Study reference

Study characteristics

Patiënt characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Bracey, 1999

Type of study: RCT

Setting: heart institute, postoperatively admitted to ICU

Country: USA

Source of funding: unknown

Inclusion criteria: Patiënts who underwent first time, elective coronary artery bypass graft surgery at the Texas Heart Institute (THI) between February 4, 1997, and November 15, 1997.

Exclusion criteria: preoperative Hb level <12 g per dL, history of bleeding diathesis, use of an intraaortic balloon pump, blood loss >2500 mL within 24 hours of operation, and the patiënt’s refusal of blood transfusion for religious reason

N total at baseline:

Intervention: 212

Control: 216

Important prognostic factors²:

age ± SD:

I: 61 ± 11

C: 62 ± 11

Sex:

I: 82% M

C: 83% M

Groups comparable at baseline? yes

Describe intervention (treatment/procedure/test):

Restrictive group received RBC transfusion at a Hb level <8.0 g/dl

RBC transfusion in the postoperative period at a Hb level less than 8 g per dL, unless the patiënt experienced blood loss >750 mL since the last transfusion, hypovolemia with hemodynamic instability and excessive acute blood loss, acute respiratory failure or inadequate cardiac output and oxygenation, or hemodynamic instability requiring the use of vasopressors.

Describe control (treatment/procedure/test):

Liberal group received transfusions at Hb level <9.0 g/dl

Patiënts who received transfusions on the instructions of their individual physicians, who considered the clinical assessment of the patiënt and the institutional guidelines, which propose a Hb level <9 g per dL as the postoperative threshold for RBC transfusion.

Length of follow-up: till discharge

Loss-to-follow-up:

-

Incomplete outcome data:

-

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality:

I: 3 (1.4%)

C: 6 (2.7%)

Length of hospital stay:

I: 7.5 ± 2.9

C:7.9 ± 4.9

Blood usage (units) (postoperatively):

mean

restrictive: 0.9 ±1.5

control: 1.4 ± 1.8

(p= 0.005).

Complications

Neurologic deficit

I: 11 (5%)

C: 9 (4%)

Pulmonary complications:

I: 57 (27%)

C: 64 (30%)

Renal failure:

I: 8 (4%)

C: 5 (2%)

Infection rates

I: 5 (2%)

C: 3 (1%)

Cardiac events

Atrial arrhythmia

I: 30 (14%)

C:40 (19%)

Ventricular arrhythmia:

I: 13 (6%)

C: 9 (4%)

Myocardial infarction:

I: 1 (0,5%)

C: 0 (0%)

Hajjar, 2010

NICE 123

Type of study: RCT

Setting: heart institute of university hospital

Country: Brazil

Source of funding: none reported

Inclusion criteria:

Patiënts who were undergoing CABG surgery or cardiac valve replacement or repair, alone or in combination

Exclusion criteria:

younger than 18 years; surgery without cardio-pulmonary bypass; emergency procedure ; ascending and descending thoracic aortic procedures; left ventricular aneurysm resection; inability to receive blood products; enrolment in another study; chronic anemia (preoperative haemoglobin concentration less than 10 g/dL); low platelet count (preoperative platelet count less than 150

N total at baseline:

Intervention: 255

Control: 257

Important prognostic factors²:

Mean age ± SD:

Restrictive: 58.6 ± 12.5

C: 60.7 ± 12.5

Sex:

Restrictive:60 % M

Liberal: 64% M

Type of surgery (restrictive vs. liberal)

GABG: 152 (61%) vs. 155 (61%)

Valve: 80 (32%) vs. 79 (31%)

CABG+Valve: 17(7%) vs.19 (8%)

Groups comparable at baseline? yes

Describe intervention (treatment/procedure/test):

Restrictive group were transfused if haematocrit values were less than 24%

Describe control (treatment/procedure/test):

Liberal group were transfused RBC if the haematocrit was less than 30% at any time from the start of surgery until discharge from the ICU

Length of follow-up:

Excluded patiënts:

10 patiënts (1.9%) (4 in the liberal-strategy group and 6 in the restrictive-strategy group) were excluded after consent because surgery was performed without cardiopulmonary bypass,

Incomplete outcome data:

-

Outcome measures and effect size (include 95%CI and p-value if available):

composite end point— (all-cause 30-day mortality, cardiogenic shock, ARDS, or acute renal injury requiring dialysis or hemofiltration during the hospital stay):

Restrictive: 11%

Liberal: 10%

Mortality (30 days):

Restrictive: 6%

Liberal: 5%

Cardiogenic shock:

Restrictive: 9%

Liberal: 5%

Cardiac complications:

Restrictive: 24%

Liberal: 21%

OR:

Infectious complications:

Restrictive: 12%

Liberal: 10%

Acute respiratory distress syndrome:

Restrictive: 2%

Liberal: 1%

Respiratory complications:

Restrictive: 11%

Liberal: 11%

No. of days in hospital (median):

Restrictive: 9

Liberal: 9

TRACS

Johnson, 1992

NICE

Type of study: RCT

Setting: operation theatre and IC afterwards

Country: USA

Source of funding: dr. Kruskall is in part supported by Transfusion Medicine Academic Award K07HL02033 from the National Institutes of Health

Inclusion criteria:

autologous blood donors undergoing elective myocardial revascularisation, who were able to donate at least three units of packed cells before their operation

Exclusion criteria: not enough donated blood, operation too soon after catheterization, excerise test on day five or six in weekend

N total at baseline:

Intervention: 20

Control: 18

Important prognostic factors²:

Mean age ± SD:

Restrictive: 58.2 ± 7.5

C: 60.5 ± 6.9

Sex:

Restrictive: 100% M

Liberal: 89% M

Groups comparable at baseline? yes

Restrictive group

Received blood for haematocrit values less than 25%

Liberal group:

Received blood to achieve a haematocrit value of 32% as long as autologous blood was available

Length of follow-up:

Till sixth postoperative day

Loss-to-follow-up:

Intervention:

N (%)

Reasons (describe)

Control: 1

N (%)

Reasons (describe): intraoperative great vessel injury necessitating all his autologous blood, eliminated because of continued hemodynamic instability

Incomplete outcome data:

-

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality (30 days):

Restrictive:

Liberal:

Perioperative myocardinfarct:

Restrictive: 0

Liberal: 1

Arrhythmias:

Restrictive:4

Liberal: 6

Infectious complications:

Not reported

Respiratory complications:

Pulmonary oedema

Restrictive: 0

Liberal: 1

Pneumothorax:

Restrictive: 2

Liberal: 0

Units transfused (mean):

Restrictive: 1.0 ± 0.86

Liberal: 2.0 ± 0.93

No. of days in hospital (median):

Restrictive: 7.9 ± 4.3

Liberal: 7.6 ± 1.9

Autologous donation

Murphy, 2015

NICE 203

Type of study: a multicenter, parallel-group, randomized,

controlled trial

Setting: nonemergency cardiac surgery

Country: Uk

Source of funding: Supported by the NIHR Health Technology Assessment program (ref: 06/402/94). Dr. Reeves and the research nurse team in Bristol were supported in part by the NIHR Bristol Biomedical Research Unit in Cardiovascular Disease, and Drs. Murphy, Angelini, and Rogers were supported by the British Heart Foundation (ref: CH/12/1/29419 and CH/92027).

Inclusion criteria: >16 years of age, undergoing nonemergency cardiac surgery

Exclusion criteria:

emergency cardiac surgery, prevented from having blood and blood products according to a system of beliefs (e.g. Jehovah’s Witnesses), congenital or acquired platelet, red cell or clotting disorders (patiënts with iron deficient anaemia are not excluded), ongoing or recurrent sepsis, unable to give full informed consent, critical limb ischaemia (defined as rest pain in affected limb associated with peripheral vascular disease), already participating in another study

N total at baseline:

Intervention: 1000

Control: 1003

Important prognostic factors²:

Age (median) ± SD:

Restrictive: 69.9

Liberal: 70.8

Sex:

Restrictive: 69% M

Liberal: 68%M

Groups comparable at baseline? Yes

Type of surgery

CABG 408 (40,8%) vs 408 (40,7%)

valves: 307 (30,7%) vs 304 (30,3%)

combined: 196 (19,5%) vs. 203(20,3%)

Major aortic: 54 (5,4%) vs. 62 (6,2%)

Describe intervention (treatment/procedure/test):

Restrictive (Experimental) Group:

Continue monitoring Hb / Hct level.

Eligible for transfusion if Hb < 7.5g/dL / Hct < 22

Describe control (treatment/procedure/test):

Liberal (Control) Group:

Eligible for transfusion when

Hb < 9.0g/dL / Hct <

Length of follow-up: 3 months

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality (30 days):

Restrictive: 26/1000 (2.6%)

Liberal: 19/2003 (1.9%)

Mortality (90 days):

Restrictive: 42/1000 (4.2%)

Liberal: 26/1003 (2.6%)

Odds ratio: 1.64 (1.00-2.67), p=0.045

Myocardinfarct:

Restrictive: 3/987 (0.3%)

Liberal: 4/981 (0.4%)

Ischemic event:

Restrictive: 156/991 (15.7%)

Liberal: 139/991 (14.0%)

OR: 1.16 (0.90-1.49)

Infectious event:

Restrictive: 238/936 (25.4%)

Liberal: 240/954 (25.2%)

Clinically significant pulmonary complications:

Restrictive: 127/979 (13%)

Liberal: 116/982 (11.8%)

Patiënts transfused:

Restrictive: 637 (63.7%)

Liberal: 952 (94.9%)

Units transfused:

Restrictive: 1494

Liberal: 2494

No. of days in hospital (median):

Restrictive: 7.0

Liberal: 7.0

Study protocol details described in Brierley et al., 2014

Most patiënts had undergone coronary-artery bypass grafting (40.7%) or valve surgery (30.5%). A total of 25.7% of the participants received a red-cell transfusion before randomization

Shehata, 2012

NICE 278

Type of study: pilot RCT

Setting: university-affiliated tertiary care centre

Country: Canada

Source of funding: This study was supported by Canadian Blood Services SPF

XT00070.

Inclusion criteria: adult patiënts undergoing cardiac surgery with a CARE score (a score for cardiac surgery patiënts used to predict morbidity and mortality) of 3 or 4,17 or patiënts of advanced age defined as greater than or equal to 80 years.

Exclusion criteria: refusal of participation, unable to receive or refused blood products, or involved in autologous predonation program

N total at baseline:

Intervention: 25

Control: 25

Important prognostic factors²:

Mean age ± SD:

I: 67.2 ± 11.2

C: 68.8 ± 9.2

Sex:

I: 68% M

C: 80% M

Surgical procedure (restrictive vs. liberal):

CABG: 13 (52%) vs. 15 (60%)

Valve: 5 (20%) vs. 1 (4%)

Combined: 7 (28%) vs. 9 (36%)

Groups comparable at baseline? yes

Describe intervention (treatment/procedure/test):

Patiënts allocated to a “restrictive” transfusion strategy received RBC transfusions if their Hb was 70 g/L or less during cardiopulmonary bypass (CPB) and 75 g/L or less postoperatively after bypass.

Describe control (treatment/procedure/test):

Patiënts allocated to a “liberal” transfusion strategy received RBC transfusions if their Hb concentration was 95 g/L or less during and less than 100 g/L after bypass. A lower Hb concentration for transfusion was used during CPB because metabolic demands are reduced during anesthesia and CPB. When a threshold was reached, each group was to receive 1 unit of RBCs administered at a time followed by determination of the Hb concentration.

Length of follow-up: unclear

Adherence

For patiënts allocated to the restrictive strategy, adherence was 69% in the operating room, 82% in the intensive care unit, and 100% on the ward. For the liberal group, adherence was 40% in the operating room, 69% in the intensive care unit, and 14% on the ward. There were 10 RBC units in the restrictive group and 3 units in the liberal group that were transfused in the absence of a Hb transfusion threshold being attained. Thus adherence

overall was 84% in the restrictive group and 41% in the liberal group. Reasons for nonadherence included patiënt refusal, the use of hemoconcentration, hemorrhage, the Hct and not the Hb concentration being used to transfuse the patiënt, confusion about the transfusion strategy and whether it was to be applied on the ward, hyperkalemia, the physician planning to repeat the complete blood count before transfusing the patiënt, and the perfusionist being occupied.

Loss-to-follow-up:

-

Incomplete outcome data:

-

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality (in hospital):

Restrictive: 4/25 (16%)

Liberal: 1/25 (4%)

Myocardinfarct:

Restrictive: 1/25 (4%)

Liberal: 0/25 0%)

Pneumonia:

Restrictive: 4/25

Liberal: 0/25 (0%)

Patiënts transfused (intraoperative):

Restrictive: 6/25 (24%)

Liberal: 15/25 (60%)

Patiënts transfused (postoperative, IC):

Restrictive: 11/25 (44%)

Liberal: 17/25 (68%)

No. of days in hospital (mean ±sd):

Restrictive: 9.0 ± 12

Liberal: 7.0 ± 4

A CARE score of 3 is defined as patiënts with uncontrolled medical problems such as unstable angina treated with intravenous heparin or nitroglycerin, use of a preoperative intraaortic balloon pump, heart failure with pulmonary or peripheral edema, uncontrolled hypertension, renal insufficiency (creatinine level greater than 140 mmol/L), debilitating systemic diseases, or patiënts in whom a complex surgery is undertaken (i.e., reoperation, combined valve and coronary artery surgery, multiple valve surgery, left ventricular aneurysmectomy, repair of ventricular septal defect after myocardial infarction, or CABG of diffuse or heavily calcified vessels). A CARE score of 4 is defined as patiënts with any uncontrolled medical problem and in whom a complex surgery is undertaken.

Willems, 2010

Type of study: RCT

Setting: tertiary pediatric intensive care units

Country: Canada

Source of funding: supported, in part, by Grants 84300 and 130770 from the Canadian Institutes of Health Research (CIHR) and Grant 13904 from the Fonds de la Recherche en Sante´ du Que´bec (FRSQ).

Inclusion criteria:

TRIPICU study enrolled stabilized critically ill children from 19 tertiary care pediatric intensive care units (ICUs). Children aged between 3 days and 14 yrs, with at least one hemoglobin concentration of <=95 g/L within the first 7 days after pediatric ICU admission

Exclusion criteria:

Expected to stay < 24 hours in PICU, lack of approval from the physician, hemodynamically unstable, acute blood loss, weight < 3 Kg , cardiovascular problems, never discharged from neonatal ICU, hemolytic anemia, , enrollment in another study, other causes see figure 1

N total at baseline:

Restrictive (I): 63

Liberal (C): 62

Important prognostic factors²:

age ± SD:

I: 31.4 ± 38.1

C: 26.4 ± 39.1

Sex:

I: 57% M

C: 57% M

Groups comparable at baseline? yes

Describe intervention (treatment/procedure/test):

restrictive

(haemoglobin <70 g/L)

Describe control (treatment/procedure/test):

liberal (haemoglobin <95 g/L)

Length of follow-up:

No patiënts lost to follow-up and adherence to research protocol was excellent

Seven patiënts in the restrictive group and one in the liberal group were suspended temporarily from the transfusion protocol. Length of suspension was comparable (1.1 ± 0.4 days in the restrictive and 1.0 day in the liberal group). During the suspension period, seven red-cell transfusions were given in the restrictive group and one in the liberal group.

Incomplete outcome data:

-

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality (30 days):

Restrictive: 2/63 (3%)

Liberal: 2/62 (3%)

Ischemic event:

Not reported

Nosocomial infections:

Restrictive: 12 (19%)

Liberal: 12 (19%)

Number of transfusions:

Restrictive: 13

Liberal: 82

No. of days IC (median):

NOTE: different numbers in abstract and table 3.

Abstract:

Restrictive: 7.0 ± 5.0

Liberal: 7.4 ± 6.4

Table (total PICU stay after randomisation):

I:4,6 ±3,1

C: 4,7±4,6

subgroup analysis focused on patiënts included in the TRIPICU study who had undergone cardiac surgery or catheterization:

Leukocyte reduced allogeneic red-cell units

Cholette, 2017

Type of study:

RCT

Setting:

hospital

Country:

USA

Source of funding:not stated

Inclusion criteria:

congenital heart disease

requiring operation, and weight of 10 kg or less for the

consistent use of our infant cardiopulmonary bypass

(CPB) circuit

Exclusion criteria:

transition to extracorporeal

membrane oxygenation (ECMO) from CPB,

Hb greater than 17.5 g/dL, isolated patent ductus arteriosus,

or non–English–speaking parent/guardian

N total at baseline: 162

Intervention: 82 (53 biventricular, 29 palliative)

Control: 80 (52 biventricular, 28 palliative)

Important prognostic factors²:

Age in days (median (range):

I: 84 (2–914)

C: 43 (1–730)

Male Sex:

I: 40 (49%)

C: 45 (56%)% M

Weight in KG (median, (range)):

I: 4.5 (2.0–10.0)

C: 4.0 (2.3–9.4)

PCICU admission Hb, g/dL

I: 11.9 ± 1.4

C: 11.9 ± 1.7

Groups comparable at baseline?

yes

Conservative

RBC transfusion for a hemoglobin less

than 7.0 g/dL for biventricular repairs or less than 9.0 g/dL

for palliative procedures plus a clinical indication

Red cell

transfusion was ordered immediately on breaching the

Hb threshold in the liberal arm. In contrast, the combination

of both breaching the Hb threshold in conjunction

with a clinical indication as determined by the

attending physician was required for transfusion in the

conservative group. “Clinically indicated” transfusions

were based on the judgment of the PCICU attending

physician (ie, hypotension, tachycardia).

Liberal

RBC transfusion for a hemoglobin less

than 9.5 g/dL for biventricular repairs or less than 12 g/dL

for palliative procedures regardless of clinical indication

Length of follow-up:

The transfusion strategy was initiated at admission to

the pediatric cardiac intensive care unit (PCICU) and was

maintained until 1) transfer from PCICU service, 2)

postoperative day (POD) 28, 3) decision to cannulate for

ECMO, or 4) death.

Loss-to-follow-up:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Incomplete outcome data:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality (30 days): not reported.

In hospital death:

Restrictive: 6 (7%)

Liberal: 5 (6%)

P=0.97

Thrombosis

Restrictive: 7 (9)

Liberal: 4 (5)

P: 0.37

Myocardinfarct:

NR

Ischemic event:

Restrictive:

Liberal:

Cerebral vascular accident

Restrictive: 1 (1)

Liberal: 0

P=0.32

Infectious event:

NR

Clinically significant pulmonary complications:

NR

Patiënts transfused:

NR

Units transfused:

Number of RBC transfusions

Restrictive: 116

Liberal: 166

Nu. Of RBC transfusions (median (range)):

I: 1 (0–10)

C: 0 (0–10), p<0.01

No. of days in hospital (median):

Restrictive: 13 (4–143)

Liberal: 13 (6–133)

P=0.42

Rapporteren ook veel niet benoemde uitkomstmaten (lactaat concentraties, hemoglobine concentraties, avO2diff)

Block

randomization (block size = 8) divided patiënts to either a

conservative or liberal transfusion strategy (stratified

randomization). Patiënts were further divided into two

strata based on the type of operation as follows: biventricular

repair or palliative (nonseptated) procedure.

Palliative procedure is defined as including all cases in

which the heart is not septated or in which there continues

to be intracardiac mixing.

All blood was prestored, leukoreduced, irradiated (CISUS

IBL 437 at 2,500 cg), ABO identical, and Rh matched.

A volume-depleted protocol was used except in rare

emergent cases when time does not allow for volume

depletion methods and citrate-phosphate-dextroseadenine

(CPDA)-1 red cells of the patiënt’s blood type

are not available. Volume depletion was accomplished by

centrifugation to remove additive solution, washing with

normal saline, or using CPDA-1 red cells that do not have

an additive solution. All RBCs were stored for less than 14

days and platelets for 3 to 5 days, and transfusion was

initiated within 60 minutes of breaching the Hb threshold.

Mazer, 2017

Type of study:

open label randomized non-inferiority trial

Setting:

hospital

Country:

International (17 countries)

Source of funding:

Canadian Institutes

of Health Research, the Canadian Blood

Services–Health Canada, the National Health and

Medical Research Council of Australia, and the

Health Research Council of New Zealand

Inclusion criteria:

18 years of age or older

who were scheduled to undergo cardiac surgery

with cardiopulmonary bypass and who had a preoperative

additive EuroSCORE I of 6 or higher.

Exclusion criteria:

unable

to receive blood products, declined blood

products, were involved in a preoperative autologous

donation program, were undergoing heart

transplantation, were having surgery solely for the

insertion of a ventricular assist device, or were

pregnant or lactating.

N total at baseline: 4860

I: 2430

C: 2430

Important prognostic factors²:

Age mean ±SD

I:72±10

C:72±10

Male Sex, n(%)

I: 1553 (63.9)

C: 1586 (65.3)

EuroSCORE I‡

I: 7.9±1.8

C: 7.8±1.9

BMI

I: 28.1±6.0

C: 28.0±5.2

Type of surgery:

CABG: 622/2429 (25.6) vs. 645/2430 (26.5)

Valve: 703/2429 (28.9) vs. 716/2430 (29.5)

CABG+ valve: 464/2429 (19.1) vs. 472/2430 (19.4)

CABG + another (non-valve): 205/2429 (8.4) vs. 203/2430 (8.4)

Other: 433/2429 (17.8) vs. 394/2430 (16.2)

Groups comparable at baseline?

yes

Restrictive

transfusion strategy received a red-cell

transfusion if their hemoglobin concentration was

less than 7.5 g per deciliter intraoperatively or

postoperatively.

If the hemoglobin concentration fell below the

appropriate threshold at any time, 1 unit of red

cells was administered at a time and was followed

with a reassessment of the hemoglobin concentration.

Red-cell transfusion was to be administered

as soon as possible after the threshold-related

hemoglobin concentration was measured and

had to be initiated within the following maximum

time frames: 2 hours for patiënts during an operation;

18 hours in the ICU, including step-down

units; or 40 hours on the non-ICU hospital ward.

Liberal transfusion strategy received

a red-cell transfusion if their hemoglobin concentration

was less than 9.5 g per deciliter intraoperatively

or postoperatively in the intensive care

unit (ICU) or if the hemoglobin concentration was

less than 8.5 g per deciliter when the patiënt was

in the non-ICU ward

Length of follow-up:

28 days or until discharche

Loss-to-follow-up:

-

Incomplete outcome data:

-

Death — no./total no. (%)

74/2427 (3.0) vs. 87/2429 (3.6), OR: 0.85 (0.62–1.16)

Stroke:

Stroke — no./total no. R: (%) 45/2428 (1.9) L: 49/2429 (2.0)

HR: 0.92 (0.61–1.38)

Myocardial infarction — no./total no. (%) 144/2428 (5.9) vs. 144/2429 (5.9) OR: 1.00 (0.79–1.27)

Lenght of stay in ICU, days median:

2.1 vs, 1.9 HR: 0.89 (0.84–0.94)

Lenght of stay in hospital days median:

8.0 vs. 8.0, HR: 0.93 (0.88–0.99)

Infection — no./total no. (%) 121/2428 (5.0) vs. 101/2429 (4.2), OR: 1.21 (0.92–1.58)

New-onset renal failure with dialysis — no./total no. (%) 61/2428 (2.5) vs. 72/2429 (3.0) HR: 0.84 (0.60–1.19)

≥1 Unit of red cells — no. (%)

1271 (52.3) vs. 1765 (72.6), OR 0.41 (0.37–0.47)

No. of units of red cells transfused, median

2 vs. 3, RR 0.85 (0.82–0.88)*

Pts receiving no RBC transfusion n(%):

1159 (47.7) vs. 665 (27.4)

Study reference

(first author, publication year)

Describe method of randomisation¹

Bias due to inadequate concealment of allocation?²

(unlikely/likely/unclear)

Bias due to inadequate blinding of participants to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of care providers to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of outcome assessors to treatment allocation?³

(unlikely/likely/unclear)

Bias due to selective outcome reporting on basis of the results?⁴

(unlikely/likely/unclear)

Bias due to loss to follow-up?⁵

(unlikely/likely/unclear)

Bias due to violation of

intention to treat analysis?⁶

(unlikely/likely/unclear)

Bracey, 1999

randomly assigned to study and control groups on the basis of the last digit of their medical record number.

likely

unlikely

likely

likely

unlikely

unlikely

unlikely

Carson, 2013

Using an automated telephone system, we implemented a permuted block randomization process stratified by clinical site and clinical diagnosis (acute coronary syndrome or stable coronary artery disease).

likely

likely

likely

unlikely

unlikely

unlikely

unlikely

Johnson, 1992

Randomized with the aid of a random numbers and an odd-even (conservative strategy-liberal strategy) designation

unlikely

unlikely

unlikely

unclear

unlikely

unlikely

unlikely

Hajjar, 2010

Opaque envelopes arranged using a random-number table were prepared by the chief statistician and opened sequentially to determine the patiënt’s treatment group. The research coordinator enrolled the participants and obtained informed consent.

unlikely

unlikely

not possible

unlikely

unlikely

unlikely

unlikely

Murphy, 2015

by means of a secure Internet-based system that concealed assignments and used cohort minimization to balance assignments according to center and type of surgery.

unlikely

unlikely

not possible

not possible

unlikely

unlikely

unlikely

Shehata, 2012

Randomization sequence was created using permuted blocks of four stratified by age and the Cardiac Anesthesia Risk Score (CARE) score.Opaque sequential sealed envelopes were used and opened at the start of surgery. The allocation sequence was generated by an independent statistician at the Ottawa Methods Centre of the Ottawa Hospital Research Institute. The research coordinator enrolled participants and assigned participants to the transfusion strategies.

unlikely

unclear

unclear

unclear

unlikely

unclear

unlikely

Cholette, 2017

Block

randomization (block size = 8)

unlikely

Unlikely

Likely (no blinding and treatment was clinically indicated by the treating physician.

Unclear (not blinded)

Unlikely

Unlikely

Unlikely

Mazer, 2018

concealed

centralized, Web-based system, stratified according

to center, with computer-generated random

permuted blocks of varying sizes from two to six.

Unlikely

Unlikely

Incluclear (no blinding, this was not possible, however might cause bias(

unlikely

Unlikely

unlikely

Per protocol analysis, however modified ITT and adjusted

analyses of the primary and secondary outcomes

was performed to examine the consistency of the treatment

effect. The modified intention-to-treat population

included all the patiënts who underwent randomization

except for those who did not undergo the

planned surgical procedure and those who withdrew

consent preoperatively.

Auteur en jaartal

Redenen van exclusie

Brierley, 2014

Protocol, geen resultaten

Murphy, 2007

Pilot study van Murphy 2015 (wel geïncludeerd) met zelfde data

Curley, 2014

Review, andere PICO

Hogervorst, 2016

Cohort study

Galas, 2013

Subgroep analyse van Hajjar, 2010 naar length of hospital stay, niet PICO

Nakamura, 2015

De studie van Nakamura et al., (2015) is een substudie/secundaire analyse van de Transfusion Requirements After Cardiac Surgery (TRACS) RCT waarin patiënten werden gerandomiseerd voor restrictief of liberaal bloedtransfusiebeleid (Hajjar, 2010). In deze subgroepanalyse werden patiënten onderverdeeld in patiënten jonger en ouder dan 60 jaar. De resultaten van de primaire RCT (Hajjar, 2010) zijn reeds geïncludeerd in deze literatuuranalyse (en die van NICE).

Reeves, 2016

Is een uitgebreide beschrijving van de TITRe2 trial (Murphy, 2015) waarbij een kosteneffectiviteitsanalyse is toegevoegd. Beschrijft dezelfde primaire data als Murphy 2015, dus exclusie.

Wang, 2016

Niet gerandomiseerd retrospectief onderzoek

Wilkinson, 2014

Gaat over congenital heart disease, sluit niet aan bij PICO. Tevens ouder dan NICE richtlijn en Cochrane analyse (2016b)

Slight. 2008;95(3):205-10

Exclusie: voldoet niet aan PICO (had ook in NICE en Cochrane moeten zitten)

Carson 2013

Cardiac catheterization is geen operatieve ingreep. Studie wordt tevens meegenomen in de analyses voor patiënten met ACS.

de Gast-Bakker. 2013;39(11):2011-9

Ouder dan NICE en Cochrane en daar zit hij niet in bij de cardiac surgery groep.

Almeida. 2011;15:30

Study design: Case control

Bagate, Intensive Care Medicine. 2013;39:S413.

Study design:: cohort studie

Balzer, Anesthesia and Analgesia. 2016;122(5):S60

Vanwege type design: retrospective study. Cardiac surgery

Costello. Annals of Thoracic Surgery. 2010;89(6):1833-41

Vanwege type design: matched case- control. Focus niet bloedtransfusie maar infectie.

Hogervorst E. Applied Cardiopulmonary Pathophysiology. 2014;18:49-50

Vanwege type design: single centre cohort study. Cardiac surgery

Hogervorst. Journal of Cardiothoracic & Vascular Anesthesia. 2016;30(2):363-72

Vanwege type design: cohort study (data 1997-2013). Cardiac surgery

Mikroulis. Evidence-Based Medicine. 2011;16(4):107-8

Onduidelijk design.

Wang. Journal of Cardiothoracic & Vascular Anesthesia. 2016;30(5):1163-6

Vanwege design: retrospective, controlled study

Galas . Cardiothorac Surg. 2013;8:54

Vanwege design: retrospective study

Bhaskar,. Annals of Thoracic Surgery, 2012. 94(2): p. 460-7

Exclusie: type studie