Study ID

Allocation method

Animal model

Intervention

Outcomes

Mortality results

Large animals (pigs and sheep)

Alspaugh 2000 USA

Not specified

19 anaesthetised swine

Animals randomly assigned to two groups

Death within 2 hours

Swine of either gender averaging 50kg in body weight faster overnight but allowed free access to water. Model of uncontrolled haemorrhage produced by inflicting splenic laceration and cryogenic brain injury.

1. Pre-hospital resuscitation with Ringer’s Lactate

1. Pre-hospital 3/9

2. Delayed resuscitation until after control of haemorrhage

2. Delayed resuscitation 4/10

Bickell 1992 USA

Not specified

24 immature Yorkshire swine

Animals divided into three groups:

Death within 2 hours

Following splenectomy a stainless steel wire was placed in the infrarenal aorta. The wire was pulled, producing a 5-mm aortotomy and spontaneous intraperitoneal haemorrhage.

1. Untreated group

1. Untreated group 0/8

2. Treated group: 4mL/kg mixture of IV 7.5% NaCl and 6% Dextran-70 over one minute

2. 7.5% NaCl and 6% Dextran-70 group 5/8

3. Treated group: 80 ml/kg lactated Ringer’s solution intravenously

3. Lactated Ringer’s 8/8

Dronen 1993 USA

Not specified

17 three to four month old Yorkshire swine

Animals were randomised into two groups:

Death within 1 hour

Immature swine were instrumented and subjected to severe blood loss (40 to 46 mL/kg), when the MAP was decreased to 30 mm Hg, a 4mm tear was created in the infra-renal aorta, allowing free intra-peritoneal bleeding.

1. No resuscitation

1. No resuscitation 7/8

2. Resuscitation with normal saline infused at a rate of 6 mL/kg/min followed by shed blood at a rate of 2mL/kg/min infused to maintain a MAP of 80 mm Hg.

2. Resuscitation 7/9

Kowalenko 1992 USA

Not specified

24 immature swine

Animals were randomly divided into three groups:

Death within 1 hour

Pre-resuscitative haemorrhage is accurately controlled from a femoral artery catheter. Once animal reaches a mean arterial pressure of 30mm Hg a 4-mm aortic tear is inflicted, allowing free intra-peritoneal haemorrhage.

1. No fluids

1. No fluids 7/8

2. Saline infusion at 6mL/kg/min to reach MAP 40mmHg

2. MAP 40mmHg 1/8

3. Saline infusion at 6mL/kg/min to reach MAP 80mmHg

3. MAP 80mmHg 5/8

Owens 1995 USA

Not specified

20 immature Yorkshire swine

Animals were randomised into three groups:

Death within 2 hours

Swine were anaesthetised, arterial and venous catheters were inserted. 25-mL/kg blood was withdrawn during a 30 minute controlled haemorrhage, followed by a 20 minute uncontrolled haemorrhage from a 5mm aortotomy.

1. No resuscitation

1. No resuscitation 1/6

2. Standard resuscitation: Lactated Ringer’s infused to achieve and maintain 100% baseline cardiac index for 20 minutes

2. Standard resuscitation 1/6

3. Limited pre-hospital resuscitation: Lactated Ringer’s infused to achieve and maintain 60% baseline cardiac index for 20 minutes In the intra-operative phase intra-operative resuscitation was continued for 120 minutes using lactated Ringers to achieve 80% baseline CI.

3. Limited pre-hospital resuscitation: 0/8

Riddez 1998 Sweden

Not specified

32 swine

Animals were randomised into four groups:

Death within 1 hour

Inducing a 5mm long laceration in the infra-renal aorta induced uncontrolled heamorrhagic shock.

1. No fluid resuscitation

1. No fluid resuscitation 4/8

2. Ringer’s solution in about the same amount as the expected blood loss per hour (1:1) over 2 hours

2. Ringer’s solution (1:1) 2/8

3. Ringer’s solution in a ratio 2:1 of the expected blood loss over 2 hours

3. Ringer’s solution (2:1) 2/8

4. Ringer’s solution in a ration 3:1 of the expected blood loss over 2 hours

4. Ringer’s solution (3:1) 4/8

Sakles 1997 USA

Not specified

16 adult sheep

Animals were randomly divided into two groups:

Death within 2 hours

Sheep were anaesthetised and lacerating a branch of the pulmonary vein through an antero-lateral thoracotomy induced uncontrolled haemorrhage.

1. No resuscitation

1. No resuscitation 1/8

2. Immediate resuscitation: 30 mL/kg Lactated Ringer’s solution over a period of 10 minutes, repeated to achieve normotension.

2. Immediate resuscitation 1/8

Silbergleit 1996 USA

Not specified

10 swine

Animals were randomly divided into two groups:

Death within 1 hour

Animals were anaesthetised and uncontrolled continuous haemorrhage was induced via catheters in the femoral vessels.

1. No fluid resuscitation

1. No fluid resuscitation 3/5

2. Fluid resuscitation: 80mL/kg Lactated Ringer’s during a resuscitation phase between 10 and 20 minutes post injury.

2. Fluid resuscitation 4/5

Stern 1993 USA

Alternation

27 immature Yorkshire swine

Animals were resuscitated with a saline infusion at 6mL/kg/min as needed to maintain the desired endpoints: 

Death within 1 hour

Near-fatal porcine aortic injury haemorrhage model. Pre-resuscitative haemorrhage is accurately controlled from a femoral artery catheter. Once animal reaches a pre-determined physiologic end point, aortic tear is inflicted, allowing free intra-peritoneal haemorrhage

1. MAP 40

1. MAP 40 1/9

2. MAP 60

2. MAP 60 2/9

3. MAP 80

3. MAP 80 7/9

After 30 minutes or a total saline resuscitation 90 mL/kg, the resuscitation fluid was changed to shed blood infused 2mL/kg/min as needed to maintain the desired MAP.

Stern 1995 USA

Alternation

54 immature Yorkshire swine

Animals were divided into three groups:

Death within 1 hour

Near-fatal porcine aortic injury haemorrhage model. Pre-resuscitative haemorrhage is accurately controlled from a femoral artery catheter. Once animal reaches a pre-determined physiologic end point, aortic tear is inflicted, allowing free intra-peritoneal haemorrhage.

1. MAP 40

1. MAP 40 2/18

2. MAP 60

2. MAP 60 3/18

3. MAP 80 mm Hg

3. MAP 80 14/18

Resuscitation fluid was either shed blood followed by normal saline or normal saline followed by shed blood.

Stern 2000 USA

Not specified

24 swine

Animals were randomly assigned to one of three groups.

Death within 150 minutes

Each swine underwent fluid percussion brain injury and uncontrolled haemorrhage to a mean arterial pressure of 30mm Hg in the presence of a 4mm aortic tear.

1. No resuscitation

1. no resuscitation 6/6

2. MAP 60 mm Hg

2. MAP 60 mm Hg 1/9

3. MAP 80 mm Hg

3. MAP 80 mm Hg 4/9

Small animals (rats)

Bilynskyj 1992 Brazil

Not specified

120 male Wistar rats

Animals were divided into four anaesthetic groups and each of these were divided into two groups:

Deaths within 4 hours

Uncontrolled haemorrhage was induced by 12% or 50% tail resection

1. No fluid resuscitation

1. No fluid resuscitation 9/60

2. Hypertonic saline (7.5%) 4mL/kg IV. 15 minutes from start of bleeding.

2. Hypertonic saline 13/60

Burris 1999 USA

Not specified

61 male Sprague-Dawley rats

Fluids infused at 2 mL/Kg per minute were turned off or on to maintain a mean arterial pressure of 40, 80, or 100 mm Hg in six groups:

Death within 70 minutes

After instrumentation the heamodynamically stable but lightly anaesthetised rats were subjected to a vascular injury leading to uncontrolled haemorrhagic shock by piercing the infra-renal aorta with a 25-guage needle, creating two standard sized holes one each side of the aorta.

1. No fluid

1. No fluid 11/11

2. Lactated Ringer’s MAP 100

2. Ringer’s (MAP 100) 6/9

3. Lactated Ringer’s MAP 80

3. Ringer’s (MAP 80) 2/11

4. Lactated Ringer’s MAP 40

4. Ringer’s (MAP 40) 7/10

5. Hypertonic NaCl/hetastarch (MAP 80)

5. NaCl/hetastarch (MAP 80) 8/11

6. Hypertonic NaCl/hetastarch (MAP 40)

6. NaCl/hetastarch (MAP 40) 3/9

Capone 1995a USA*

Randomised in blocks of four, allocation concealment not specified

40 male Sprague-Dawley rats

Experimental design consisted of three phases: Pre-hospital phase: from insult to haemostasis 60 minutes later. During this phase uncontrolled haemorrhagic shock was produced so that different fluid resuscitation regimens could be tested. Hospital phase: started with haemostasis and continued for another 120 minutes, with unrestricted fluid resuscitation, including infusion of the donor blood to Hct 30% plus lactated Ringer’s solution to achieve normotension. 

Death within 3 days

Uncontrolled haemorrhagic shock produced by a preliminary bleed (3 mL/100g) followed by 75% tail amputation.

Observation phase extended from the end of the hospital phase to 3 days to evaluate outcome.

1. untreated controls

2. no pre-hospital fluid

2. no fluid 9/10

3. LR pre-hospital to maintain MAP at 40 mm HG

3. LR pre-hospital (MAP 40) 4/10

4. LR pre-hospital to maintain MAP at 80 mm HG

4. LR pre-hospital (MAP 80) 10/10

Group 1 not resuscitated in hospital phase is excluded.

Capone 1995b USA

Not specified

40 male Sprague-Dawley rats

Experimental design consisted of three phases: a ‘pre-hospital phase’ (90 minutes of uncontrolled bleeding with or without Lactated Ringer’s solution) followed by a ‘hospital phase’ (60 minutes including control of bleeding and fluid resuscitation including blood) 

Death within 3 days

Haemorrhage induced by an oblique sterile tail amputation of 75% its length, as measured from the tip.

1. untreated controls

2. no pre-hospital fluid

2. no pre-hospital fluid 1/10

3. LR pre-hospital to maintain MAP at 40 mm HG

3: LR pre-hospital (MAP 40) 0/10

4. LR pre-hospital to maintain MAP at 80 mm HG

4. LR pre-hospital (MAP 80) 3/10

Groups 2,3 and 4 only were resuscitated in the hospital phase. Group 1 therefore not included.

Capone 1995c USA*

Not specified

20 male Sprague-Dawley rats

Experimental design consisted of three phases:

Death within 3 days

Volume controlled haemorrhage of 3mL/100g over 15 minutes followed by uncontrolled haemorrhage induced by an oblique sterile tail amputation of 75% its length, as measured from the tip.

Pre-hospital phase: from insult to haemostasis 90 minutes later. During this phase uncontrolled haemorrhagic shock was produced so that different fluid resuscitation regimens could be tested.

Hospital phase: started with haemostasis and continued for another 60 minutes, with unrestricted fluid resuscitation, including infusion of the donor blood to Hct 30% plus lactated Ringer’s solution to achieve normotension.

Observation phase extended from the end of the hospital phase to 3 days to evaluated outcome.

1. no fluid resuscitation in either phase

1. no resuscitation 10/10

2. no resuscitation in pre-hospital phase, then in hospital phase all out fluid resuscitation as above.

2. hospital resuscitation 9/10

Craig 1994 USA

Not specified

45 male Sprague-Dawley rats

Animals randomly divided into five groups:

Death within 3 hours

Rats were anaesthetised and a femoral artery and vein were cannulated for blood pressure monitoring and fluid infusion. Through a mid-line abdominal incision the distal ileocolic artery and vein were cut and allowed to bleed freely into the abdominal cavity. The abdomen was closed and the animals were randomised into five groups.

1. no resuscitation

1. no resuscitation 2/9

2. small volume lactated Ringer’s solution

2. small volume lactated Ringer’s solution 3/9

3. large volume lactated Ringer’s solution

3. large volume lactated Ringer’s solution 4/9

4. small volume hetastarch

4. small volume hetastarch 8/9

5. large volume hetastarch

5. large volume hetastarch 4/9

Elgjo 1996 Norway

Not specified

39 male Wistar-Kyoto rats

Animals in each group were randomised into two groups:

Death within 4 hours

Arterial haemorrhage

Hypotension was induced by controlled withdrawal of blood until MAP reached 50 mm Hg. After 5 min subjects were randomised into two groups for uncontrolled haemorrhage: (1) arterial haemorrhage – the abdominal aorta puncture with a 21,22, or 23 gauge hypodermic needle; (2) venous haemorrhage – inferior caval vein puncture with a 18 or 19 gauge hypodermic needle.

1. No treatment

1. No treatment: 5/12

2. Saline 8.0% 2mL/kg IV at 0.4mL/min.

2. Saline 8.0%: 6/10

Venous haemorrhage

1. No treatment: 0/7

2. Saline 8.0%: 1/10

Feldman 1997 Israel

Not specified

56 male Sprague-Dawley rats

Randomly allocated to early or late fluid resuscitation with warmed lactated Ringer’s solution. Volume infused was three times the total amount of blood lost.

Death within 2 hours

No closed head injury

Halothane anaesthesia and divided into groups with or without closed head trauma. Head trauma was delivered to the skull over the frontal portion of the left cerebral hemisphere by a weight drop device. The 30 rats that survived were head injury were randomly divided into groups. Haemorrhagic hypotension was produced by 12% tail resection.

1. Early resuscitation: IV fluids started 1.25 hours after injury.

1. Early resuscitation 0/10

2. Delayed resuscitation: IV fluids started 2 hours after injury.

2. Delayed resuscitation 0/10

Closed head injury

1. Early resuscitation 3/10

2. Delayed resuscitation 3/10

Greene 1998 USA

Not specified

63 female Sprague-Dawley rats

Rats were randomly divided into three groups:

Death within 6 hours

The day before the study the animals were cannulated through the jugular vein. Under light ether anaesthesia the rats were restrained and haemorrhage was initiated by a 75% tail resection with a guillotine.

1. not resuscitated