Preconditioning for Renal Damage Under Contrast Examination
(Preconditioning-ReDUCE)
A Multi-Centre Feasibility Trial of Preconditioning Against ContrastInduced Acute Kidney Injury
Protocol Date: 22nd October 2013
Protocol Version:
3
STUDY COORDINATION CENTRE:
For general queries, supply of study documentation, and collection of data, please contact the HRB
Clinical Research Facility Galway:
Name:
Address:
Contact Number:
Email:
Clinical queries should be directed to the Central Study Co-ordinator who will direct the query to the
appropriate person:
Name: Mary Clarke-Moloney
Address: Vascular Research Unit, University Hospital Limerick
Contact Number: 061 482736
Email: mary.clarkemoloney@hse.ie
This study is supported by NCSRN Ireland.
This protocol describes Preconditioning for Renal Damage Under Contrast Examination, a multicentre feasibility trial of preconditioning against contrast-induced acute kidney injury in surgical
patients undergoing contrast-enhanced CT scans and provides information about study processes
and procedures.
This study will adhere to the principles of the Declaration of Helsinki and Good Clinical Practice
Guidelines. It will be conducted in compliance with the protocol, the Data Protection Acts and other
regulatory requirements as appropriate.
CLINICAL STUDY PROTOCOL
Protocol Number:
4
Study Title:
Renal Damage Under Contrast Examination (ReDUCE): a multicentre
feasibility randomised, sham-controlled clinical trial
Investigational product: N/A
Chief Investigator:
Professor Stewart R Walsh
Chair of Vascular Surgery, NUI Galway
E-mail: stewart.walsh@ul.ie
Co-investigators:
Professor John Forbes, HRB Chair of Health Economics, University of
Limerick
Professor Austin Stack, University of Limerick
Dr John Newell, Health Research Board Clinical Research Facility, Galway
Dr Mary Clarke-Moloney, University Hospital Limerick
NCSRN Network Lead: Ms Siobhan O’Daly, Irish Heart Foundation
Local Investigators:
Professor J Calvin Coffey, University Hospital Limerick
Professor Simon Cross, Waterford Regional Hospital, Waterford
Professor Michael Kerin, University College Hospital Galway
Collaborators:
Professor Walter Cullen, University of Limerick
Dr Phil Hodnett, University Hospital Limerick
Dr Tim Scanlon, University Hospital Limerick
Study Location:
Multicentre
Medical contact:
Professor Stewart R Walsh
Contact on site:
Local investigators, as above
CONTENTS
1
INTRODUCTION
Page 6
2
RATIONALE
Page 9
3
TRIAL OBJECTIVES
Page 10
4
TRIAL DESIGN
Page 11
5
TRIAL OUTCOME MEASURES
Page 15
6
TRIAL OUTCOME ASSESSMENT
Page 20
7
TRIAL INTERVENTION
Page 21
8
RANDOMISATION
Page 22
9
SAMPLE SIZE
Page 23
10
TRIAL RECRUITMENT
Page 26
11
TRIAL DURATION AND TIMETABLE
Page 27
12
ADAPTIVE DESIGN
Page 30
13
STATISTICAL ANALYSIS PLAN
Page 32
14
HEALTH ECONOMIC ANALYSIS
Page 35
15
ASSESSMENT OF SAFETY
Page 36
16
DIRECT ACCESS TO SOURCE DATA / DOCUMENTS
Page 41
17
ETHICAL CONSIDERATIONS
Page 42
18
DATA HANDLING AND RECORD KEEPING
Page 43
19
INSURANCE
Page 45
20
DISSEMINATION PLAN
Page 46
21
TRIAL ORGANISATION, ROLES & RESPONSIBILITIES
Page 47
22
EXTERNAL TRIAL CONSULTANTS
Page 50
23
TRIAL STAFFING
Page 51
24
TRIAL GOVERNANCE
Page 53
25
USER & STAKEHOLDER INVOLVEMENT
Page 55
INTRODUCTION
1.1 Background
Intravenous contrast for diagnostic and therapeutic interventions is a leading cause of hospitalacquired acute renal failure (1). Contrast-induced acute kidney injury (CI-AKI) occurs through a
complex mechanism including toxic and hypoxic renal tubular injury, diminished renal parenchymal
circulation and renal endothelial dysfunction with the production of oxygen free radicals due to
post-ischaemia oxidative stress (2). With increasing numbers of diagnostic and therapeutic
interventions, the incidence of CI-AKI will rise over the next few decades. Already, 40% of general
surgical in-patients undergo cross-sectional imaging with computed tomography (CT), largely using
contrast (3). Current CI-AKI preventive measures comprise identification of at-risk patients,
minimisation of contrast dose and the use of intravenous volume expansion (4). Simple, costeffective methods to reduce CI-AKI are required.
Risk factors for CI-AKI include advanced age, diabetes mellitus, pre-existing renal insufficiency and
congestive cardiac failure. Dehydration also increases the risk (5). While patients over the age of 65
years currently account for 11% of the Irish population, they account for 30% of surgical admissions
(6). In a recent cohort study of 1800 surgical in-patients in a major Irish teaching hospital, acute
renal failure was a common major complication, occurring in 2.4% of patients (6). Multivariate
analysis of this cohort demonstrated that increasing age and emergency admission were both
independent risk factors for complications (6). Emergency surgical patients constitute a high-risk
group in whom simple methods to minimise complications are required.
Remote ischaemic preconditioning (RIPC) is a simple technique whereby brief periods of skeletal
muscle ischaemia and reperfusion triggers a period of resistance to ischaemia-reperfusion injury in
distant tissues e.g. heart or kidney (7). It reduces renal damage following endovascular aneurysm
repair, a procedure requiring considerable contrast-volumes (8). We hypothesise that RIPC induced
using brief periods of upper limb ischaemia-reperfusion will reduce AKI in patients with mild-tomoderate renal impairment undergoing contrast enhanced abdomino-pelvic CT scans.
1.2 Proof-of-concept trials
The hypothesis that RIPC will reduce CI-AKI was first explored by Whittaker and Pryzklenk (9). They
utilised available data from a patient cohort undergoing emergency coronary angioplasty. Patients
with 1 to 3 balloon inflations in the coronary artery served as the control group whilst those with >3
balloon inflations served as the RIPC group. In this ‘natural experiment’, brief cardiac ischaemia
induced by balloon inflation in the coronary arteries during stenting served as the RIPC stimulus.
Both groups displayed an immediate improvement in estimated glomerular filtration rate (eGFR).
However, the control group then displayed a statistically significant decrease in eGFR by day 3 postprocedure (77 +14 ml/min/1.73m2 versus 70 +12 ml/min/1.73m2). The RIPC group displayed no such
difference by day 3 (81 +21 ml/min/1.73m2 versus 80 +14 ml/min/1.73m2) despite having received a
larger contrast volume.
The hypothesis was further evaluated by means of a randomised controlled trial in patients at highrisk of CI-AKI undergoing elective coronary angiography (10). The trial randomised 100 patients in a
1:1 ratio to either intermittent arm ischaemia with 4 cycles of 5 minutes blood pressure cuff inflation
followed by 5 minutes deflation (RIPC group) or a sham control. The primary outcome was CI-AKI
defined as an increase in serum creatinine of > 25% or >0.5 mg/dL above baseline at 48 hours after
contrast exposure. The incidence of CI-AKI was significantly reduced by RIPC (12% versus 40%; p =
0.002).
Igarashi et al undertook a further randomised trial in patients undergoing coronary angiography who
were at low- to moderate-risk of CI-AKI (11). This group randomised 60 patients in a 1:1 ratio to
undergo RIPC using the same protocol as Er et al (10) but omitted the sham control. The trial was
primarily designed to evaluate the effect of RIPC on renal injury biomarkers. CI-AKI was reduced
from 27% of controls to 8% in the RIPC group. This difference was not statistically significant but this
probably represents a type 2 error as the effect size is similar to that by Er et al (10).
1.3
Pilot Trial in patients undergoing Contrast-Enhanced CT Scans
Our group has recently completed a pilot trial of RIPC in patients undergoing contrast-enhanced CT
of the abdomen and pelvis (NCT01741896; manuscript in preparation). Surgical in-patients
undergoing contrast-enhanced CT of the abdomen and pelvis (ce-CTAP) were randomly allocated to
undergo RIPC one hour pre-scan or usual care. The trial was stratified for diabetes and pre-existing
renal impairment. Patients with pre-existing renal impairment were pre-hydrated with isotonic
saline. Over a three month period, 100 patients were randomised. There were three dropouts. Of
the remaining 97 patients, 45 were allocated to RIPC and 52 to control. Both groups were wellmatched with respect to baseline renal function, medication use, diabetes and pre-hydration. The
rate of change of creatinine over the first 24 hours was significantly lower in the RIPC group (-0.08 vs
=0.125 micromols/ml/hr; p=0.02). Peak post-scan creatinine as a percentage of baseline was lower
in the RIPC group (99% versus 107%; p=0.01) while the trough eGFR as a percentage of baseline was
higher in the RIPC group (97% versus 90%); p=0.06).
2
RATIONALE
Greater availability of CT technology has led to a dramatic increase in the number of patients
undergoing ce-CTAP in recent years. Between 1996 and 2010, the number of patients undergoing CT
scans in the United States tripled, increasing by 8% per annum (12). Simultaneously, an increasingly
elderly and sick population means that many patients now have multiple co-morbidities, increasing
their risk of contrast-induced acute kidney injury. Approximately 6.5% of patients undergoing ce-CT
develop acute kidney injury defined as a >25% increase in serum creatinine from baseline (13).
Simple methods to reduce CI-AKI are required.
Ideally, a clinical trial of RIPC in patients undergoing ce-CT would be powered to detect a
significant reduction in CI-AKI. Weisbord et al reported a 6.5% incidence of CI-AKI in their
cohort of 421 patients undergoing CT (13). In the Limerick pilot trial described above, 10 of
97 patients (10.3%) developed CI-AKI. Assuming that 6% develop CI-AKI following ce-CT,
1962 patients would be required in each arm of a trial designed to detect a reduction in CIAKI from 6% with standard care to 4% with RIPC (alpha 0.05, beta 0.8).
The proposed work builds on the previous trials of RIPC conducted by the chief investigator in this
area, a programme consistent with the ‘MRC Framework for the Design and Evaluation of Complex
Interventions to Improve Health’, which suggests core phases to the development of health
interventions. This proposal describes the development and piloting phases of the MRC framework.
Conducting a full-scale RCT, and economic evaluation, of remote preconditioning versus ‘standard
care’ in this population is likely to require at least 4000 participants and to be costly. As there are
uncertainties regarding rates of eligibility, consent, participation in the intervention and retention
for follow-up - and regarding the feasibility / acceptability of the intervention - this feasibility study is
essential to inform the design and conduct of a larger scale study.
This feasibility trial aims to address a number of key points:
1. To date, the role of RIPC in preventing CI-AKI in patients receiving contrast has been evaluated in
three trials (two published and one unpublished pilot). Only the unpublished pilot evaluated the
potential effect of RIPC in patients undergoing ce-CT. A phase 2B trial to further evaluate potential
efficacy using a surrogate marker is required to justify phase 3 work. This will also provide an
accurate estimate of the likely effect size that can be expected in any phase 3 trial
2. As identified in recent HRB reports, clinical trial capability in Ireland is underdeveloped relative to
other Western countries. The chief investigator and most of the ReDUCE team are responsible for a
multicentre feasibility trial of RIPC in major vascular surgery (Preconditioning SAVES), funded under
the HRB Health Research Awards 2013. The ReDUCE trial will allow the group to build upon its
experience and further develop the multicentre network essential for clinical trials in Ireland. An
examination of the feasibility of a large multicentre surgical trial particularly with respect to
recruitment, drop-off rates, compliance, data collection and randomisation would be beneficial in
advance of proceeding with a phase 3 trial.
3. It is expected that any phase 3 trial will run through the National Cardiovascular and Stroke
Research Network based at the Irish Heart Foundation. The NCSRN is a partner organisation in the
Preconditioning SAVES trial. The ReDUCE feasibility trial will provide the NCSRN team with the
opportunity to further refine the processes and procedures required for the logistical organisation
required for a multicentre randomised trial of a complex intervention, in advance of any full-scale
trials.
4. CI-AKI is usually transient. However, data regarding the effect on medium-term renal function
are lacking. Professor Austin Stack leads the Mid-West Renal Data Registry, funded by the HRB
Health Research Awards 2013. The ReDUCE trial provides an opportunity to evaluate the effect of
ce-CT on medium term renal function utilising the Mid-West Renal Data Registry.
5. Professor John Forbes has been appointed to the University of Limerick as the HRB Chair of
Health Economics under the Research Leader Award Scheme. He has extensive trials experience.
The ReDUCE trial provides an opportunity to incorporate his expertise into the design and conduct of
clinical trials in Ireland, thereby meeting a fundamental objective of the Leader Award Scheme
(building trial capability and expertise).
TRIAL OBJECTIVES
The single main research question for the feasibility trial, in terms of PICO (Population; Intervention;
Comparator; Outcome) is as follows: ‘In adult surgical in-patients undergoing contrast-enhanced CT
scans of the abdomen and pelvis, does RIPC induced by brief arm ischaemia and reperfusion, when
compared to control, reduce the proportion of patients developing a significant reduction in
estimated glomerular filtration rate in the first 3 post-scan days? ‘. Whilst powered to address this
specific research question, this phase 2b trial will also provide insight into the feasibility of running a
larger, phase 3 study and will provide data regarding the proposed primary efficacy endpoint (CI-AKI)
3.1
Primary research objective
To determine whether RIPC induced using brief arm ischaemia and reperfusion reduces the
proportion of patients who develop a reduction in estimated glomerular filtration rate (eGFR) >20%
in the first 3 days following the CT scan.
3.2
Secondary research objectives
(i)
To evaluate the proportion of patients who develop CI-AKI defined as a rise in serum
creatinine > 25% of baseline in the first 3 days post-scan in order to inform the design of
a phase 3 trial
(ii)
To determine whether RIPC reduces length of hospital stay
(iii)
To determine whether RIPC reduces ITU stay
(iv)
To determine whether RIPC reduces unplanned critical care admission
(v)
To determine the proportion of patients who develop renal impairment within 1 year of
ce-CT
(vi)
To evaluate the likely recruitment rates for a phase 3 trial
(vii)
To evaluate compliance and logistic issues with respect to multicentre trials in the Irish
context
(viii)
To examine the acceptability of the intervention to clinicians and patients utilising
qualitative methods
4
TRIAL DESIGN
4.1
Statement of design
A randomised controlled pilot trial of the effect of RIPC induced by brief arm ischaemia and
reperfusion on eGFR levels in the first 3 days post-scan amongst surgical in-patients undergoing
contrast-enhanced CT scans of the abdomen and pelvis.
4.2
Trial treatment
The intervention being tested in non-pharmacological. Patients will be randomised 1:1 into one of
two groups: RIPC or control. Preconditioning will be performed in the same manner as several
previous trials. One hour before the CT scan, a standard, CE-approved tourniquet cuff will be placed
around one arm of the patient. It will then be inflated to a pressure of 200mmHg for 5 minutes. For
patients with a systolic blood pressure >185mmHg, the cuff will be inflated to at least 15mmHg
above the patient’s systolic blood pressure. The cuff will then be deflated and the arm allowed
reperfuse for 5 minutes. This will be repeated so that each patient receives a total of 4 ischaemiareperfusion cycles.
Patients randomised to the control arm will receive a sham intervention. A blood pressure cuff will
be applied and inflated to a level 10mmHg below the patients systolic blood pressure for 5 minutes,
then deflated for 5 minutes. This cycle will be repeated so that the patient receives a total of 4 sham
cycles.
Patients with pre-existing renal impairment (defined as a baseline eGFR of 30 to 60 mls/min/1.73m2)
will receive pre-scan volume loading with isotonic normal saline which will be continued for at least 24
hours post-scan. In all other respects, the procedure and peri-scan care will follow the routine
practices of the surgeons and radiologists involved.
4.3
Inclusion criteria
Surgical in-patients aged 40 years or over scheduled to undergo contrast-enhanced CT scans of the
abdomen and pelvis will be eligible for inclusion.
4.4
Exclusion criteria
Patients will be excluded if any of the following apply:
(i)
Pregnancy
(ii)
Significant upper limb peripheral arterial disease
(iii)
Previous history of upper limb deep vein thrombosis
(iv)
Patients on glibenclamide or nicorandil (these medications may interfere with RIPC)
(v)
Patients with an estimated pre-operative glomerular filtration rate <
30mls/min/1.73m2
(vi)
Patients with a known history of myocarditis, pericarditis or amyloidosis
(vii)
Patients who have received intravenous contrast in the previous year.
(viii)
Patients with severe hepatic disease defined as an international normalised ratio >2
in the absence of systemic anticoagulation
(ix)
4.5
Patients previously enrolled in the trial representing for a further scan
Selection of patients
In each centre, potentially eligible patients will be selected from in-patients under the care of
consultant surgeons who are scheduled to undergo contrast-enhanced CT of the abdomen and
pelvis.
4.6
Informed consent
The process of obtaining informed consent will be conducted in compliance with the principals of
good clinical practice and requirements of the approving research ethics committee and other
regulatory requirements as appropriate. Consent to enter the study must be sought from each
subject only after a full explanation has been given and time allowed for consideration. Written
information will be provided for each potential participant. Signed subject consent will be obtained
and a copy given to the subject. It is a right of the subject to refuse to participate or to withdraw
without at any time from the protocol without giving reasons and without prejudicing treatment.
4.7
Baseline data
Patients will all have a full medical history taken and clinical examination as part of their usual care.
The following will be recorded:
1. Weight
2. Height
3. Blood pressure
4. Heart rate
5. ECG
6. Gender
7. Ethnicity
8. Date of birth
9. Diabetes mellitus
10. Hypercholesterolaemia
11. Hypertension
12. Previous myocardial infarction
13. Previous coronary revascularisation
14. Previous stroke
15. Atrial fibrillation
16. Peripheral arterial disease
17. Smoking history
18. NYHA class
19. Medication at time of consent (aspirin, clopidogrel, beta-blocker, calcium-channel
antagonist, nitrates, cholesterol-lowering agent, ACE inhibitor / A2 receptor antagonist,
insulin, metformin, sulphonylurea, warfarin)
20. Serum creatinine and troponin.
4.8
Estimated Glomerular Filtration Rate
This will be assessed by measuring serum creatinine samples at 24, 48 and 72 hours post-scan. The
samples will be analysed in the local laboratory at each site, using local protocols. The creatinine
values will be used to calculate the eGFR at each time point. This calculation will be performed
centrally using the abbreviated MDRD (Modification of Diet in Renal Disease) equation (14) as
recommended by the UK National Institute of Health and Clinical Excellence.
4.9
Length of hospital / ITU stay
The duration of hospital stay and ITU stay have a major impact on health service resource utilisation,
and are factors which can be influenced by acquired kidney injury.
4.10
Acute Kidney Injury Score
The AKI score will be calculated over the first three post-scan days. Creatinine will be measured
daily as part of routine care. Urine volumes will be calculated from the fluid balance charts
maintained as part of usual care.
4.11
Post-scan data
On the day of their scan, the patient will be randomised to either the RIPC or control arm using a
web-based system. The following information regarding the scan will be recorded for all patients:
scanner model, contrast type, contrast dose.
4.13
Data at discharge
At discharge, duration of hospital stay and ITU stay will be recorded. The final diagnosis for each
patients’ admission, requirement for renal replacement therapy and any surgical procedures
performed will be recorded. Data regarding complications occurring during the patients’ hospital
stay will also be recorded.
4.14
Data at 1 year
For patient recruited in the Mid-West, one year renal function data will be collected via the MidWest Renal Data Registry and the proportion of patients developing chronic renal impairment within
1 year of ce-CT will be evaluated.
4.15
Data at 5 years
For patients recruited in the Mid-West, five-year renal function and survival data will be obtained
through the Mid-West Renal Data Registry.
TRIAL OUTCOME MEASURES
5.1
Primary efficacy endpoint
The trial is intended to pragmatically evaluate the potential of RIPC to improve clinical outcomes
among patients undergoing ce-CT in routine clinical practice. For the pilot trial, a surrogate marker
of efficacy will be used, namely the proportion of patients developing a > 20% reduction in eGFR in
the first three days post-scan. The primary efficacy outcome will be a comparison of the proportion
of patients in each arm of the trial who develop a > 20% reduction in eGFR in the first three days
post-scan.
5.2
Primary safety endpoint
The intervention under investigation carries a theoretical risk of causing acute upper limb ischaemia
or an upper limb deep vein thrombosis. Such an event has not been reported by any of the phase 1
and 2 trials of RIPC in cardiac patients. In addition, there is a theoretical risk of minor adverse events
such as upper limb bruising. The primary safety endpoint will again be a composite of major life- or
limb-threatening adverse events which could be attributed to the preconditioning stimulus. The
components are:

Acute upper limb ischaemia

Acute upper limb deep vein thrombosis
The components of the primary safety endpoint are defined as follows:
Acute upper limb ischaemia
This is defined as the development of ischaemia in the arm used for the preconditioning stimulus
requiring systemic anti-coagulation, radiological intervention or surgical intervention
Acute upper limb deep vein thrombosis
This is defined as the development of thrombus within the subclavian, axillary or brachial vein
confirmed in duplex ultrasound and in the same arm as used for the RIPC stimulus
Secondary endpoints
These will be:
(i)
The proportion of patients who develop CI-AKI defined as a rise in serum creatinine >
25% of baseline in the first 3 days post-scan in order to inform the design of a phase 3
trial
(ii)
Length of hospital stay
(iii)
Length of ITU stay
(iv)
Unplanned critical care admissions
(v)
The proportion of patients who develop renal impairment within 1 year of ce-CT
(vi)
Acute kidney injury score in first three peri-scan days (Table 1)
(vii)
In-hospital mortality
(viii)
Short-term cost-effectiveness and cost-effectiveness at 1 year
(ix)
Other complications during admission (myocardial infarction, cerebrovascular accident,
transient ischaemic attack, respiratory failure, cardiac arrest, wound infection, chest
infection, urinary tract infection, abdominal infection, other sepsis, pulmonary embolus,
deep vein thrombosis, prolonged ileus)
AKI
Creatinine criteria
Urine output criteria
1
Rise > 26.4micromols/L or 150 to 200% of baseline
<0.5ml/kg/hr for >6hrs
2
Rise of 200 to 300% of baseline
<0.5ml/kg/hr for >12hrs
3
Increase of. 300% or creatinine > 354 micromols/L with
<0.3ml/kg/hr for >24 hrs or
acute rise of at least 44 micromols/L
anuria for 12 hrs
Grade
Table 1: Acute kidney injury criteria
TRIAL OUTCOME ASSESSMENT
The predefined outcomes will be recorded in an electronic case record form developed by the HRB
Clinical Research Facility Galway.
6.1
Evaluation of primary outcome
Calculation of the eGFR values and the determination of a decrease >20% from baseline will be
undertaken by the trial statistical team who will be blinded to trial allocation. Thus, evaluation of
the primary outcome will be double-blinded.
6.2
Evaluation of other primary and secondary outcomes
The development of the remaining primary and secondary outcomes will be determined by a trial
research nurse at each of the three sites. The EQ-5D-5L self-complete quality-of-life evaluation will
be administered by the research nurse immediately prior to discharge from the index admission.
The one-year follow-up EQ-5D-%L instrument will be sent to surviving patients from the central trial
office.
TRIAL INTERVENTION
The intervention being tested is a simple intervention which can be undertaken using existing
equipment. About 1 hour before the scan, patients randomised to the RIPC arm will have a
standard, CE-approved tourniquet cuff will be placed around one arm of the patient. It will then be
inflated to a pressure of 200mmHg for 5 minutes. For patients with a systolic blood pressure
>185mmHg, the cuff will be inflated to at least 15mmHg above the patient’s systolic blood pressure.
The cuff will then be deflated and the arm allowed reperfuse for 5 minutes. This will be repeated so
that each patient receives a total of 4 ischaemia-reperfusion cycles.
Patients randomised to the control arm will receive a sham intervention. A blood pressure cuff will
be applied and inflated to a level 10mmHg below the patients systolic blood pressure for 5 minutes,
then deflated for 5 minutes. This cycle will be repeated so that the patient receives a total of 4 sham
cycles.
RANDOMISATION
Randomisation for the trial will be undertaken using a web-based randomisation system
administered by the HRBCRF Galway. The use of third-party randomisation will maintain allocation
concealment. A unique trial number will be assigned at the time of randomisation. Once recruited,
patients will be randomised to one of two groups: either to RIPC or to the sham control arm.
Randomisation will be stratified by centre using randomly selected blocks of 4, 6 and 8. In addition,
each centre will be stratified by diabetes and baseline renal impairment. The randomisation will be
accessed by a designated member of the research team at each site who will not be involved with
data collection.
SAMPLE SIZE
Currently, the projected sample size for a full phase 3 trial using CI-AKI as the primary endpoint is
4000 patients in total. This would provide sufficient power to detect a reduction in the proportion of
patients sustaining CI-AKI from 6% to 4% (alpha 0.05, beta 0.8). The assumed 6% figure in the
control arm is based upon data from a sample of only 100 patients randomised at University Hospital
Limerick. Before proceeding with a full-scale trial, and in accordance with the MRC framework for
the evaluation of complex interventions, it would be preferable to (i) explore the feasibility of
randomising a significant number of surgical patients in the Irish healthcare system context (ii)
undertake further work using a surrogate marker of efficacy to more robustly evaluate the apparent
protective effect of RIPC in patients undergoing ce-CT (iii) use such a surrogate marker to evaluate
the likely effect size of RIPC and (iv) evaluate the incidence of CI-AKI following ce-CT
The feasibility trial design utilises a > 20% decrease in eGFR from baseline as the primary outcome.
The sample size calculation for the pilot trial is based upon data from the Limerick pilot trial of RIPC
in surgical in-patients undergoing ce-CT. In this trial, 16 of 97 patients (16%) developed such an
eGFR decrease. In order to demonstrate a reduction from 16% to 10% (alpha 0.05, beta 0.8), 492
patients are required in each arm of the trial. The trial aims to recruit 1000 patients. At 25% of the
proposed full trial recruitment, the investigators feel it is sufficiently large to allow adequate testing
of trial procedures and compliance at multiple sites in line with objective (i) above.
AVAILABLE PILOT DATA
Some pilot work has been undertaken at University Hospital Limerick. Over a 3-month period in
2013, 97 surgical in-patients undergoing ce-CT were randomly allocated to either RIPC or standard
care. The randomisation was stratified for pre-existing renal impairment and also for diabetes.
Patients with baseline renal impairment received pre-scan volume expansion with isotonic saline
which was continued for 24 hours post-scan. The results are summarised in Table 2. The groups
were well-matched. Of note, the RIPC group displayed greater preservation of post-procedure eGFR.
Median age
Diabetes
Pre-existing renal
impairment
Hypertension
Statin use
ACE use
Diuretic use
ARB use
Prehydrated
Baseline creatinine
(median)
Baseline eGFR
(median)
Creatinine rise >20%
from baseline
eGFR fall > 20% from
baseline
Creatinine AUC
eGFR AUC
Rate of change
creatinine 1st 24 hours
Rate of change
creatinine 1st 48 hours
Peak creatinine as %
baseline
Trough eGFR as %
baseline
Control (n = 52)
62
8 (15%)
7 (13%)
RIPC (n = 45)
65
7 (15%)
8 (18%)
p
0.31
0.99
0.76
20 (38%)
15 (27%)
7 (13%)
8 (15%)
5 (9%)
18 (35%)
75 microlmols/dl
22 (48%)
14 (30%)
5 (11%)
13 (28%)
7 (15%)
22 (48%)
74.5 micromols/dl
0.41
0.98
0.97
0.17
0.56
0.22
0.79
99 mls/kg/min
83 mls/kg/min
0.15
5 (9%)
5 (11%)
0.99
9 (17%)
7 (15%)
0.99
5256
6208
+0.125
micromols/ml/hr
+0.02 micromols/ml/hr
5616
5888
-0.08 micromols/ml/hr
0.77
0.34
0.02
-0.04 micromols/ml/hr
0.48
107%
99%
0.01
90%
97%
0.06
Table 2: Data from Limerick pilot
While some preliminary data have been generated, a larger feasibility cohort is required. Robust
data regarding network level recruitment, compliance and dropouts are lacking. The early pilot was
run using sequentially numbered sealed opaque envelopes rather than web-based randomisation.
Logistic issues around allocation concealment and centralised randomisation have not been
evaluated. Consequently, a feasibility trial appears necessary before proceeding with a full phase 3
trial.
EVALUATION OF ACCEPTABILITY AND BARRIERS TO IMPLEMENTATION
Whilst RIPC is simple to induce and inexpensive, it does add to the burden of clinical staff in advance
of CT scans. Perceptions that the intervention is onerous will significantly impede adoption into
routine practice should a clinical benefit be demonstrated. It may also significantly hinder
recruitment in any phase 3 trial, as reluctance to complete the preconditioning protocol may result
in numerous protocol breaches in the intervention arm. For patients, the intervention may be
burdensome and uncomfortable, again negatively impacting upon likely adoption in to routine
practice.
In order to explore these potential issues in advance of any full-scale trial, the feasibility trial will
include a qualitative evaluation of acceptability to both patients and clinical staff together with a
qualitative evaluation of any perceived barriers to implementation. This insights may prove valuable
in the subsequent design of a full-scale trial and ensuring adoption in to routine practice should the
intervention prove clinically and economically-efficacious.
Healthcare professionals at participating practices will be asked to complete a self-administered
questionnaire at the end of the study period. The questionnaire will elicit data on profession and
practice details, their perceived experience of trial involvement, and open-ended questions to elicit
information regarding attitudes to trial involvement, willingness to recruit participants, difficulties
that arose during the trial and potential barriers to further research or routine clinical use of the trial
intervention.
A convenience sample of 40 patients (10 from each site) will be contacted following their admission
to conduct a brief interview. This will again involve the use of open-ended questions in order to
elicit information regarding patients’ attitudes towards the intervention and perceived barriers to
implementation. These interviews will be conducted by Dr Clarke-Moloney, under the supervision of
Professor Walter Cullen. Professor Cullen is currently chief investigator of the the recently
established HRB-funded (feasibility) trial of 'Psychosocial Interventions for Problem Alcohol Use
Among Problem Drug Users'. He has established expertise in qualitative research methods and will
lead the qualitative aspects of the SAVES feasibility trial.
STATISTICAL ANALYSIS PLAN
The statistical analysis plan has been developed in conjunction with the HRBCRF Galway. The
statisticians will be blinded to the trial allocation for all analyses, with the two trial arms being
labelled A and B.
Interim analysis
No interim analysis is planned for this one-year feasibility trial
Final analysis
Once the trial has closed, a final analysis of the data will be conducted by Dr Newell, the trial
biostatistician. The trial arms will be labelled A and B, to maintain blinding. Baseline characteristics
will be compared between the trial arms to evaluate the efficacy of randomisation. The proportions
of patients with primary and secondary end-points will be compared between the two groups.
These results will all be presented in the final trial report.
Subgroup analyses
Experimental data suggest that the protective effect of RIPC may be attenuated in diabetic patients.
This trial provides an opportunity to evaluate the relationship between RIPC and diabetes in clinical
practice. There will be two components to this analysis. In the diabetic subgroup, the proportion of
patients suffering a n eGFR decrease will be compared between those patients who receive RIPC and
patients who are controls. In addition, the proportion of patients suffering an eGFR decrease will be
compared between diabetic RIPC patients and non-diabetic RIPC patients, in order to determine
whether the effect size of RIPC differs significantly in diabetic patients. These data will inform
subsequent design of a phase 3 trial, should it appear justified.
There have also been suggestions that RIPC is attenuated in older patients. Again, this trial provides
an opportunity to investigate this potential relationship further. Proportions of the patients
sustaining an eGFR decrease will be compared between patients aged > 80 years at randomisation
and patients aged < 80 years at randomisation. The proportion with an eGFR decrease will also be
compared between RIPC and control patients in the > 80 subgroup. A between group analysis will
compare the proportions of eGFR decreases between RIPC patients in the > 80 and < 80 groups.
These data will again inform subsequent design of a phase 3 trial, should it appear justified.
Qualitative data analysis
Each interview will be transcribed verbatim, following which the transcript will be reviewed by the
researchers for accuracy. Each interview will be ‘openly coded’ (to allow development of categories
of concepts, without making any prior assumptions). The researcher will analyse half the interviews
first to identify common codes for each group of respondents. These coded transcripts will be
reviewed by a second researcher to ensure the interviews are coded correctly and consistently and
the codes identified by the researcher were the same as those identified by the expert. The second
half of the transcripts will be consequently added to the analysis one-by-one allowing for the
monitoring of data saturation.
For the purpose of this research, thematic analysis will be used to analyse qualitative data. This
approach has many benefits for studies such as this which are interpretive in nature, as it is a
‘method for identifying, analysing and reporting patterns (themes) within data’(15). The process of
thematic analysis is concerned with the basic to advanced encoding of data, which are subsequently
developed to themes. Themes are described as ‘patterns found in the information’ which can be
developed in an inductive manner from raw information but also deductively from prior / existing
research and knowledge. This flexible approach can also be seen in how themes identified at one
level can help the researcher describe their observations and at a more advanced level allow the
researcher to interpret aspects of the phenomenon under study(15).
Qualitative data analysis will be systematic and organised in order to easily locate information within
the data set when tracing results, providing examples in context(15). The qualitative research
software Nvivo v8 will be used to facilitate the coding of this data. Such research software allows the
researcher to manage, shape and make sense of unstructured information; it also facilitates the
complex organisation and retrieval of data (www.qsrinternational.com). The analysis will follow a “5Step Analysis” approach whereby data is reviewed, examined, coded, themes generated and
defined(15). To achieve validity in the coding / analysis of data, two reviewers will code data
independently and inter-rater reliability measures will be computed based on this coding. Coding
consistency will be maintained throughout the coding process and will be reviewed by regular
meetings between researchers and principal investigator. The findings will be compared with other
study findings (validity and credibility). The researchers will present the findings to participants to
determine if the study findings reflect their experience of the topic under study (member checking).
Illustrative quotes will be used to emphasise points made by the participants.
HEALTH ECONOMIC ANALYSIS
The short-term cost-effectiveness of RIPC will be investigated using a within-trial analysis of the
difference or increment in mean effectiveness and the increment in mean costs. We anticipate that
the direct costs of the intervention itself will be modest but that the net benefits (i.e., economic
value of health benefits less costs) could be substantial if the risk of renal injury is reduced,
subsequent health outcomes improve and less costly patterns of treatment and resource use
emerge. We will measure resource use from the perspective of the health service using individual
level data on cumulative hospital stay and ICU stay up to one year from randomisation. Health
related quality of life will be estimated using the self-complete version of the EQ-5D-5L instrument
collected at around the time of hospital discharge and the one year follow-up. Expecting correlation
between the distribution of costs and benefits we will use a multivariate modelling framework that
allows for estimation of a marginal density for costs and a conditional density for benefits, given the
costs. Probabilistic sensitivity analysis of parameter uncertainty will be conducted to assess the
robustness of the cost effectiveness model and the expected value of information.
ASSESSMENT OF SAFETY
Definitions
This is not a trial of an investigational medicinal product. By definition, therefore, all untoward
events will be adverse events rather than adverse reactions. Safety assessments will be from the
time of randomisation to completion of follow-up.
Adverse event
An adverse event is defined as any untoward medical occurrence affecting a patient which does not
necessarily have a causal relationship with the RIPC stimulus. The terms ‘mild, moderate or severe’
will be used to describe the intensity of a specific event. This is not the same as serious which is
based on criteria outlined below. An adverse event can therefore be any unfavourable and
unintended sign (including an abnormal laboratory finding), symptom or disease temporally
associated with the use of RIPC whether or not it is considered to be related to the technique.
Serious adverse event
A serious adverse event is defined as any untoward medical occurrence / effect that:
(i)
Results in death]
(ii)
Is life-threatening
(iii)
Requires hospitalisation or prolongs the patient’s hospital stay
(iv)
Results in persistent or significant disability or incapacity
‘Life-threatening’ in the definition of a serious adverse event refers to an event in which the subject
was at risk of death at the time of the event; it does not refer to an event which hypothetically might
have caused death were it more severe.
Unexpected adverse event
This is defined as an adverse event, the nature or severity of which is not consistent with an
expected consequence of RIPC.
Expected adverse events (recognised to be caused by the RIPC stimulus)
The benign nature of the RIPC stimulus excludes there any expected serious adverse events. The
following are expected non-serious events in response to the RIPC stimulus and will be recorded on
the case report form. They do not to be reported to the trial office.
(i)
Skin petechiae secondary to cuff inflation
Expected serious adverse events related to usual clinical care
These events are recognised complications of contrast-enhanced CT scans. They will be recorded on
the case report form. They do not need to be reported separately on a serious adverse event form.
(i)
Anaphylaxis due to contrast allergy
(ii)
Hypersensitivity reaction to contrast
(iii)
Phlebitis at intravenous cannula sites
The following events are recognised complications of routine clinical care and for the purposes of
this trial will not be designated as SAEs. They do not need to be reported.
(i)
Complications of surgery
(ii)
Complications of routine clinical care e.g.
(iii)
Complications due to administration of anaesthetic drugs
(iv)
Known adverse effects of other drugs used in routine clinical care
Reporting unexpected adverse events
Investigators will make their reports of all unexpected adverse events, whether serious or not, to the
trial office.
Unexpected Serious Adverse Events
Serious adverse events (other than those described 11.6 above) should be reported to the trial office
within 24 hours. The report should include an assessment of causality by the Principal Investigator
at each site. The Chief Investigator will be responsible for the prompt notification of findings that
could adversely affect the health of subjects or impact on the conduct of the trial. Notification of
confirmed unexpected SAEs will be to the sponsor, Ethics Committee and Data Monitoring
Committee. All deaths will be reported to the sponsor irrespective of whether the death is related
to a contrast-enhanced CT scan or to an unrelated event.
Unexpected Non-serious Adverse Events
Unexpected non-serious adverse events will be evaluated by the Local Principal Investigator. This
should include an assessment of causality and intensity 9See sections 11.7.3 and 11.7.4). Reports
should be lodged with the trial office within 14 days. The trial office will maintain detailed records of
all unexpected adverse events reported. Reports will be reviewed by the Chief Investigator to
consider intensity, causality and expectedness. As appropriate, these will be reported to the
sponsor, the DMC and the Research Ethics Committee.
Intensity Assessment
The intensity of adverse events will be classified into mild, moderate or severe.
(i)
Mild: The subject is aware of the event or symptom but the event or symptom is easily
tolerated.
(ii)
Moderate: The subject experiences sufficient discomfort to interfere with or reduce his
or her usual activity level.
(iii)
Severe: Significant functional impairment; the subject is unable to carry out usual
activities and / or the subject’s life is at risk from the event.
Causality Assessment
Causality will be classified as probable, possible, unlikely or unrelated.
(i)
Probable; a causal relationship is clinically / biologically highly plausible and there is a
plausible time sequence between onset of the adverse event and administration of the
intervention.
(ii)
Possible: A causal relationship is clinically / biologically plausible and there is a plausible
time sequence between onset of the event and administration of the intervention.
(iii)
Unlikely; A causal relationship is improbable and another documented cause of the
adverse event is most plausible.
(iv)
Unrelated: A causal relationship can be definitely excluded and another documented
cause of the adverse event is most plausible.
Summary of Adverse Event Reporting
Report to Chief Investigator and
Event
Study Coordination Centre
Related SAE
Unanticipated
related
to
Within 24 hours
problems
study
and Within 24 hours
indicates risk to subjects
Related AE
Within 7 days
Privacy violation or breech Within 24 hours
Report to Ethics Committee
Within 7 days
As per Chief Investigator and
requirements of REC
As per Chief Investigator and
requirements of REC
As per Chief Investigator and
of confidentiality
Protocol Deviations
requirements of REC
Within 7 days
As per Chief Investigator and
requirements of REC
With the exception of related SAEs, the Chief Investigator will assess all other AEs for
appropriateness for reporting to the Ethics Committee based on importance of event, requirements
of the REC, regulatory and GCP guidelines.
Safety Monitoring Plan
The Trial Manager and Trial Co-ordinator will contact each site bi-weekly to assess for adverse
events, protocol deviations, progress, recruitment, accrual, and unanticipated problems. The Chief
Investigator will meet bi-weekly with the Central Study Coordinator to review these data. Adverse
events will be reported as per Adverse Events Section 7. At least one site visit will be performed by
the Trial Manager to each enrolling site to review consent forms for completion.
Measures to minimise risks
Risk
Risk to:
RIPC Intervention: Cuff Inflation
to
250mmHg
leading
to
Subject
bruising/redness on the arm
Plan to minimize risk
All personnel performing RIPC will be experienced in
correct procedures.
Research personnel will comply with Data Protection Acts.
Research documents containing any identifiable patient
Breach of confidentiality (rare)
Subject
information will be stored in locked areas. Data stored on
computers will be password protected and where
possible, identifiable patient information will be kept to a
minimum.
DIRECT ACCESS TO SOURCE DATA / DOCUMENTS
Local investigators will ensure that all study data are available for trial related monitoring, audits and
research ethics committee review.
ETHICAL CONSIDERATIONS
Consent
All patients will freely give their informed consent to participate in the trial. A patient may decide to
withdraw from the study at any time without prejudice to their future care. Only patients who
provide written consent will be randomised in the trial. If fully informed consent is not possible, the
patient will not be recruited into the study. Inpatients awaiting ce-CT will be given an information
sheet to read, following which informed consent will be obtained.
Declaration of Helsinki and Good Practice Guidelines
The study will conform to the letter and spirit of the Declaration of Helsinki and the HIQA guidelines
‘Evaluating the clinical effectiveness of health technologies in Ireland’.
Ethical committee review
Site-specific approval will be obtained from the REC for each clinical site prior to recruitment
commencing. Amendments to the protocol that require REC approval will not be instituted until the
amendment and, if applicable, revised consent form/information leaflet have been approved by the
REC. An amendment to address urgent safety measures in order to protect research participants
against immediate risks to their health or safety may be implemented immediately provided the REC
is notified within 3 days and approval sought.
DATA HANDLING AND RECORD KEEPING
Electronic data will be returned to the trial office. Data will be stored for a period of 15 years in the
Graduate Entry Medical School, University of Limerick. The use of data from the trial will be
controlled by the Chief Investigator in conjunction with the trial office.
Confidentiality
The Chief Investigator and all members of the research team will preserve the confidentiality of
participants taking part in the study and in compliance with the Data Protection Acts.
Research documents containing any identifiable patient information will be stored in locked areas.
Data stored on computers will be password protected and where possible, identifiable patient
information will be kept to a minimum.
Each participant will be identified in trial records only by a unique trial identifier. Each centre will be
responsible for maintaining a record of individual participants’ trial number, name and hospital
number, to allow later identification of participants should it prove necessary. This record of trial
numbers will be retained in the care of the lead clinician for each unit for a period of 5 years
following completion of the trial.
The study database will be maintained by the HRBCRF Galway. Adequate data back-up procedures
will be in place. A patient can request that their data be deleted at any time.
Database Development and Data Management provision for the ReDUCE Trial
All data provided to the CRFG data management team and statistics will be anonymised. Prior to
project start-up, a Service Level Agreement will be developed to define the exact responsibilities of
CRFG Data Management and the rest of the project team. A Project plan according to PRINCE
methodology will be developed for the Data Management portion of the trial.
The CRFG Data management team will develop an eCRF (electronic Case Report Form) accessible via
a standard Internet browser from all the locations. The database used will be Infermed MACRO, an
eCRF system widely used for clinical trials. It supports adherence to Clinical Trial regulations and
Good Clinical Practice (GCP) guidelines. It can export to XML, CSV, SPSS, STATA and SAS. CDISC
exports of the database structure are supported. The data will be held in a secure location with
required backups etc. The database has been validated by the software provider and CRFG has
performed the requisite Operational and Performance Qualification. The validation also covered
Data Protection requirements. Database development will follow CRFG Standard Operational
Procedures (SOP). Access to the database is provided as per SOP and an access signoff process will
be defined at start of trial.
Input will be sought from the PI’s, Statistics, Data Manager and end users to inform the eCRF
development. Once live the Data Manager will take the lead on ensuring that the eCRF follows the
trial workflow to ease data entry. As soon as is practicable a test export will be performed to ensure
that data required to measure primary and secondary endpoints is been provided. If amendments
are required the Data Manager will work with the PI’s , Statistics and Database developer to ensure
that amendments are signed off and go live as early as possible in accordance with a quality based
change management process.
The Data Manager will develop a Data Management Plan which will detail all activities surrounding
the database and management of the data.
The Data Manager will train the end users in the relevant location and provide training
documentation. It will provide advice if any location specific IT infrastructure issues are encountered
but this is unlikely considering the locations and the fact that the software uses a thin-client
infrastructure. The Data Manager will work with PI, Stats and Database developer to develop reports
to track data quality and missing data. During the course of the trial the Data Manager will review
these quality reports to detect site specific data quality issues and liaise with the site to find
solutions. Interim analyses required will be agreed at the start and the Data Manager will cleanse
the data prior to export to Stats. The Data Manager will cleanse the data when trial is completed and
assist with Database Lock procedures (as per SOP’s) before data export occurs.
CRFG will not be involved in data entry.
19
INSURANCE
This is an investigator-led trial which will be investigator-self sponsored and covered under the
auspices of the Clinical Indemnity Scheme once ethical approval has been obtained. Centres will be
covered by HSE indemnity for negligent harm providing researchers hold a contract of employment
with the HSE, including honorary contracts held by academic staff. Medical co-investigators will also
be covered by their own medical defence insurance for non-negligent harm.
20
DISSEMINATION PLAN
It is the intention of the trial team to disseminate the results of the trial as widely as possible. This is
likely to be through publication in a peer-reviewed journal, and also via presentations at National
and International Cardiology Conferences. Publication will be in accordance with the CONSORT
guidelines.
The trial will be registered before randomisation commences. The results will be
submitted for presentation at major vascular surgery conferences. It is anticipated that the trial
manuscript will be submitted for publication by 2016. The paper will be attributed to the ReDUCE
Trial Group. Individual clinicians involved in the trial will be listed by centre at the end of the trial.
REFERENCES
1.
McCullough PA. Acute kidney injury with iodinated contrast. Critical care medicine.
2008;36(4 Suppl):S204-11. Epub 2008/04/11.
2.
Brezis M, Rosen S. Hypoxia of the renal medulla--its implications for disease. The New
England journal of medicine. 1995;332(10):647-55. Epub 1995/03/09.
3.
Cadwallader RA, Walsh SR, Burrows B, Chappell CL, Cousins C. Prospective audit of crosssectional imaging and radiation exposure in general surgical patients. Annals of the Royal College
of Surgeons of England. 2011;93(1):6-8. Epub 2010/10/20.
4.
Goldfarb S, McCullough PA, McDermott J, Gay SB. Contrast-induced acute kidney injury:
specialty-specific protocols for interventional radiology, diagnostic computed tomography
radiology, and interventional cardiology. Mayo Clinic proceedings Mayo Clinic. 2009;84(2):170-9.
Epub 2009/02/03.
5.
Tehrani S, Laing C, Yellon DM, Hausenloy DJ. Contrast-induced acute kidney injury
following PCI. European journal of clinical investigation. 2013;43(5):483-90. Epub 2013/02/28.
6.
McVeigh TP, Al-Azawi D, O'Donoghue GT, Kerin MJ. Assessing the impact of an ageing
population on complication rates and in-patient length of stay. Int J Surg. 2013. Epub 2013/08/07.
7.
Walsh SR, Tang TY, Sadat U, Gaunt ME. Remote ischemic preconditioning in major vascular
surgery. Journal of vascular surgery. 2009;49(1):240-3. Epub 2008/10/03.
8.
Walsh SR, Boyle JR, Tang TY, Sadat U, Cooper DG, Lapsley M, et al. Remote ischemic
preconditioning for renal and cardiac protection during endovascular aneurysm repair: a
randomized controlled trial. Journal of endovascular therapy : an official journal of the
International Society of Endovascular Specialists. 2009;16(6):680-9. Epub 2009/12/10.
9.
Whittaker P, Przyklenk K. Remote-conditioning ischemia provides a potential approach to
mitigate contrast medium-induced reduction in kidney function: a retrospective observational
cohort study. Cardiology. 2011;119(3):145-50. Epub 2011/09/29.
10.
Er F, Nia AM, Dopp H, Hellmich M, Dahlem KM, Caglayan E, et al. Ischemic preconditioning
for prevention of contrast medium-induced nephropathy: randomized pilot RenPro Trial (Renal
Protection Trial). Circulation. 2012;126(3):296-303. Epub 2012/06/28.
11.
Igarashi G, Iino K, Watanabe H, Ito H. Remote Ischemic Pre-Conditioning Alleviates
ContrastInduced Acute Kidney Injury in Patients With Moderate Chronic Kidney Disease.
Circulation journal : official journal of the Japanese Circulation Society. 2013. Epub 2013/08/30.
12.
Smith-Bindman R, Miglioretti DL, Johnson E, Lee C, Feigelson HS, Flynn M, et al. Use of
diagnostic imaging studies and associated radiation exposure for patients enrolled in large
integrated health care systems, 1996-2010. JAMA : the journal of the American Medical
Association. 2012;307(22):2400-9. Epub 2012/06/14.
13.
Weisbord SD, Mor MK, Resnick AL, Hartwig KC, Palevsky PM, Fine MJ. Incidence and
outcomes of contrast-induced AKI following computed tomography. Clinical journal of the
American Society of Nephrology : CJASN. 2008;3(5):1274-81. Epub 2008/05/09.
14.
Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to
estimate glomerular filtration rate from serum creatinine: a new prediction equation.
Modification of Diet in Renal Disease Study Group. Annals of internal medicine. 1999;130(6):46170. Epub 1999/03/13.
15.
Braun V, Clarke V. Using thematic analysis in psychology. Qualitative Research in
Psychology. 2006;3(2):77-101.