15 Study Design

16 Reduced Quality Factors

17 0 = no serious, - = serious,

18 - - = very serious

19 Positive Quality Factors

20 0 = none, + = one, ++ = multiple

21 Dichotomous Outcome Summary

22 Non-Dichotomous Outcome Summary

23 Brief description

24 Overall Quality

25 High, moderate, low,
very low, none

26 No of studies

27 Study Type

28 RoB

29 Indirectness

30 Imprecision

31 Inconsistency

32 Large Effect

33 Dose-Response

34 Confounder

35 # Intervention with Outcome

36 # Control with Outcome

37 RR (95% CI)

38 Outcome: ROSC

39 1

40 EXPT

41 0

42 - -

43 0

44 0

45 0

46 0

47 0

48 14/18 successful CPR and no difference in body position

49 Very low

50 1

51 OBS

52 -

53 -

54 0

55 0

56 0

57 0

58 +

59 OR 46.6, 95% CI 4.1-535.6

60 Very low

61 Outcome: Surrogate markers of perfusion

62 1

63 EXPT

64 0

65 0

66 - -

67 0

68 0

69 0

70 0

71 No statistical analysis; statement that pressure & flow improved with dogs in lateral recumbency. Chest conformation unknown.

72 Very low

74 Introduction

75 In clinical veterinary patients, chest compressions during CPR are most commonly performed with animals in lateral recumbency, in contrast to experimental animal and clinical human studies, in which dorsal recumbency is most common. Dorsal recumbency for chest compressions in non-wide-chested dogs is challenging without restraint or additional equipment, given their conformation. The previous veterinary guidelines make no comment regarding the role of sternal compressions in dorsal recumbency for non-wide-chested dogs.1

76 Consensus on science

77 For the most critical outcomes of favorable neurologic outcome and survival to discharge, we identified no studies addressing the PICO question.

78 Outcome 3: ROSC:

79 For the next most important outcome of ROSC we found 1 experimental animal study (very low quality of evidence, downgraded for very serious indirectness) and 1 observational study (very low quality of evidence, downgraded for serious risk of bias and serious indirectness, upgraded for effect despite confounding).2,3 One experimental study in cats with an asphyxiation model of cardiopulmonary arrest failed to demonstrate a difference between chest compressions in dorsal vs lateral recumbency.2 One 2009 observational study in a veterinary teaching hospital documented that dogs in CPA receiving chest compressions in lateral recumbency had higher odds of ROSC OR 46.6, 95% CI (4.1 – 535.6) in multivariate logistic regression. Body conformation was not reported in this study.3

80 Outcome 4: Surrogate markers of perfusion:

81 For the important outcome of surrogate markers of perfusion, we identified 1 experimental study in dogs (very low quality of evidence, downgraded for very serious imprecision) that addressed the PICO question.4 This study included 24 “mongrel dogs” weighing 25 – 35 kg whose conformations were not described, and found that manual compressions administered in lateral recumbency resulted in higher LV pressure and aortic flow than those delivered with the dog in supine position. Comparative numerical values were not reported, and no statistical analyses appear to have been performed.

82 Treatment recommendation

83 We recommend performing chest compressions in lateral recumbency in non-wide-chested dogs (strong recommendation, very low quality of evidence).

84 Justification of treatment recommendation

85 There are no studies directly comparing outcomes of dorsal versus lateral body position for chest compressions in dogs of any conformation. Despite the very low quality of evidence, this strong treatment recommendation is based on the logistical challenges of performing sternal chest compressions in non-wide-chested dogs in dorsal recumbency and the high compliance of the chest wall of many dogs in lateral recumbency. This recommendation is supported by 1 veterinary observational study that reported a higher odds of ROSC and 1 experimental study in dogs that found improved LV pressure and aortic flow when chest compressions were performed with dogs in lateral recumbency compared to dorsal recumbency.

86 Knowledge gaps

87 There are no studies that clearly report on any critical outcome for chest compressions performed in dorsal vs lateral recumbency in non-wide-chested dogs.