Study Design

Reduced Quality Factors

0 = no serious, - = serious,

- - = very serious

Positive Quality Factors

0 = none, + = one, ++ = multiple

Dichotomous Outcome Summary

Non-Dichotomous Outcome Summary

Brief description

Overall Quality

High, moderate, low,
very low, none

No of studies

Study Type

RoB

Indirectness

Imprecision

Inconsistency

Large Effect

Dose-Response

Confounder

# Intervention with Outcome

# Control with Outcome

RR (95% CI)

Outcome: Favorable neurologic outcome

2

OBS

-

- -

-

0

0

0

0

Two human clinical observational trials - one found increased CC depth increased odds of survival with a favorable neurological outcome. The second, smaller study showed no difference in neurologic outcome with CC depth

Very low

Outcome: Survival to discharge

3

OBS

0

- -

-

0

0

0

0

Three human observational studies (1 large), two of which found increased CC depth increased survival to discharge. One very small study found no difference

Very low

Outcome: ROSC

4

OBS

-

- -

-

-

0

0

0

Three human studies found increasing CC depth was associated with ROSC, one very small study found no difference

Very low

2

EXP

0

- -

-

-

0

0

0

Two swine VF studies - one found increasing CC depth was associated with ROSC, the other did not.

Very low

Outcome: Surrogate markers of perfusion

6

EXP

0

- -

-

0

0

0

0

Five swine and one canine study all found improved hemodynamic parameters with increased depth of CC.

Very low

Outcome: Complications

1

OBS

0

-

0

0

0

0

0

One human clinical study found in male patients increasing depth of CC was associated with injury but not in women.

Very low

3

EXP

0

-

0

-

0

0

0

Three swine studies, two showed no relationship between CC depth and injury. One documented an increased incidence of epicardial hemorrhage with increasing CC depth

Very low

Introduction

Chest compressions during CPR aim to generate blood flow through either direct compression of the heart or secondary to global increases in intrathoracic pressure. The depth of chest compressions is likely to have some relationship with cardiac output during CPR but this benefit needs to be weighed against the potential for harm with increasing compression depth. The current human CPR guidelines recommend a chest compression depth of approximately 5 cm while avoiding excessive compression depths (greater than 6 cm).1 The previous veterinary guidelines suggested that a compression depth of one third to one half of the width of the chest was reasonable.2

Consensus on science

Outcome 1: Favorable neurologic outcome

For the most critical outcome of FNO, we identified 2 observational studies of OHCA in people (very low quality of evidence, downgraded for serious risk of bias, very serious indirectness, and serious imprecision) that address the PICO question. 3,4 One before-and-after study of 593 adults comparing CC depth prior to and after institution of a comprehensive CPR quality improvement initiative found that each 5 mm increase in mean CC depth significantly increased the odds of survival with favorable functional outcome with an adjusted OR of 1.30 (95% CI 1.00–1.70).4 In another before-and-after study of 32 people with OHCA, the use of a real-time audiovisual feedback (RTAVF) device increased CC depth from 38.8±11.5 mm to 48.0±9.2 mm, while no change was noted in CC depth when no feedback was provided. No difference in FNO was found between the 2 groups of 16 people each.3

Outcome 2: Survival to discharge

For the next critical outcome of survival to discharge, we identified 3 observational studies (very low quality of evidence, downgraded for very serious indirectness and serious imprecision) that address the PICO question.3–5 In a large observational study of 9136 adults with OHCA, the adjusted odds ratio for survival to discharge was 1.04 (95% CI, 1.00–1.08) for each 5-mm increment in compression depth, 1.45 (95% CI, 1.20–1.76) for cases with a depth range (> 38 mm), and 1.05 (1.03 – 1.08) for percentage of minutes within depth range. Covariate-adjusted spline curves revealed that the maximum survival in these adult people is at a depth of 45.6 mm (15-mm interval with highest survival between 40.3 and 55.3 mm); no differences were found between male and female patients.5 A smaller before-and-after study of 593 adults comparing CC depth prior to and after institution of a comprehensive CPR quality improvement initiative found that each 5 mm increase in mean CC depth increased the odds of Survival to discharge with an adjusted OR of 1.29 (95% CI 1.00 – 1.65).4 A very small before-and-after study of 32 people with OHCA found that the use of a real-time audiovisual feedback (RTAVF) device increased CC depth from 38.8±11.5 mm to 48.0±9.2 mm, while no change was noted in CC depth when no feedback was provided. No difference in survival was found between the 2 groups of 16 people each.3

Outcome 3: ROSC

For the critical outcome of ROSC, we identified 4 observational studies in people (very low quality of evidence, downgraded for serious risk of bias, very serious indirectness, serious imprecision, and serious inconsistency)3,5–7 and 2 experimental swine studies (very low quality of evidence, downgraded for very serious indirectness, serious imprecision, and serious inconsistency).8,9 In a large observational study of 9136 adults with OHCA, the adjusted OR for ROSC was 1.06 (1.04 – 1.08) for each 5-mm increment increase in compression depth.5 The remaining 3 studies are much smaller, 2 supporting these findings and one finding no difference. In a before-and-after, observational study of 284 OHCA events comparing the use of an automated feedback system to no feedback, the feedback group had greater compression depth (38+/-6 mm vs 34+/-9 mm); logistic regression found that the average compression depth (per mm increase) had an OR of 1.05 (1.01 – 1.09, P = 0.02) for ROSC.6 In an observational study in 60 people with IHCA or OHCA, logistic regression analysis demonstrated that successful defibrillation was associated with higher mean compression depth during the 30 seconds of CPR preceding the pre-shock pause with an adjusted OR of 1.99 (1.08 – 3.66) for every 5 mm increase.7 A very small before-and-after study of 32 people with OHCA found that the use of a real-time audiovisual feedback (RTAVF) device increased CC depth from 38.8±11.5 mm to 48.0±9.2 mm, while no change was noted in CC depth when no feedback was provided. No difference in ROSC was found between the 2 groups of 16 people each.3

An experimental swine VF model compared optimal CC depth (25% = 6 cm) in an anterior-posterior (A-P) direction with a CC depth of 4.2cm (70% of optimal = ~17% chest diameter); this study found that greater A-P CC depth was associated with ROSC (P = 0.0004).8 In an experimental ventricular fibrillation (VF) model in which swine underwent CC in dorsal recumbency, delivering CC to a depth of 35.2 – 57.0 mm (rescuer targeting 50 mm or 25% of the A-P diameter of the chest) resulted in no difference in ROSC than delivery of CC to a depth of 19.0 – 38.5 mm.9

Outcome 4: Surrogate markers of perfusion

For the important outcome of surrogate markers of perfusion, we identified 6 experimental studies (very low quality of evidence, downgraded for very serious indirectness and serious imprecision) that address the PICO question.8–13 In dogs weighing 6-12 kg positioned in dorsal recumbency and receiving 62 compressions per minute with varied CC depths, cardiac output varied with chest displacement. Mean CC depth of 1.8 +/-0.85 cm was required to achieve MAP > 0 mmHg. Increasing MAP was associated with increasing compression depth.11 A swine experimental model comparing 20% to < 14% A-P compression depth showed higher systolic arterial pressure, CoPP, ETCO2, and central venous O2 for the 20% compression group.10 In a swine experimental model, compressions to a depth of 35.2 – 57.0 mm (rescuer targeting a depth of 50 mm or 25% of the A-P diameter of the chest) resulted in a significantly higher CoPP than compressions to a depth of 19.0 – 38.5 mm (rescuer targeting 70% of “good” CPR depth, equivalent to a depth of 35 mm or ~17% of the A-P diameter of the chest) in dorsal recumbency.9 In arrested piglets receiving 3 cm compressions vs. 5 cm compressions, DAP and CoPP were significantly higher in piglets receiving 5 cm compression depths.13 Another swine study found that CoPP was better with 5 cm CC depth than with 3 cm CC depth.12 In a swine experimental VF model comparing optimal A-P CC depth (25% = 6cm) with conventional depth (4.2cm - 70% of optimal = ~17%), CoPP and ETCO2 were both significantly higher with greater depth of CC.8

Outcome 5: Complications

For the important outcome of complications, we identified 1 observational (very low quality of evidence downgraded for serious indirectness)14 and 3 experimental studies (very low quality of evidence, downgraded for serious indirectness and serious imprecision) that address the PICO question.8,9,15 Among male human patients, CPR-related injuries were associated with deeper mean and peak compression depths (P < 0.05). No such association was observed in women. The frequency of injuries in mean compression depth categories < 5, 5 – 6 and > 6 cm, was 28%, 27%, and 49%, respectively (P = 0.06).14 In a 1-2-week-old swine model, there was a significantly higher incidence of epicardial hemorrhage in the intervention group (ETCO2-guided CPR, resulting in deeper CC) compared to the control group (standard CPR).15 In an experimental swine VF model, CC to a depth of 35.2 – 57.0 mm (rescuer targeting a depth of 50 mm or 25% of the A-P diameter of the chest) found no rib fractures were evident in any pig.9 In an experimental swine study comparing optimal CC depth (25% = 6 cm) in the A-P direction with conventional depth (4.2cm - 70% of optimal = ~17%), no evidence of CPR-related injury was found on necropsy in any animal.8

Treatment recommendation

In dogs and cats that are positioned in lateral recumbency, we recommend providing chest compressions to a depth of one-third to one-half of the lateral diameter of the chest at the compression point.(strong recommendation, very low quality of evidence)

In dogs and cats that are positioned in dorsal recumbency, we recommend providing chest compressions to a depth of one-quarter the anterior-posterior diameter of the chest at the compression point.(strong recommendation, very low quality of evidence)

Justification of treatment recommendation

Available evidence shows that in people receiving anterior-posterior chest compressions, the ideal compression depth is approximately 20-25% the depth of the A-P diameter at the compression point. Considering that the spine and epaxial musculature is thicker in most animals than the lateral body wall, and considering that the main compressible anatomical part of the thorax is the lung, the committee estimates that the degree of thoracic space reduction achieved by 33 – 50% compression of the lateral diameter of the chest would achieve the same degree of thoracic space reduction as a 20 – 25% compression depth in the A-P orientation. Unfortunately, all of this evidence is based on human or animal studies in dorsal recumbency receiving sternal compressions, including the single canine study identified, which leaves open the question of specifically how deep compressions should be in the target species delivered in the presumed optimal state of lateral recumbency.

Knowledge gaps

There are no direct studies evaluating the optimal compression depth in dogs and cats of varying size and conformation in lateral recumbency, and there is limited information for dorsal recumbency.

A maximum safe compression depth in dogs and cats is unknown, and the relationship between compression depth and risk of complications in animals in lateral recumbency is likely different than that in people and pigs in dorsal recumbency.

Further, the risk of complications associated with hands over the heart (cardiac pump) versus thoracic pump (hands over the widest part of the chest) likely varies, and is unknown.