continued )
Pneumonia (f )
Congestive cardiac failure (f )
Renal failure
Pulmonary embolism (f )
Notes Published in German
Abstract and diagrams are contradictory for endocrine (ADH, adrenalin) results
Allocation concealment C
Study Berggren 1987

Methods Randomised trial: method not stated
Participants Orthopaedic hospital in Umea, Sweden.
57 patients with a femoral neck fracture
Mean age 77/78 years (range 65-92 years).
Male: 19%
Number lost to follow-up: 4 (7%)
Interventions Both groups premedicated with pethidine 25-50mg.
Spinal anaesthesia with 2% prilocaine in the epidural space, mean volume used 12.5ml
versus
General anaesthesia with thipopentone 3-4mg/kg, atropine 0.25-0.5mg iv, suxemethonium, ventilated with
nitrous oxide and oxygen and halothane and suxamethonium infusion.
Outcomes Length of follow-up: 12 months
Mortality - 1 year (see notes)
Length of operation
Operative hypotension
Intraoperative blood loss (not split by treatment groups)
Hypoxaemia
Pneumonia
Cerebrovascular accident
Congestive cardiac failure
Confusional state
Urine rention
Urinary tract infection
Pulmonary embolism
Total medical complications
Notes 4 died by 1 year, 1 in the epidural group on 1st post-op day, the other 3 (group not given) by 5 months.
Patients were interviewed at 6 and 12 months regarding living conditions and walking ability - data not
presented.
Allocation concealment B
Study Bigler 1985

Participants Place and country of study not stated
40 patients with a proximal femoral fracture
Mean age 79 years.
Male: 17.5%
Loss to follow-up: not known

Interventions Spinal anaesthesia with 3ml of 0.75% bupivacaine
versus
General anaesthesia using atropine, thoiopentane, fentanyl, pancuronium, nitrous oxide/oxygen
Mortality - early
Length of operation
Hypotension (maximum drop in systolic blood pressure)
Fall in haemaglobin
Pneumonia
Confusional state
Urine rention
Post-operative vomiting
Pulmonary embolism
Time till ambulation
Mental function
Headache
Notes
Study Bredahl 1991

Participants Orthopaedic hospital Aalborg, Denmark
30 female patients with a proximal femoral fracture
Mean age 79 years (range 60-90).
Male: 0%
Loss to follow-up: not stated, but 2 excluded due to incomplete data.
Interventions Spinal anaesthesia with 2.5-3ml of 0.5% bupivacaine
versus
General anaesthesia using thoiopentane, pethidine, pancuronium, nitrous oxide/oxygen, IPPV
Outcomes Length of follow-up: 3 days
Length of operation
Change in body temperature (up to 3 hours)
Notes
Study Brichant 1995

Participants Orthopaedic hospital in Brussels, Belgium
106 patients with proximal femoral fracture
Age: not stated.
Male: % not stated
Interventions Spinal (subarachroid or epidural) anaesthesia with bupivacaine
versus
General anaesthesia administered according to ’local practice’
Deep vein thrombosis (venography)
Pulmonary embolism
Haemorrhagic complications
Thrombocytopenia
Notes Conference abstract only
All patients had subcutaneous nadroparin for DVT prophylaxis
Study Brown 1994

Methods Randomised trial: use of random numbers table
Participants Orthopaedic hospital in Hong Kong
Male: 50%
Interventions Spinal (subarachnoid) anaesthesia with 0.2mg/kg 0.5% bupivacaine
versus
General anaesthesia using thiopentone or propofol, isoflurane or enflurane and pre-medication with pethidine
or temazepam
Outcomes Length of follow-up: 2 days (up to 44 hours)
Hypotension
Oxygen saturation
Notes
Study Couderc 1977

Methods Randomised study: by ’drawing of lots’
Methodogical score: 4
Participants Orthopaedic hospital in Paris, France
Mean age 86 years. (Inclusion criterion: 80+ years; range not stated).
Male: 14%
Interventions Spinal anaesthesia with 0.5% bupivacaine and adrenaline
versus
General anaesthesia with thiopentone, pancuronium, dextromoramide or methoxyflurane, nitrous ox-
ide/oxygen
Mortality - 11 days, 3 months
Hypotension
Oxygenation and carbon dioxide levels
Myocardial infarction (f )
Cerebrovascular accident (f )
Notes In French
Complete data for fatal myocardial infarction, congestive heart failure and pulmonary embolism not provided.
Study Davis 1981

Participants Orthopaedic hospital Christchurch, New Zealand
Mean age 81/78 years (Inclusion criterion: 50+, range not given).
Male: 15%
Number lost to follow-up: 0
Interventions Spinal anaesthesia using tetracaine 0.5% in 51 patients and 0.5% cinchocaine in 13 patients. Ketamine also
used for sedation in 8 patients and diazapam (mean dose 9mg).
versus
General anaesthesia with diazapam (2.5-30mg) mean dose 9.5mg. Fentanyl 1-3mcg/kg, nitrous oxide and
oxygen, IPPV, pancuronium mean dose 6mg.
Outcomes Length of follow-up : 1 month
Mortality - 1 month
Duration of anaesthesia (Length of operation)
Postoperative blood gases
Hypotension
Fall in haematocrit
Pneumonia (f )
Aspiration pneumonia (f )
Renal failure
Cardiac arrthymias
Deep vein thrombosis (fibrinogen)
Notes 8 failed spinals who had a general anaesthesia were placed in the general anaesthesia group.
Results for DVT were available for 76 out of a sub-group of 90 patients who were monitored using I125
fibrogen scanning
Study Davis 1987

Participants Orthopaedic hospitals in New Zealand - multicentre study
Mean age 79.5 years (range not stated).
Male: 22%
Number lost to follow-up: 0, but 11 excluded
Interventions Spinal anaesthesia with sedation with diazapam. Tetracaine, nupercaine or bupivacaine for spinal
versus
General anaesthesia with pre-oxygenation, iv induction with thiopentone, IPPV maintained with nitrous
oxide/oxygen, non-depolarizing neuromuscular blocker, fentanyl
Outcomes Length of follow-up: 3 to 30 months
Mortality - 1 month, 3 & 6 months (estimated from graph)
Hypotension
Pneumonia (f )
Renal failure (f )
Notes 11.3% of patients originally allocated to spinal anaesthesia were given general anaesthesia due to failed spinals.
These were retained in the spinal group for analysis purposes.
There was 1 non fatal anaphylactoid reaction at induction of general anaesthesia
Study Eyrolle 1998

Participants Orthopaedic hospital in Paris, France
Mean age 82 years (range not stated)
Male: % not stated
Number lost to follow-up: none probably
Interventions Spinal anaesthesia with 0.5% bupivacaine
versus
lumber plexus block using 2% lidocaine, 0.5% bupivacaine with 1:200,000 epinephrine.
A light sedation with propofol intravenously, as required.
Outcomes Length of follow-up: not stated
Ease of insertion
Hypotension
Use of propofol during surgery (associated with discomfort)
Use of epinephrine during surgery
Post-operative cognitive function
Pain levels post-operatively Adverse effects (including urinary retention)
Study Juelsgaard 1998

Participants Orthopaedic hospital in Aarhus, Denmark
29 followed-up out of 54 patients with proximal femoral fracture and known coronary artery disease
For 29 patients included in this review:
Age: mean 80.9 years (range 65-99)
Male: 13%
Number lost to follow-up: 0, but 11 excluded from original trial population

Interventions Spinal anaesthesia with 2.5ml of 0.5% bupivacaine in the subarachnoid space
versus
General anaesthesia with fentanyl 1-2mcg/kg, 1-4mg/kg thiopentone, 0.5mg/kg atracurium, nitrous oxide
and oxygen, enflurane.
Outcomes Length of follow-up: 1 month
Mortality - 1 month
Length of operation
Hypotension (33% reduction from baseline)
Peri and post operative blood loss
Pneumonia (f )
ECG analysis
Notes The study also included 14 patients allocated to incremental spinal anaesthesia. These patients have not been
included in this review
Study Maurette 1988

Methods Randomised trial: by ’random draw’
Participants Orthopaedic hospital Bordeaux, France
Mean age 83 years (range not stated).
Male: % not stated
Number lost to follow-up: not stated, but 2 excluded as they failed to participate in post-op tests
Interventions Spinal anaesthesia with 1.5mg/kg prilocaine
versus
General anaesthesia using thiopentone, spontaneous ventilation, nitrous oxide/oxygen, enflurane, dextro-
moramide
Length of operation
Hypotension
Psychological evaluation
Notes In French
Study McKenzie 1984

Methods Randomised trial: use of envelopes containing random numbers Methodological score: 6
Participants Orthopaedic hospital in Glasgow, Scotland
150 patients with fractured neck of femur.
Male: % not stated
Number lost to follow-up: 0, but 2 excluded due to postponement of operation
Interventions Spinal anaesthesia with 0.5% hyperbaric cinchocaine 1.3-1.5ml. Supplemented by small doses of diazapam
if required
versus
General anaesthesia induced with althesin 1-3ml, suxamethonium 50mg, nitrous oxide and oxygen, halothane
and spontaneous respiration.
Mortality - at 1, 3, 6 and 12 months
Length of operation
Pneumonia (f )
Deep vein thrombosis (venography)
Location at 12 months
Notes Additional information supplied by Dr McLaren indicated that all the references refered to one study.
Additional data on mortality supplied.
The venography study for DVT detection involved a subgroup of 40 patients
Allocation concealment A
Study McLaren 1978

Participants Orthopaedic hospital in Glasgow, Scotland
55 patients with fractured neck of femur
Mean age 76 years.
Male: % not stated.
Interventions No premedicaton
Spinal anaesthesia with 0.5ml hyperbaric cinchocaine 0.5%. Patients sedated with 10% Althesin in 5%
dextrose during operation.
versus
General anaesthesia with Althesin 50mcg/kg, Pancuronium bromide 0.1 mg/kg, IPPV, nitrous oxide, oxygen
and Fentanyl 0.05mg as needed.
Outcomes Length of follow-up: 1 month minimum
Mortality - 1 month
Length of operation
Hypotension
Post-operative oxygenation
Blood loss
Pneumonia (respiratory infections)
Vomiting
Deep vein thrombosis (f )
Headache (none)
Notes Addendum in paper indicated that data for a further 20 patients were available - there were 2 more deaths
in the general anaesthesia group
Study Racle 1986

Methods Randomised study: use of random numbers table
Participants Orthopaedic hospital in Cedex, France.
Mean age: 82 years (Inclusion criterion: 75+, range not given).
Male: 0%
Interventions Spinal anaesthesia with 3ml 0.5% Bupivacaine + adrenaline
versus
General anaesthesia using thiopentone, vecuronium, fentanyl, nitrous oxide/oxygen, enflurane
Mortality - 1, 3 months
Length of operation
Hypotension
Pneumonia
Renal failure (f )
Confused state
Pulmonary embolism
Notes In French
Study Spreadbury 1980

Methods Randomised: method not stated
Participants Orthopaedic hospital in Warwick, England
Male: % not stated
Number lost to follow-up: none
Interventions Ketamine anaesthesia using atropine pre-medication, ketamine 2mg/kg at induction then ketamine 1mg/kg
as required
versus
General anaesthesia using premedication of atropine 0.6mg then a general anaesthetic using drugs and
method chosen by the anaesthetist
Mortality - 14 days, during hospital stay
Time to mobilisation
Return of patients back home
Occurrence of dreams or hallucinations after operation
Unsatisfactory surgical results
Notes

Study Tasker 1983

Participants Orthopaedic hospital in Leicester, England
Mean age not stated.
Male: % not stated
Interventions Spinal versus general anaesthesia
Exact method of anaesthesia not stated
Mortality
Plasma catecholamines, cortisol
Study Ungemach 1993

Methods Randomised trial: method not stated, mention of pairs
Participants Orthopaedic hospital in Mannheim, Germany
Male: 16%
Interventions Spinal anaesthesia with 3-4ml of 0.5% hyperbaric bupivacaine
versus
General anaesthesia with isoflurane, fentanyl, nitrous oxide/oxygen
Outcomes Length of follow-up: 2 weeks
Mortality - 2 weeks
Score based on conscious level, respiration, circulation, blood lost and laboratory tests taken at 2 hours.
Score based on lab tests, cardiopulmonary situation and complications (e.g. heart failure, thrombosis and
apoplexy) at 2 weeks post-operatively
Study Valentin 1986

Participants Orthopaedic hospital in Hellerup, Denmark
Mean age 79 years (range 50 - 100).
Male: 20%
Number lost to follow-up: 2 (0.3%), 84 patients excluded
Interventions Spinal anaesthesia with 3-4ml isotonic Bupivacaine and sedation with Fentanyl 0.05-0.1mg IV
versus
General anaesthesia with enflurane and nitrous oxide/oxygen with or without thiopentone at induction or
neurolept anaesthesia with droperidol, fentanyl and nitrous oxide/oxygen
Mortality - 1 month, 3, 6 and 12 months (read from graphs)
Length of operation
Time to ambulation
Notes
Study White 1980

Participants Orthopaedic hospital in Cape Town, South Africa
40 of 60 patients in trial with a proximal femoral fracture.
Male: 8%
Interventions Spinal anaesthesia with 0.6-0.8ml hyperbaric cinchocaine and ’light’ general anaesthesia with althesin, fen-
tanyl, nitrous oxide/oxygen
versus
General anaesthesia with thiopentone, suxamethonium, nitrous oxide/oxygen, halothane, fentanyl
versus
Psoas nerve block with 30ml 2% mepivacaine and ’light’ general anaesthesia with fentanyl and althesin (not
included in review)
Outcomes Length of follow-up: minimum 4 weeks
Mortality - 1 month
Length of operation
Post operative blood gases (oxygen and carbon dioxide)
Pneumonia
Confusional state
Deep vein thrombosis
Vomiting
Notes The 20 Psoas nerve block group patients were not included in this review
Study de Visme 2000

Methods Randomised trial: method by ’hospital pharmacy before transfer to the operating theatre’
Participants Orthopaedic hospital in Brest, France
Male: 17%
Number lost to follow-up: none
Interventions Spinal anaesthesia with sedation using alfentanil and 3ml 0.5% plain bupivacaine for the spinal
versus
lumber plexus, sacral plexus and iliac crest block first with sedation using alfentanil. 30ml and 10ml of 1.33%
lidocaine and epinephrine were used for the lumbar and sacral blocks and 5ml 1% lidocaine for the iliac crest
block (for lateral cutaneous nerve).
Outcomes Length of follow-up: not stated but probably 5 days
Length of operation
Time to perform the anaesthetic
Hypotension
Use of epinephrine during surgery
Post-operative cognitive function
Pain levels in the recovery room
Need for supplemention of analgesia
Notes
(f ) = fatal: outcome such as pneumonia only appears as a reason for death
Characteristics of excluded studies
Study Reason for exclusion

Barna 1981 Translation of the article from Hungarian revealed it is a comparative study of 100 spinal anaesthetics and 100
general anaesthetics for hip fracture patients. The study was excluded as there was no randomisation of patients.
Coleman 1988 A randomised trial of 152 patients comparing general anaesthesia with spontaneous respiration with general anaes-
thesia with controlled ventilation. The study was excluded as it involved a change in the types of drugs used only,
not a change in the method of anaesthesia.
Critchley 1995 A randomised trial of 30 hip fracture patients comparing spinal anaesthesia with ephedrine alone or with ephedrine
and colloid. The trial was excluded as it was not a trial of different types of anaesthesia but a comparison of different
drugs within one form of anaesthesia.
Darling 1994 A randomised trial of 10 patients with spinal anaesthetic and 10 with general anaesthesia to assess the rate of
clearance of a bolus dose of Indocyanine green between the two anaesthetic techniques. There was no difference
in the rate of disappearance of the indocyanine green between the two techniques and no other outcomes were
reported. The study was excluded as it was not felt relevant to this review as no clinical outcomes were reported.
Dyson 1988 A randomised trial of 60 patients which tested the use of postoperative oxygen in two groups that had already been
divided into those receiving general anaesthesia and those receiving spinal anaesthesia. No results were provided
for the anaesthetic comparison save the general statement that there was no statistical difference in mean oxygen
tensions between the two anaesthesia groups. The trial was excluded due to the lack of outcome data for the
anaesthesia comparison.
El-Zahaar 1995 This study was a randomised comparison of general versus epidural anaesthesia in 214 patients undergoing either
hip or femoral surgery (117 patients), or tibial surgery (97 patients). This trial was excluded because separate results
for patients having surgery for a hip fracture were not presented.
Favarel 1996 A randomised trial of 60 hip fracture patients comparing the haemodynamic effects of a single dose of spinal
bupivacaine versus a continuous titrated dose. Outcome measures were the onset of anaesthesia and haemodynamic
variables. The trial was excluded as it was not considered a comparison of different forms of anaesthesia, only of a
modification of anaesthetic technique.
Maurette 1993 A randomised trial of 34 hip fracture patients comparing continous spinal anaesthesia with lidocanine alone versus
lidocaine with meperidine. The trial was excluded as it was a trial of different drugs with the same anaesthetic
technique, not a comparision of different types of anaesthesia.
Owen 1982 A randomised trial of a single dose of doxapram on the post-operative arterial oxygen tension in hip fracture patients.
The trial was excluded as it was not a comparison of anaesthetic techniques.

Characteristics of excluded studies (Continued )
Sinclair 1997 A randomised trial of 40 patients with a hip fracture surgically treated under general anaesthesia. Patients were
randomised to have either conventional intra-operative fluid management or colloid fluid challenges. The study
was excluded as it was not a comparison of different types of anaesthesia.
Sutcliffe 1994 A comparative study of 1333 patients with general versus spinal anaesthesia. The study was excluded as there was
no randomisation of patients.
Tonczar 1981 A randomised trial of 14 patients comparing neuroleptic anaesthesia with spinal anaesthesia. The study was excluded
as it involved a neuroleptic anaesthesia and the only outcome measures were plasma catecholamines, cortisol, blood
pressure and changes in heart rate.
Ungemach 1987 A randomised trial of 50 hip fracture patients using either enflurane or enflurane and fentanyl. The trial was excluded
as it was a comparison of different drugs within one type of anaesthesia (general anaesthesia) and not a comparison
of different anaesthetic techniques.
Van Gessel 1989 A randomised trial of 30 hip fracture patients comparing spinal anaesthesia with either hypobaric tetracaine or
hypobaric bupivacaine. The trial was excluded as it was a not a trial of different types of anaesthesia but a comparison
of different drugs within one form of anaesthesia.
Wickstrom 1982 This was a report of two quasi-randomised trials (based on dates of birth) with a month in-between, reported as one
study. The first study compared epidural versus ketamine intravenous infusion versus neurolept general anaesthesia
in 129 hip fracture patients. The second study compared enflurane general anaesthesia versus halothane general
anaesthesia in 40 hip fracture patients. The first study was excluded as it was considered that neuroleptic anaesthesia
was no longer applicable or relevant for hip fracture surgery. A comparison of non-concurrent treatment groups was
also not considered appropriate. The second study was excluded as it was a comparison of different drugs within
one type of anaesthesia (general anaesthesia) and not a comparison of different anaesthetic techniques.
GRAPHS
Comparison 01. Regional (spinal or epidural) versus general anaesthesia
No. of No. of
Outcome title studies participants Statistical method Effect size
01 Mortality - 1 month 8 1607 Relative Risk (Fixed) 95% CI 0.72 [0.51, 1.00]
02 Mortality - 1 month (random 8 1607 Relative Risk (Random) 95% CI 0.73 [0.47, 1.12]
effects model)
03 Mortality - 3 months 6 1491 Relative Risk (Fixed) 95% CI 0.92 [0.71, 1.21]
06 Mortality - early and up to 1 11 1817 Relative Risk (Fixed) 95% CI 0.76 [0.56, 1.04]
month
07 Length of operation (mins) 6 376 Weighted Mean Difference (Fixed) 95% CI 4.82 [1.08, 8.56]
08 Operative hypotension 8 902 Relative Risk (Fixed) 95% CI 1.31 [1.09, 1.58]
09 Operative hypotension 8 902 Relative Risk (Random) 95% CI 1.18 [0.87, 1.60]
(random effects model)
10 Operative blood loss (mls) 3 308 Weighted Mean Difference (Random) 95% CI -81.24 [-216.01,
53.54]
11 Patients receiving blood 3 228 Relative Risk (Fixed) 95% CI 1.01 [0.82, 1.24]
transfusion
12 Transfusion requirements (mls) 3 203 Weighted Mean Difference (Random) 95% CI 140.69 [-40.33,
321.71]
13 Post-operative hypoxia 1 57 Relative Risk (Fixed) 95% CI 0.74 [0.40, 1.38]
14 Length of hospital stay 2 218 Weighted Mean Difference (Fixed) 95% CI -0.21 [-5.21, 4.78]
15 Pneumonia 9 1125 Relative Risk (Fixed) 95% CI 0.99 [0.62, 1.59]
16 Myocardial infarction 5 917 Relative Risk (Fixed) 95% CI 0.70 [0.26, 1.85]
17 Cerebrovascular accident 7 1085 Relative Risk (Fixed) 95% CI 1.51 [0.64, 3.57]
18 Congestive cardiac failure 7 931 Relative Risk (Fixed) 95% CI 1.05 [0.49, 2.23]
19 Renal failure 4 796 Relative Risk (Fixed) 95% CI 0.86 [0.22, 3.41]
20 Acute confusional state 3 167 Relative Risk (Fixed) 95% CI 0.53 [0.27, 1.07]
21 Urine retention 2 97 Relative Risk (Fixed) 95% CI 1.02 [0.47, 2.23]
22 Vomiting 2 95 Relative Risk (Fixed) 95% CI 0.70 [0.12, 3.94]
23 Deep vein thrombosis 4 259 Relative Risk (Fixed) 95% CI 0.64 [0.48, 0.86]
24 Pulmonary embolism 9 1184 Relative Risk (Fixed) 95% CI 0.90 [0.42, 1.94]
25 Pulmonary embolism (random 9 1184 Relative Risk (Random) 95% CI 0.98 [0.37, 2.64]
effects model)
26 Pulmonary embolism (fatal and Relative Risk (Fixed) 95% CI Subtotals only
non fatal)
Comparison 02. Spinal and ’light’ general anaesthetic versus general anaesthetic
No. of No. of
01 Mortality - 1 month 1 40 Relative Risk (Fixed) 95% CI Not estimable
02 Length of operation 1 40 Weighted Mean Difference (Fixed) 95% CI 0.00 [-14.89, 14.89]
03 Pneumonia 1 40 Relative Risk (Fixed) 95% CI 0.80 [0.25, 2.55]
04 Confusional state 1 40 Relative Risk (Fixed) 95% CI 1.00 [0.23, 4.37]
05 Deep vein thrombosis 1 40 Relative Risk (Fixed) 95% CI 0.33 [0.01, 7.72]
Comparison 03. Regional (spinal or epidural) versus lumbar plexus nerve blocks
No. of No. of
01 Incomplete or unsatisfactory 2 79 Relative Risk (Fixed) 95% CI 0.23 [0.10, 0.50]
analgesia
02 Operative hypotension 1 50 Relative Risk (Fixed) 95% CI 6.00 [2.02, 17.83]
03 Mean fall in arterial blood 1 29 Weighted Mean Difference (Fixed) 95% CI 16.00 [1.31, 30.69]
pressure (mmHg)
04 Mean dose of ephedrine used 2 79 Weighted Mean Difference (Fixed) 95% CI 5.96 [4.46, 7.45]
(mg)
05 Adverse effects 2 79 Relative Risk (Fixed) 95% CI 6.00 [0.78, 46.29]
06 Post-operative confusion 1 29 Relative Risk (Fixed) 95% CI 0.89 [0.35, 2.28]
Comparison 04. Intravenous ketamine versus general anaesthesia
No. of No. of
01 Mortality - during hospital stay 1 60 Relative Risk (Fixed) 95% CI 1.00 [0.46, 2.17]
02 Myocardial infarction 1 60 Relative Risk (Fixed) 95% CI 0.33 [0.01, 7.87]
03 Congestive cardiac failure 1 60 Relative Risk (Fixed) 95% CI 0.20 [0.01, 4.00]
04 Pulmonary embolism 1 60 Relative Risk (Fixed) 95% CI 0.14 [0.01, 2.65]
05 Length of hospital stay 1 39 Weighted Mean Difference (Fixed) 95% CI 12.00 [5.57, 18.43]
(discharge home)
INDEX TERMS
Medical Subject Headings (MeSH)
Adult; ∗ Anesthesia, Conduction; Anesthesia, Epidural; ∗ Anesthesia, General; Anesthesia, Spinal; Clinical Trials; Hip Fractures
[∗ surgery]; Length of Stay; Postoperative Complications; Randomized Controlled Trials
MeSH check words
Human
COVER SHEET
Title Anaesthesia for hip fracture surgery in adults
Authors Parker MJ, Handoll HHG, Griffiths R, Urwin SC
Contribution of author(s) Martyn Parker (MP) initiated the review and wrote the first draft of the protocol. Helen
Handoll (HH) identified the trial studies. Susan Urwin and Richard Griffiths indepen-
dently assessed trial quality and extracted data. The other two reviewers (HH and MP)
independently checked these results and entered the review into RevMan. All reviewers
critically reviewed successive drafts of the review. The updates were compiled by MP and
HH with RG independently extracting data. Susan Urwin was not available to contribute
to or comment on the second update. Martyn Parker is the guarantor of the review.
Issue protocol first published 1997/4
Review first published 1999/4
Date of most recent amendment 06 August 2003
Date of most recent 04 July 2001

SUBSTANTIVE amendment
What’s New The second update, first appearing in Issue 4, 2001, involved an expansion of the scope of
the review to include comparisons of all forms of anaesthesia; as reflected in the changed
review title. Three new trials were included; one comparing general versus spinal anaesthesia
(Ungemach 1993) and two (Eyrolle 1998; de Visme 2000) comparing spinal anaesthesia
with lumbar plexus blocks. Considerations of surrogate outcomes led to a slight amendment
to the conclusions of the review.
Date new studies sought but Information not supplied by author

none found
Date new studies found but not Information not supplied by author
yet included/excluded
Date new studies found and 01 March 2001

included/excluded
Date authors’ conclusions Information not supplied by author

section amended
DOI 10.1002/14651858.CD000521
Cochrane Library number CD000521
Editorial group Cochrane Musculoskeletal Injuries Group
Editorial group code HM-MUSKINJ

GRAPHS AND OTHER TABLES
Comparison 04. 01 Mortality - 1 month
Review: Anaesthesia for hip fracture surgery in adults
Comparison: 01 Regional (spinal or epidural) versus general anaesthesia
Outcome: 01 Mortality - 1 month
Study Regional General Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI
Berggren 1987 1/28 0/29 0.6 3.10 [ 0.13, 73.12 ]
Davis 1981 3/64 9/68 11.4 0.35 [ 0.10, 1.25 ]
Davis 1987 17/259 16/279 20.2 1.14 [ 0.59, 2.22 ]
Juelsgaard 1998 4/15 2/14 2.7 1.87 [ 0.40, 8.65 ]
McKenzie 1984 8/73 13/75 16.8 0.63 [ 0.28, 1.44 ]
McLaren 1978 1/26 9/29 11.1 0.12 [ 0.02, 0.91 ]
Racle 1986 2/35 5/35 6.5 0.40 [ 0.08, 1.93 ]
Valentin 1986 17/281 24/297 30.6 0.75 [ 0.41, 1.36 ]
Total (95% CI) 781 826 100.0 0.72 [ 0.51, 1.00 ]

Total events: 53 (Regional), 78 (General)
Test for heterogeneity chi-square=9.05 df=7 p=0.25 I =22.7%
Test for overall effect z=1.96 p=0.05
0.01 0.1 1 10 100

Favours regional Favours general

Comparison 04. 02 Mortality - 1 month (random effects model)
Outcome: 02 Mortality - 1 month (random effects model)
Study Regional General Relative Risk (Random) Weight Relative Risk (Random)
n/N n/N 95% CI (%) 95% CI
Berggren 1987 1/28 0/29 1.8 3.10 [ 0.13, 73.12 ]
Davis 1981 3/64 9/68 9.7 0.35 [ 0.10, 1.25 ]
Davis 1987 17/259 16/279 24.5 1.14 [ 0.59, 2.22 ]
Juelsgaard 1998 4/15 2/14 6.9 1.87 [ 0.40, 8.65 ]
McKenzie 1984 8/73 13/75 18.7 0.63 [ 0.28, 1.44 ]
McLaren 1978 1/26 9/29 4.3 0.12 [ 0.02, 0.91 ]
Racle 1986 2/35 5/35 6.6 0.40 [ 0.08, 1.93 ]
Valentin 1986 17/281 24/297 27.4 0.75 [ 0.41, 1.36 ]
Total (95% CI) 781 826 100.0 0.73 [ 0.47, 1.12 ]

0.01 0.1 1 10 100

Comparison 04. 03 Mortality - 3 months

Outcome: 03 Mortality - 3 months
n/N n/N 95% CI (%) 95% CI
Berggren 1987 1/28 0/29 0.5 3.10 [ 0.13, 73.12 ]
Couderc 1977 7/50 12/50 12.5 0.58 [ 0.25, 1.36 ]
Davis 1987 36/259 31/279 31.0 1.25 [ 0.80, 1.96 ]
McKenzie 1984 16/73 17/75 17.4 0.97 [ 0.53, 1.77 ]
Racle 1986 4/35 5/35 5.2 0.80 [ 0.23, 2.73 ]
Valentin 1986 22/281 33/297 33.4 0.70 [ 0.42, 1.18 ]
Total (95% CI) 726 765 100.0 0.92 [ 0.71, 1.21 ]

0.01 0.1 1 10 100


n/N n/N 95% CI (%) 95% CI
Davis 1987 44/259 42/279 39.6 1.13 [ 0.77, 1.66 ]
McKenzie 1984 20/73 21/75 20.3 0.98 [ 0.58, 1.65 ]
Valentin 1986 39/281 42/297 40.0 0.98 [ 0.66, 1.47 ]
Total (95% CI) 613 651 100.0 1.04 [ 0.81, 1.33 ]

0.01 0.1 1 10 100


n/N n/N 95% CI (%) 95% CI
McKenzie 1984 26/73 25/75 32.4 1.07 [ 0.69, 1.67 ]
Valentin 1986 54/281 53/297 67.6 1.08 [ 0.76, 1.52 ]
Total (95% CI) 354 372 100.0 1.07 [ 0.82, 1.41 ]

0.01 0.1 1 10 100


Comparison 04. 06 Mortality - early and up to 1 month
Outcome: 06 Mortality - early and up to 1 month
n/N n/N 95% CI (%) 95% CI
Adams 1990 4/24 3/32 3.1 1.78 [ 0.44, 7.21 ]
Berggren 1987 1/28 0/29 0.6 3.10 [ 0.13, 73.12 ]
Bigler 1985 1/20 1/20 1.2 1.00 [ 0.07, 14.90 ]
Davis 1981 3/64 9/68 10.5 0.35 [ 0.10, 1.25 ]
Davis 1987 17/259 16/279 18.6 1.14 [ 0.59, 2.22 ]
Juelsgaard 1998 4/15 2/14 2.5 1.87 [ 0.40, 8.65 ]
McKenzie 1984 8/73 13/75 15.5 0.63 [ 0.28, 1.44 ]
McLaren 1978 1/26 9/29 10.3 0.12 [ 0.02, 0.91 ]
Racle 1986 2/35 5/35 6.0 0.40 [ 0.08, 1.93 ]
Ungemach 1993 3/57 3/57 3.6 1.00 [ 0.21, 4.75 ]
Valentin 1986 17/281 24/297 28.1 0.75 [ 0.41, 1.36 ]
Total (95% CI) 882 935 100.0 0.76 [ 0.56, 1.04 ]

0.01 0.1 1 10 100


Comparison 04. 07 Length of operation (mins)
Outcome: 07 Length of operation (mins)
Study Regional General Weighted Mean Difference (Fixed) Weight Weighted Mean Difference (Fixed)
N Mean(SD) N Mean(SD) 95% CI (%) 95% CI
Berggren 1987 28 35.00 (10.00) 29 31.00 (10.00) 51.9 4.00 [ -1.19, 9.19 ]
Bigler 1985 20 67.00 (35.80) 20 59.00 (44.70) 2.2 8.00 [ -17.10, 33.10 ]
Bredahl 1991 15 60.00 (22.90) 13 65.00 (22.00) 5.0 -5.00 [ -21.65, 11.65 ]
Maurette 1988 18 80.50 (12.80) 15 71.50 (20.90) 9.5 9.00 [ -3.12, 21.12 ]
McKenzie 1984 73 82.20 (22.20) 75 77.20 (27.70) 21.5 5.00 [ -3.08, 13.08 ]
Racle 1986 35 125.00 (35.50) 35 116.00 (5.90) 9.8 9.00 [ -2.92, 20.92 ]
Total (95% CI) 189 187 100.0 4.82 [ 1.08, 8.56 ]

-10.0 -5.0 0 5.0 10.0

Comparison 04. 08 Operative hypotension

Outcome: 08 Operative hypotension
n/N n/N 95% CI (%) 95% CI
Berggren 1987 18/28 13/29 10.4 1.43 [ 0.88, 2.34 ]
x Brown 1994 0/10 0/10 0.0 Not estimable
Couderc 1977 14/50 20/50 16.2 0.70 [ 0.40, 1.22 ]
Davis 1987 98/259 67/279 52.4 1.58 [ 1.21, 2.04 ]
Juelsgaard 1998 12/15 9/14 7.6 1.24 [ 0.78, 1.98 ]
Maurette 1988 3/18 6/15 5.3 0.42 [ 0.12, 1.39 ]
McLaren 1978 3/26 1/29 0.8 3.35 [ 0.37, 30.21 ]
Racle 1986 10/35 9/35 7.3 1.11 [ 0.51, 2.40 ]
Total (95% CI) 441 461 100.0 1.31 [ 1.09, 1.58 ]

0.01 0.1 1 10 100


Comparison 04. 09 Operative hypotension (random effects model)
Outcome: 09 Operative hypotension (random effects model)
n/N n/N 95% CI (%) 95% CI
Berggren 1987 18/28 13/29 18.5 1.43 [ 0.88, 2.34 ]
x Brown 1994 0/10 0/10 0.0 Not estimable
Couderc 1977 14/50 20/50 16.1 0.70 [ 0.40, 1.22 ]
Davis 1987 98/259 67/279 28.0 1.58 [ 1.21, 2.04 ]
Juelsgaard 1998 12/15 9/14 19.3 1.24 [ 0.78, 1.98 ]
Maurette 1988 3/18 6/15 5.4 0.42 [ 0.12, 1.39 ]
McLaren 1978 3/26 1/29 1.8 3.35 [ 0.37, 30.21 ]
Racle 1986 10/35 9/35 10.8 1.11 [ 0.51, 2.40 ]
Total (95% CI) 441 461 100.0 1.18 [ 0.87, 1.60 ]

0.01 0.1 1 10 100

Comparison 04. 10 Operative blood loss (mls)

Outcome: 10 Operative blood loss (mls)
Study Regional General Weighted Mean Difference (Random) Weight Weighted Mean Difference (Random)
Bredahl 1991 15 190.00 (186.00) 13 321.00 (432.70) 18.8 -131.00 [ -384.35, 122.35 ]
Davis 1981 64 304.00 (232.00) 68 468.00 (445.30) 38.7 -164.00 [ -284.14, -43.86 ]
McKenzie 1984 73 277.70 (308.40) 75 261.70 (317.80) 42.5 16.00 [ -84.89, 116.89 ]
Total (95% CI) 152 156 100.0 -81.24 [ -216.01, 53.54 ]

-1000.0 -500.0 0 500.0 1000.0


Comparison 04. 11 Patients receiving blood transfusion
Outcome: 11 Patients receiving blood transfusion
n/N n/N 95% CI (%) 95% CI
Adams 1990 9/24 9/32 11.8 1.33 [ 0.63, 2.84 ]
Bigler 1985 9/20 7/20 10.7 1.29 [ 0.60, 2.77 ]
Davis 1981 45/64 52/68 77.4 0.92 [ 0.75, 1.13 ]
Total (95% CI) 108 120 100.0 1.01 [ 0.82, 1.24 ]

0.01 0.1 1 10 100

Comparison 04. 12 Transfusion requirements (mls)

Outcome: 12 Transfusion requirements (mls)
Study Regional General Weighted Mean Difference (Random) Weight Weighted Mean Difference (Random)
Couderc 1977 50 1100.00 (400.00) 50 1000.00 (300.00) 32.2 100.00 [ -38.59, 238.59 ]
Maurette 1988 18 600.00 (150.00) 15 300.00 (150.00) 35.2 300.00 [ 197.22, 402.78 ]
Racle 1986 35 488.60 (282.80) 35 480.00 (292.80) 32.6 8.60 [ -126.26, 143.46 ]
Total (95% CI) 103 100 100.0 140.69 [ -40.33, 321.70 ]

-100.0 -50.0 0 50.0 100.0


Comparison 04. 13 Post-operative hypoxia
Outcome: 13 Post-operative hypoxia
n/N n/N 95% CI (%) 95% CI
Berggren 1987 10/28 14/29 100.0 0.74 [ 0.40, 1.38 ]
Total (95% CI) 28 29 100.0 0.74 [ 0.40, 1.38 ]

Test for heterogeneity: not applicable
0.01 0.1 1 10 100

Comparison 04. 14 Length of hospital stay

Outcome: 14 Length of hospital stay
Study Regional General Weighted Mean Difference (Fixed) Weight Weighted Mean Difference (Fixed)
McKenzie 1984 73 38.80 (55.50) 75 42.90 (69.30) 6.1 -4.10 [ -24.30, 16.10 ]
Racle 1986 35 20.09 (10.60) 35 20.05 (11.40) 93.9 0.04 [ -5.12, 5.20 ]
Total (95% CI) 108 110 100.0 -0.21 [ -5.21, 4.78 ]

-100.0 -50.0 0 50.0 100.0


Comparison 04. 15 Pneumonia
Outcome: 15 Pneumonia
n/N n/N 95% CI (%) 95% CI
01 Fatal (reason for death only)

Adams 1990 1/24 1/32 2.8 1.33 [ 0.09, 20.26 ]
Davis 1981 2/64 4/68 12.7 0.53 [ 0.10, 2.80 ]
Davis 1987 5/259 4/279 12.6 1.35 [ 0.37, 4.96 ]
Juelsgaard 1998 2/15 0/14 1.7 4.69 [ 0.24, 89.88 ]
McKenzie 1984 5/73 3/75 9.7 1.71 [ 0.42, 6.91 ]
Subtotal (95% CI) 435 468 39.4 1.32 [ 0.63, 2.74 ]

02 Other (non fatal or fatal)

Berggren 1987 1/28 2/29 6.4 0.52 [ 0.05, 5.40 ]
Bigler 1985 1/20 2/20 6.5 0.50 [ 0.05, 5.08 ]
McLaren 1978 9/26 7/29 21.6 1.43 [ 0.62, 3.30 ]
Racle 1986 3/35 8/35 26.1 0.38 [ 0.11, 1.30 ]
Subtotal (95% CI) 109 113 60.6 0.78 [ 0.42, 1.45 ]

Total (95% CI) 544 581 100.0 0.99 [ 0.62, 1.59 ]
Test for overall effect z=0.04 p=1
0.01 0.1 1 10 100


Comparison 04. 16 Myocardial infarction
Outcome: 16 Myocardial infarction
n/N n/N 95% CI (%) 95% CI

Davis 1981 0/64 1/68 15.5 0.35 [ 0.01, 8.53 ]
Davis 1987 2/259 1/279 10.2 2.15 [ 0.20, 23.62 ]
McKenzie 1984 0/73 2/75 26.2 0.21 [ 0.01, 4.21 ]
Subtotal (95% CI) 396 422 52.0 0.63 [ 0.15, 2.62 ]


Juelsgaard 1998 1/15 0/14 5.5 2.81 [ 0.12, 63.83 ]
Racle 1986 2/35 4/35 42.5 0.50 [ 0.10, 2.56 ]
Subtotal (95% CI) 50 49 48.0 0.76 [ 0.20, 2.96 ]

Total (95% CI) 446 471 100.0 0.70 [ 0.26, 1.85 ]
0.01 0.1 1 10 100


Comparison 04. 17 Cerebrovascular accident
Outcome: 17 Cerebrovascular accident
n/N n/N 95% CI (%) 95% CI

Couderc 1977 2/50 2/50 23.7 1.00 [ 0.15, 6.82 ]
Davis 1987 3/259 0/279 5.7 7.54 [ 0.39, 145.24 ]
McKenzie 1984 0/73 1/75 17.6 0.34 [ 0.01, 8.27 ]
Racle 1986 0/35 1/35 17.8 0.33 [ 0.01, 7.91 ]
Subtotal (95% CI) 417 439 64.8 1.22 [ 0.40, 3.71 ]


Berggren 1987 3/28 0/29 5.8 7.24 [ 0.39, 134.12 ]
Bigler 1985 0/20 1/20 17.8 0.33 [ 0.01, 7.72 ]
Davis 1981 2/64 1/68 11.5 2.13 [ 0.20, 22.87 ]
Subtotal (95% CI) 112 117 35.2 2.07 [ 0.53, 8.06 ]

Total (95% CI) 529 556 100.0 1.51 [ 0.64, 3.57 ]
0.01 0.1 1 10 100


Comparison 04. 18 Congestive cardiac failure
Outcome: 18 Congestive cardiac failure
n/N n/N 95% CI (%) 95% CI

Adams 1990 2/24 2/32 13.6 1.33 [ 0.20, 8.80 ]
Davis 1987 3/259 3/279 22.8 1.08 [ 0.22, 5.29 ]
Juelsgaard 1998 1/15 0/14 4.1 2.81 [ 0.12, 63.83 ]
Subtotal (95% CI) 298 325 40.5 1.34 [ 0.44, 4.10 ]


Berggren 1987 2/28 0/29 3.9 5.17 [ 0.26, 103.18 ]
Bigler 1985 1/29 1/20 9.4 0.69 [ 0.05, 10.39 ]
Davis 1981 2/64 5/68 38.4 0.43 [ 0.09, 2.11 ]
Racle 1986 1/35 1/35 7.9 1.00 [ 0.07, 15.36 ]
Subtotal (95% CI) 156 152 59.5 0.85 [ 0.30, 2.40 ]

Total (95% CI) 454 477 100.0 1.05 [ 0.49, 2.23 ]
0.01 0.1 1 10 100


Comparison 04. 19 Renal failure
Outcome: 19 Renal failure
n/N n/N 95% CI (%) 95% CI

Davis 1987 1/259 0/279 11.2 3.23 [ 0.13, 78.95 ]
Racle 1986 0/35 1/35 34.9 0.33 [ 0.01, 7.91 ]
Subtotal (95% CI) 294 314 46.1 1.04 [ 0.15, 7.15 ]


Adams 1990 1/24 1/32 20.0 1.33 [ 0.09, 20.26 ]
Davis 1981 0/64 1/68 33.9 0.35 [ 0.01, 8.53 ]
Subtotal (95% CI) 88 100 53.9 0.72 [ 0.10, 5.13 ]

Total (95% CI) 382 414 100.0 0.86 [ 0.22, 3.41 ]
0.01 0.1 1 10 100

Comparison 04. 20 Acute confusional state

Outcome: 20 Acute confusional state
Study Regional Control Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI
Berggren 1987 4/28 7/29 36.4 0.59 [ 0.19, 1.80 ]
Bigler 1985 1/20 1/20 5.3 1.00 [ 0.07, 14.90 ]
Racle 1986 5/35 11/35 58.3 0.45 [ 0.18, 1.17 ]
Total (95% CI) 83 84 100.0 0.53 [ 0.27, 1.07 ]

Total events: 10 (Regional), 19 (Control)
0.01 0.1 1 10 100


Comparison 04. 21 Urine retention
Outcome: 21 Urine retention
Study Regional Control Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI
Berggren 1987 5/28 6/29 59.6 0.86 [ 0.30, 2.51 ]
Bigler 1985 5/20 4/20 40.4 1.25 [ 0.39, 3.99 ]
Total (95% CI) 48 49 100.0 1.02 [ 0.47, 2.23 ]

Total events: 10 (Regional), 10 (Control)
0.01 0.1 1 10 100

Comparison 04. 22 Vomiting

Outcome: 22 Vomiting
n/N n/N 95% CI (%) 95% CI
Bigler 1985 1/20 2/20 67.9 0.50 [ 0.05, 5.08 ]
McLaren 1978 1/26 1/29 32.1 1.12 [ 0.07, 16.95 ]
Total (95% CI) 46 49 100.0 0.70 [ 0.12, 3.94 ]

0.01 0.1 1 10 100


Comparison 04. 23 Deep vein thrombosis
Outcome: 23 Deep vein thrombosis
n/N n/N 95% CI (%) 95% CI
01 Fatal (underlying reason for death only)

McLaren 1978 0/26 2/29 3.9 0.22 [ 0.01, 4.43 ]
Subtotal (95% CI) 26 29 3.9 0.22 [ 0.01, 4.43 ]

02 Other: venography diagnosis

Brichant 1995 14/46 13/42 22.2 0.98 [ 0.52, 1.84 ]
McKenzie 1984 8/20 16/20 26.2 0.50 [ 0.28, 0.89 ]
Subtotal (95% CI) 66 62 48.4 0.72 [ 0.47, 1.11 ]

03 Other: fibrinogen scan diagnosis

Davis 1981 17/37 30/39 47.8 0.60 [ 0.40, 0.88 ]
Subtotal (95% CI) 37 39 47.8 0.60 [ 0.40, 0.88 ]

Total (95% CI) 129 130 100.0 0.64 [ 0.48, 0.86 ]
0.01 0.1 1 10 100


Comparison 04. 24 Pulmonary embolism
Outcome: 24 Pulmonary embolism
n/N n/N 95% CI (%) 95% CI
Adams 1990 1/24 0/32 3.2 3.96 [ 0.17, 93.17 ]
Berggren 1987 2/28 0/29 3.6 5.17 [ 0.26, 103.18 ]
Bigler 1985 2/20 0/20 3.7 5.00 [ 0.26, 98.00 ]
Brichant 1995 1/46 0/42 3.8 2.74 [ 0.11, 65.59 ]
Davis 1981 0/64 4/68 32.1 0.12 [ 0.01, 2.15 ]
Davis 1987 0/259 1/279 10.6 0.36 [ 0.01, 8.77 ]
McKenzie 1984 1/73 3/75 21.8 0.34 [ 0.04, 3.22 ]
McLaren 1978 0/26 2/29 17.4 0.22 [ 0.01, 4.43 ]
Racle 1986 1/35 0/35 3.7 3.00 [ 0.13, 71.22 ]
Total (95% CI) 575 609 100.0 0.90 [ 0.42, 1.94 ]

0.01 0.1 1 10 100


Comparison 04. 25 Pulmonary embolism (random effects model)
Outcome: 25 Pulmonary embolism (random effects model)
n/N n/N 95% CI (%) 95% CI
Adams 1990 1/24 0/32 9.6 3.96 [ 0.17, 93.17 ]
Berggren 1987 2/28 0/29 10.6 5.17 [ 0.26, 103.18 ]
Bigler 1985 2/20 0/20 10.8 5.00 [ 0.26, 98.00 ]
Brichant 1995 1/46 0/42 9.5 2.74 [ 0.11, 65.59 ]
Davis 1981 0/64 4/68 11.3 0.12 [ 0.01, 2.15 ]
Davis 1987 0/259 1/279 9.4 0.36 [ 0.01, 8.77 ]
McKenzie 1984 1/73 3/75 18.6 0.34 [ 0.04, 3.22 ]
McLaren 1978 0/26 2/29 10.7 0.22 [ 0.01, 4.43 ]
Racle 1986 1/35 0/35 9.5 3.00 [ 0.13, 71.22 ]
Total (95% CI) 575 609 100.0 0.98 [ 0.37, 2.64 ]

0.01 0.1 1 10 100


Comparison 04. 26 Pulmonary embolism (fatal and non fatal)
Outcome: 26 Pulmonary embolism (fatal and non fatal)
n/N n/N 95% CI (%) 95% CI

Adams 1990 1/24 0/32 3.6 3.96 [ 0.17, 93.17 ]
Bigler 1985 1/20 0/20 4.1 3.00 [ 0.13, 69.52 ]
Davis 1981 0/64 4/68 36.2 0.12 [ 0.01, 2.15 ]
Davis 1987 0/259 1/279 12.0 0.36 [ 0.01, 8.77 ]
McKenzie 1984 1/73 3/75 24.5 0.34 [ 0.04, 3.22 ]
McLaren 1978 0/26 2/29 19.6 0.22 [ 0.01, 4.43 ]
Subtotal (95% CI) 466 503 100.0 0.48 [ 0.18, 1.28 ]

02 Non fatal
Berggren 1987 2/28 0/29 24.4 5.17 [ 0.26, 103.18 ]
Bigler 1985 1/20 0/20 24.8 3.00 [ 0.13, 69.52 ]
Brichant 1995 1/46 0/42 25.9 2.74 [ 0.11, 65.59 ]
Racle 1986 1/35 0/35 24.8 3.00 [ 0.13, 71.22 ]
Subtotal (95% CI) 129 126 100.0 3.46 [ 0.74, 16.29 ]

0.01 0.1 1 10 100


Comparison 04. 01 Mortality - 1 month
Comparison: 02 Spinal and ’light’ general anaesthetic versus general anaesthetic
Outcome: 01 Mortality - 1 month
Study Spinal (+) General Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI
x White 1980 0/20 0/20 0.0 Not estimable
Total (95% CI) 20 20 0.0 Not estimable

Total events: 0 (Spinal (+)), 0 (General)
Test for overall effect: not applicable
0.01 0.1 1 10 100

Favours spinal (+) Favours general
Comparison 04. 02 Length of operation

Outcome: 02 Length of operation
Study Spinal (+) General Weighted Mean Difference (Fixed) Weight Weighted Mean Difference (Fixed)
White 1980 20 58.00 (23.00) 20 58.00 (25.00) 100.0 0.00 [ -14.89, 14.89 ]
Total (95% CI) 20 20 100.0 0.00 [ -14.89, 14.89 ]

-100.0 -50.0 0 50.0 100.0

Comparison 04. 03 Pneumonia

Outcome: 03 Pneumonia
n/N n/N 95% CI (%) 95% CI
White 1980 4/20 5/20 100.0 0.80 [ 0.25, 2.55 ]
Total (95% CI) 20 20 100.0 0.80 [ 0.25, 2.55 ]

0.01 0.1 1 10 100


Comparison 04. 04 Confusional state
Outcome: 04 Confusional state
n/N n/N 95% CI (%) 95% CI
White 1980 3/20 3/20 100.0 1.00 [ 0.23, 4.37 ]
Total (95% CI) 20 20 100.0 1.00 [ 0.23, 4.37 ]

0.01 0.1 1 10 100

Comparison 04. 05 Deep vein thrombosis

Outcome: 05 Deep vein thrombosis
n/N n/N 95% CI (%) 95% CI
White 1980 0/20 1/20 100.0 0.33 [ 0.01, 7.72 ]
Total (95% CI) 20 20 100.0 0.33 [ 0.01, 7.72 ]

0.01 0.1 1 10 100


Comparison 04. 01 Incomplete or unsatisfactory analgesia
Comparison: 03 Regional (spinal or epidural) versus lumbar plexus nerve blocks
Outcome: 01 Incomplete or unsatisfactory analgesia
Study Regional Nerve blocks Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI
01 Regional (spinal) block versus lumbar plexus block

Eyrolle 1998 5/25 19/25 78.1 0.26 [ 0.12, 0.59 ]
Subtotal (95% CI) 25 25 78.1 0.26 [ 0.12, 0.59 ]

Total events: 5 (Regional), 19 (Nerve blocks)
02 Regional (spinal) block versus lumbar plexus, sacral and iliac crest block
de Visme 2000 0/14 5/15 21.9 0.10 [ 0.01, 1.61 ]
Subtotal (95% CI) 14 15 21.9 0.10 [ 0.01, 1.61 ]

Total (95% CI) 39 40 100.0 0.23 [ 0.10, 0.50 ]
0.001 0.01 0.1 1 10 100 1000

Favours regional Favours nerve blocks
Comparison 04. 02 Operative hypotension

Outcome: 02 Operative hypotension
Study Regional (spinal) Nerve block Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI

Eyrolle 1998 18/25 3/25 100.0 6.00 [ 2.02, 17.83 ]
Total (95% CI) 25 25 100.0 6.00 [ 2.02, 17.83 ]

Total events: 18 (Regional (spinal)), 3 (Nerve block)
0.01 0.1 1 10 100

Favours regional Favours nerve block

Comparison 04. 03 Mean fall in arterial blood pressure (mmHg)
Outcome: 03 Mean fall in arterial blood pressure (mmHg)
Study Regional Nerve block Weighted Mean Difference (Fixed) Weight Weighted Mean Difference (Fixed)
de Visme 2000 14 46.00 (22.00) 15 30.00 (18.00) 100.0 16.00 [ 1.31, 30.69 ]
Total (95% CI) 14 15 100.0 16.00 [ 1.31, 30.69 ]

-100.0 -50.0 0 50.0 100.0

Comparison 04. 04 Mean dose of ephedrine used (mg)

Outcome: 04 Mean dose of ephedrine used (mg)
Study Regional Nerve block Weighted Mean Difference (Fixed) Weight Weighted Mean Difference (Fixed)

Eyrolle 1998 25 7.00 (2.80) 25 1.20 (2.70) 96.3 5.80 [ 4.28, 7.32 ]
Subtotal (95% CI) 25 25 96.3 5.80 [ 4.28, 7.32 ]

Test for overall effect z=7.46 p<0.00001
de Visme 2000 14 13.00 (14.00) 15 3.00 (5.00) 3.7 10.00 [ 2.24, 17.76 ]
Subtotal (95% CI) 14 15 3.7 10.00 [ 2.24, 17.76 ]

Total (95% CI) 39 40 100.0 5.96 [ 4.46, 7.45 ]
Test for overall effect z=7.80 p<0.00001
-100.0 -50.0 0 50.0 100.0


Comparison 04. 05 Adverse effects
Outcome: 05 Adverse effects
Study Regional (spinal) Nerve block Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI

Eyrolle 1998 6/25 1/25 100.0 6.00 [ 0.78, 46.29 ]
Subtotal (95% CI) 25 25 100.0 6.00 [ 0.78, 46.29 ]

x de Visme 2000 0/14 0/15 0.0 Not estimable
Subtotal (95% CI) 14 15 0.0 Not estimable

Test for overall effect: not applicable
Total (95% CI) 39 40 100.0 6.00 [ 0.78, 46.29 ]
0.01 0.1 1 10 100

Comparison 04. 06 Post-operative confusion

Outcome: 06 Post-operative confusion
Study Regional Nerve block Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI
de Visme 2000 5/14 6/15 100.0 0.89 [ 0.35, 2.28 ]
Total (95% CI) 14 15 100.0 0.89 [ 0.35, 2.28 ]

Total events: 5 (Regional), 6 (Nerve block)
0.01 0.1 1 10 100

Favours regional Favours nerve blocks

Comparison 04. 01 Mortality - during hospital stay
Comparison: 04 Intravenous ketamine versus general anaesthesia
Outcome: 01 Mortality - during hospital stay
Study Ketamine General Relative Risk (Fixed) Weight Relative Risk (Fixed)
n/N n/N 95% CI (%) 95% CI
Spreadbury 1980 9/30 9/30 100.0 1.00 [ 0.46, 2.17 ]
Total (95% CI) 30 30 100.0 1.00 [ 0.46, 2.17 ]

Total events: 9 (Ketamine), 9 (General)
0.01 0.1 1 10 100

Favours ketamine Favours general
Comparison 04. 02 Myocardial infarction

Outcome: 02 Myocardial infarction
n/N n/N 95% CI (%) 95% CI
Spreadbury 1980 0/30 1/30 100.0 0.33 [ 0.01, 7.87 ]
Total (95% CI) 30 30 100.0 0.33 [ 0.01, 7.87 ]

0.01 0.1 1 10 100


Comparison 04. 03 Congestive cardiac failure
Outcome: 03 Congestive cardiac failure
n/N n/N 95% CI (%) 95% CI
Spreadbury 1980 0/30 2/30 100.0 0.20 [ 0.01, 4.00 ]
Total (95% CI) 30 30 100.0 0.20 [ 0.01, 4.00 ]

0.01 0.1 1 10 100

Comparison 04. 04 Pulmonary embolism

Outcome: 04 Pulmonary embolism
n/N n/N 95% CI (%) 95% CI
Spreadbury 1980 0/30 3/30 100.0 0.14 [ 0.01, 2.65 ]
Total (95% CI) 30 30 100.0 0.14 [ 0.01, 2.65 ]

0.01 0.1 1 10 100

Comparison 04. 05 Length of hospital stay (discharge home)

Outcome: 05 Length of hospital stay (discharge home)
Study Ketamine General Weighted Mean Difference (Fixed) Weight Weighted Mean Difference (Fixed)
Spreadbury 1980 19 36.00 (12.00) 20 24.00 (8.00) 100.0 12.00 [ 5.57, 18.43 ]
Total (95% CI) 19 20 100.0 12.00 [ 5.57, 18.43 ]

-100.0 -50.0 0 50.0 100.0



Anaesthesia For Hip Fracture Surgery in Adults Cochrane 2004

Enviado por

Dados do documento

Descrição original:

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Anaesthesia For Hip Fracture Surgery in Adults Cochrane 2004

Enviado por

Direitos autorais:

Formatos disponíveis

Anaesthesia for hip fracture surgery in adults (Review)

Parker MJ, Handoll HHG, Griffiths R, Urwin SC

Anaesthesia for hip fracture surgery in adults (Review) 1

Anaesthesia for hip fracture surgery in adults (Review) ii

Parker MJ, Handoll HHG, Griffiths R, Urwin SC

This record should be cited as:

This version first published online: 23 October 2001 in Issue , .

BACKGROUND an airway, intra-operative hypotension, aspiration of gastric con-

DESCRIPTION OF STUDIES Additional information on trial methodology and results would be

Post-operative vomiting Composite outcome

Deep vein thrombosis d) Final outcome measures

Anaesthesia for hip fracture surgery in adults (Review) 12

Melton 1993 WHO study group 1994

Parker 1993 References to other published versions of this review

Characteristics of included studies

Study Adams 1990

Study Berggren 1987

Study Bigler 1985

Anaesthesia for hip fracture surgery in adults (Review) 16

Study Bredahl 1991

Study Brichant 1995

Study Brown 1994

Study Couderc 1977

Study Davis 1981

Study Davis 1987

Study Eyrolle 1998

Study Juelsgaard 1998

Anaesthesia for hip fracture surgery in adults (Review) 20

Study Maurette 1988

Study McKenzie 1984

Study McLaren 1978

Study Racle 1986

Study Spreadbury 1980

Anaesthesia for hip fracture surgery in adults (Review) 23

Study Tasker 1983

Study Ungemach 1993

Study Valentin 1986

Study White 1980

Study de Visme 2000

Characteristics of excluded studies

Study Reason for exclusion

Anaesthesia for hip fracture surgery in adults (Review) 26

Comparison 01. Regional (spinal or epidural) versus general anaesthesia

Comparison 04. Intravenous ketamine versus general anaesthesia

Title Anaesthesia for hip fracture surgery in adults

Authors Parker MJ, Handoll HHG, Griffiths R, Urwin SC

Issue protocol first published 1997/4

Review first published 1999/4

Date of most recent amendment 06 August 2003

Date of most recent 04 July 2001

Date new studies sought but Information not supplied by author

Date new studies found and 01 March 2001

Date authors’ conclusions Information not supplied by author

Cochrane Library number CD000521

Editorial group Cochrane Musculoskeletal Injuries Group

Editorial group code HM-MUSKINJ

Berggren 1987 1/28 0/29 0.6 3.10 [ 0.13, 73.12 ]

Davis 1981 3/64 9/68 11.4 0.35 [ 0.10, 1.25 ]

Davis 1987 17/259 16/279 20.2 1.14 [ 0.59, 2.22 ]

Juelsgaard 1998 4/15 2/14 2.7 1.87 [ 0.40, 8.65 ]

McKenzie 1984 8/73 13/75 16.8 0.63 [ 0.28, 1.44 ]

McLaren 1978 1/26 9/29 11.1 0.12 [ 0.02, 0.91 ]