RECOVER 2.0 Worksheet

QUESTION ID: MON-03A

PICO Question:
In cats and dogs that have experienced ROSC after CPA (P), does measurement of glucose (I) as opposed to non-measurement (C), improve ... (O)?

Outcomes:
Duration of post-arrest hospitalization, Favorable neurologic outcome, Prediction of recurrent CPA, Survival to Discharge

Prioritized Outcomes (1= most critical; final number = least important):

1. Favorable neurologic outcome

2. Survival to Discharge

3. Prediction of recurrent CPA

4. Duration of post-arrest hospitalization

Domain chairs: Selena Lane, Ben Brainard, reviewed by MB

Evidence evaluators: Stefania Scarabelli, Mathieu Raillard

Conflicts of interest: none

Search strategy: See attached document

Evidence Review:

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: Survival to Discharge
5	OB	0	--	0	0	0	0	0				There is a consistent negative association between PCA hyperglycemia and survival to discharge across all studies	Very low
Outcome: Favorable neurologic outcome
7	OB	0	--	0	0	0	0	0					Very low
1	EXP	0	--	-	0	0	0	0					Very low

Outcome: Prediction of recurrent CPA

0

Outcome: Duration of post-arrest hospitalization

0

PICO Question Summary

Introduction

Blood glucose monitoring is widely available in veterinary hospitals, inexpensive, and minimally invasive. Severe hypoglycemia and hyperglycemia, as well as targeted protocols to tightly regulate glucose concentrations in hospitalized patients have been associated with both worse and better outcomes among the critically ill.1–3 While glucose derangements can occur in CPA patients during the post-cardiac arrest period, the role of blood glucose in the outcome of these patients is unknown.

Consensus on science

Outcome 1: Survival to discharge We identified 5 observational studies in people that examined the association between post-cardiac arrest BG concentrations and survival to discharge (very low quality of evidence, downgraded for very serious indirectness).4–8 Four of these 5 studies concern OHCA patients. Skrifvars et al (2003), in a small observational study including 96 patients, found that the mortality rate at 6 months increased with the average BG concentration over the first 72 hours after ROSC: 9% mortality (5.5-6.8 mmol/L), 23% (6.9-7.9 mmol/L), 50% (7.9-8.9 mmol/L), and 64% (9.1-27.9 mmol/L) (P<0.05).8 In another observational study including 2028 patients with ROSC after OHCA, Zhou (2020) reported increased odds for in-hospital mortality in patients with a mean BG of 7.8 - 10 mmol/L (OR, 1.62; 95%CI, 1.26 - 2.08, P<0.001) and those with a mean BG > 10 mmol/L (OR, 1.80; 95%CI, 1.40 - 2.30, P<0.001) when compared to patients with normal BG.4 Two additional observational studies including a total of 1003 OHCA patients suggest a similar association between hyperglycemia and non-survival.5,6 Hyperglycemia was also found to be common after in-hospital cardiac arrest (IHCA) in both diabetic and non-diabetic humans, with non-diabetic patients appearing to be more sensitive to variations in blood glucose concentrations, as both hypo- and hyperglycemia were associated with decreased survival in non-diabetics.7 Outcome 2: Favorable neurologic outcome With regard to neurologic outcome, we identified 7 observational studies in people (very low quality of evidence, downgraded for very serious indirectness) and 1 experimental study in swine (very low quality of evidence, downgraded for very serious indirectness and imprecision.5,6,9–14 Human studies assessing the association of BG concentrations with neurologic outcome in patients during the post-cardiac arrest period generally show that patients with hypoglycemia, hyperglycemia, and/or high variability of blood glucose concentrations have worse neurologic outcomes compared with patients with normal blood glucose and less variability of blood glucose concentrations in the post-cardiac arrest period.5,6,9–13 In an observational study from registry data including 883 OHCA patients treated with targeted therapeutic hypothermia, the mean BG concentration obtained 1 hour after ROSC was significantly lower in those with FNO (CPC 1-2) (12.8 ± 5.0 mmol/L) compared to those with poor neurological outcome (14.6 ± 7.6 mmol/L; P=0.001).6 This association between post-cardiac arrest BG and FNO was maintained when adjusting for confounders in multivariate analysis (OR, 0.955; 95% CI, 0.918-0.994, P=0.024). In an additional post-hoc analysis including 234 OHCA patients undergoing TTM, subjects were categorized according to quartiles of median glucose concentrations at 12 hours post ROSC: QI 5.6 (3.7—6.4 mmol/L), QII 7.2 (6.4—7.9 mmol/L), QIII 9.0 (7.9—10.7 mmol/L) and QIV 14.7 (10.7—25.8 mmol/L).13 In a multivariate analysis adjusting findings for relevant confounders including duration of untreated CPA and duration of CPR, those in QI (OR, 4.55; 95% CI, 1.28—16.12), and QII (OR, 3.02; 95% CI, 3.29—49.9) were more likely to survive with good FNO compared to those in QIV. Although not the objective of this study, the authors suggest that a more liberal BG target (e.g., 3.7 to 7.9 mmol/L) might be warranted, and that tight glycemic control in its original definition (i.e., 4.4 to 5.6 mmol/L) might not be necessary.1 A further post-hoc study including 939 patients hospitalized after OHCA found that median post-arrest BG concentrations were higher in patients with poor neurologic outcomes (12 mmol/L [IQR, 8.9–15.9 mmol/L]) compared to those with FNO (10 mmol/L [IQR, 7.71–13.5 mmol/L], P<0.0001) and that 1 mmol/L increase in glucose concentration was associated with a 22% increase in the probability of a poor neurologic outcome (P=0.010).11 An additional 4 observational studies, all small single center retrospective studies including a combined total of 337 patients, showed a similar association between post-cardiac arrest hyperglycemia and FNO.5,9,10,12 We identified one experimental study, using an OHCA VF pig model to determine whether early post-ROSC hyperglycemia affects neurologic function at 72 hours post-arrest.14 Animals were left untreated for 7 minutes after induction of VF, and CPR was continued for up to 15 minutes, at which time 21 of 22 animals achieved ROSC. Blood glucose was determined within the first hour of ROSC. The investigators found no association between BG and neurological function at 72 hours post ROSC, as 20 of 21 resuscitated animals recovered with FNO. We did not identify relevant studies that report the less critical outcomes of prediction of recurrent CPA, and duration of post-cardiac arrest hospitalization.

Treatment recommendation

We suggest measuring blood glucose in all dogs and cats as early as possible after return of spontaneous circulation (weak recommendation, very low quality of evidence). We recommend measuring blood glucose in dogs and cats after ROSC in which hypoglycemia or hyperglycemia are known or suspected (strong recommendation, expert opinion).

Justification of treatment recommendation

None of the studies we identified directly answer the PICO question on whether the measurement of BG itself impacts critical outcomes in dogs, cats, or any other species recovering from CPA. However, there is ample evidence to suggest that BG abnormalities in the post-cardiac arrest period are common and that there is a consistent association between high BG concentrations and reduced survival rates and poor neurological outcomes. While we did not find any veterinary studies that report on the relationship between post-cardiac arrest BG and relevant outcomes, there are studies in dogs and cats that document hyperglycemia post-ROSC. One experimental study in a VF dog model (n=26) evaluated BG and other metabolic parameters after resuscitation from 3 minutes of untreated arrest followed by 10 minutes of CPR. Blood glucose concentrations prior to CPA were 6.0 ± 1.9 mmol/L and increased to 12.2 ± 3.7 mmol/L 10 minutes after ROSC (P<0.05).15 Hopper et al (2014) retrospectively reviewed acid-base, electrolyte, glucose, and lactate values during or immediately after CPR in dogs and cats in a clinical context.16 Both hypoglycemia (9 of 42 patients; 21%) and hyperglycemia (26 of 42 patients; 62%) were identified during CPR and following ROSC in this patient population. The hypoglycemia was a result of pre-existing underlying disease and or was iatrogenic due to insulin and dextrose therapy for hyperkalemia and not the consequence of CPA. The median BG 5 minutes after ROSC was 11.6 mmol/L (range, 0.7 – 22.8 mmol/L). Taken together, these two studies suggest that post-arrest hyperglycemia is likely occurring to a similar extent in dogs and possibly cats as in people. Given the possible benefit of prognostication and the feasibility of determining BG in dogs and cats, we suggest determining post-resuscitation BG routinely. The relationship between high post-resuscitation BG concentrations and worse neurologic outcome is well established in people. Hyperglycemia may be a sequela of increased severity of injury rather than the cause of further injury, and a benefit of treatment cannot be concluded from these studies. Experimental studies show that induction of hyperglycemia leads to worse neurologic outcome compared to controls.17–19 However, treatment of post-arrest hyperglycemia in people has not been shown to improve outcome.20,21

Knowledge gaps

While there is evidence in humans to support the association between post-ROSC glucose derangements and lower survival or poor neurologic outcomes, we did not identify any veterinary studies that address the relationship between glucose concentrations in the PCA period and relevant outcomes. Future research should focus on whether blood glucose measurement after ROSC is of value to predict survival or time spent in hospital post-arrest, if there are “cut-off” blood glucose concentrations associated with FNO, and if glycemic control in the peri-arrest period has any effect on patient outcomes.

References:

1. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001;345(19):1359-1367.

2. The NICE-SUGAR Study Investigators. Intensive versus Conventional Glucose Control in Critically Ill Patients. N Engl J Med. 2009;360(13):1283-1297.

3. Yatabe T, Inoue S, Sakaguchi M, Egi M. The optimal target for acute glycemic control in critically ill patients: a network meta-analysis. Intensive Care Med. 2017;43(1):16-28.

4. Zhou D, Li Z, Shi G, Zhou J. Proportion of time spent in blood glucose range 70 to 140 mg/dL is associated with increased survival in patients admitted to ICU after cardiac arrest: A multicenter observational study. Medicine (Baltimore). 2020;99(33):e21728.

5. Russo JJ, James TE, Hibbert B, et al. Hyperglycaemia in comatose survivors of out-of-hospital cardiac arrest. Eur Heart J Acute Cardiovasc Care. 2018;7(5):442-449.

6. Kim SH, Choi SP, Park KN, et al. Association of blood glucose at admission with outcomes in patients treated with therapeutic hypothermia after cardiac arrest. Am J Emerg Med. 2014;32(8):900-904.

7. Beiser DG, Carr GE, Edelson DP, Peberdy MA, Hoek TLV. Derangements in blood glucose following initial resuscitation from in-hospital cardiac arrest: a report from the national registry of cardiopulmonary resuscitation. Resuscitation. 2009;80(6):624-630.

8. Skrifvars MB, Pettilä V, Rosenberg PH, Castrén M. A multiple logistic regression analysis of in-hospital factors related to survival at six months in patients resuscitated from out-of-hospital ventricular fibrillation. Resuscitation. 2003;59(3):319-328.

9. Müllner M, Sterz F, Binder M, Schreiber W, Deimel A, Laggner AN. Blood glucose concentration after cardiopulmonary resuscitation influences functional neurological recovery in human cardiac arrest survivors. J Cereb Blood Flow Metab. 1997;17(4):430-436.

10. Steingrub JS, Mundt DJ. Blood glucose and neurologic outcome with global brain ischemia. Crit Care Med. 1996;24(5):802-806.

11. Borgquist O, Wise MP, Nielsen N, et al. Dysglycemia, Glycemic Variability, and Outcome After Cardiac Arrest and Temperature Management at 33°C and 36°C. Crit Care Med. 2017;45(8):1337-1343.

12. Minha S, Taraboulos T, Elbaz-Greener G, Kalmanovich E, Vered Z, Blatt A. Routine Laboratory Indices as Predictor of Neurological Recovery in Post-Resuscitation Syndrome Patients Treated with Therapeutic Hypothermia. Isr Med Assoc J. 2017;19(5):296-299.

13. Losert H, Sterz F, Roine RO, et al. Strict normoglycaemic blood glucose levels in the therapeutic management of patients within 12h after cardiac arrest might not be necessary. Resuscitation. 2008;76(2):214-220.

14. Niemann JT, Youngquist S, Rosborough JP. Does early postresuscitation stress hyperglycemia affect 72-hour neurologic outcome? Preliminary observations in the Swine model. Prehosp Emerg Care. 2011;15(3):405-409.

15. Bleske BE, Song J, Chow MS, Kluger J, White CM. Hematologic and chemical changes observed during and after cardiac arrest in a canine model--a pilot study. Pharmacotherapy. 2001;21(10):1187-1191.

16. Hopper K, Borchers A, Epstein SE. Acid base, electrolyte, glucose, and lactate values during cardiopulmonary resuscitation in dogs and cats. J Vet Emerg Crit Care (San Antonio). 2014;24(2):208-214.

17. D’Alecy LG, Lundy EF, Barton KJ, Zelenock GB. Dextrose containing intravenous fluid impairs outcome and increases death after eight minutes of cardiac arrest and resuscitation in dogs. Surgery. 1986;100(3):505-511.

18. Nakakimura K, Fleischer JE, Drummond JC, et al. Glucose administration before cardiac arrest worsens neurologic outcome in cats. Anesthesiology. 1990;72(6):1005-1011.

19. Molnar M, Bergquist M, Larsson A, Wiklund L, Lennmyr F. Hyperglycaemia increases S100β after short experimental cardiac arrest. Acta Anaesthesiol Scand. 2014;58(1):106-113.

20. Oksanen T, Skrifvars MB, Varpula T, et al. Strict versus moderate glucose control after resuscitation from ventricular fibrillation. Intensive Care Med. 2007;33(12):2093-2100.

21. Callaway CW, Soar J, Aibiki M, et al. Part 4: Advanced Life Support: 2015 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations. Circulation. 2015;132(16 Suppl 1):S84-145.

Supplemental Material:

Survival to discharge

• Skrifvars 2003, 2780: Patients with OHCA, VF only, witnessed only, and admission to ICU for PCA care, Finnland. N=98. Association between blood glucose at admission and survival to 6 months: (mmol/l, median) 9.8 in survivors, and 12.4 in non-survivors (P=0.008). The amount of pre-hospitally administered adrenaline (epinephrine) correlated (Spearman’s Rank) with blood glucose levels on admission (P=0.004) as well as with the mean 72 h blood glucose levels (P=0.0001). The mortality rate (at 6 months) was associated with the mean 72-hour glucose levels: 9% mortality (5.5-6.8 mmol/L), 23% (6.9-7.9 mmol/L), 50% (7.9-8.9 mmol/L), and 64% (9.1-27.9 mmol/L). Uncertain on whether the glucose itself is causative for the differences in survival or whether it is an epiphenomenon of severity of injury.
• Kim 2014, 2775: Observational study from registry data. Glucose measurements collected within 1 hour after ROSC in 883 OHCA patients treated with TTM. The mean BG concentration was significantly lower in the survival group (13.4 ± 6.5 vs 15.1 ± 7.6 mmol/L; P=0.001. Multivariate analysis: negative association between BG and survival (OR, 0.974; 95% CI, 0.952-0.996, P=0.021).
• Beiser 2009, 2777: BG after IHCA in adult patients from registry data (n=3218 events) during the first 24 hours after ROSC. Post-ROSC glucose levels were increased in both diabetics (12.6 mmol/L [9.2–17.1 mmol/L]) and non-diabetics (9.8 mmol/L [7.5–13.3 mmol/L]). Decreased survival odds were observed with maximum glucose values outside the range of 6.2–13.3 mmol/L.
• Russo 2018, 2774: Retrospective observational study in 122 patients OHCA and ROSC, in which BG was determined during the first 96 hours of hospitalization. Mean BG over these 96 hours categorized: <6 mmol/L, 6 to <8 mmol/L and ⩾8 mmol/L. Increased BG was associated with increased odds of death (OR 1.50; 95% CI 1.17–1.92; P=0.001) In multivariate model, this effect persisted for death (OR:1.35; 95%CI: 1.04–1.76; p=0.02).
• Sunde 2007, 2779: PCA protocol included a component that adjusted the BG concentration between 5-8 mmol/L. This is part of a bundle including other interventions, including standardized management of respiration and circulation, and TTM. Survival to discharge and FNO improved after implementation of the bundle.
• Zhou 2020, 2770: Mulitcentre retrospective observational study including 2028 patients with ROSC. BG values were categorized into 4 BG categories: ≤3.9 mmol/L, 3.9 to 7.8 mmol/L, 7.8 to 10 mmol/L, and >10 mmol/L. Mild hypoglycemia was defined as BG ≤3.9 mmol/L, while severe hypoglycemia was defined as ≤2.2 mmol/L. Calculated the time spent at a certain level of BG over the course of the first 24 hours of hospitalization. OR for mortality with mean BG 7.8 to 10 mmol/L: 1.62 [1.26, 2.08]; and for mean BG > 10 mmol/L1.80 [1.40, 2.30], both P<0.001 when compared to normal BG.

Favorable neurologic outcomes

• Kim 2014, 2775: Observational study from registry data. Glucose measurements collected within 1 hour after ROSC in 883 OHCA patients treated with TTM. Outcomes: survival to discharge and CPC at discharge. The mean BG concentration was significantly lower in the favorable neurologic outcome group (CPC 1-2) (12.8 ± 5.0 vs 14.6 ± 7.6 mmol/L; P=0.001). Multivariate analysis: negative association between BG and FNO (OR, 0.955; 95% CI, 0.918-0.994, P=0.024).
• Russo 2018, 2774: Retrospective observational study in 122 patients OHCA and ROSC, in which BG was determined during the first 96 hours of hospitalization. Mean BG over these 96 hours categorized: <6 mmol/L, 6 to <8 mmol/L and ⩾8 mmol/L. Increased BG was associated with poor FNO (OR: 1.42; 95%CI: 1.11–1.80; P=0.004). In multivariate model, this effect persisted for low FNO (OR: 1.28; 95%CI: 1.00–1.64; p=0.05)
• Sunde 2007, 2779: PCA protocol included a component that adjusted the BG concentration between 5-8 mmol/L. This is part of a bundle including other interventions, including standardized management of respiration and circulation, and TTM. Survival to discharge and FNO improved after implementation of the bundle.
• Steingrub 1996, 2782: Humans, IHCA and OHCA mix, patients with arrest time of more 5 mins (n equals 17), a mean post-CPR blood glucose concentration (samples obtained within 30 minutes after ROSC) of >11.1 mmol/L were associated with poor functional outcome (P<0.03).
• Muellner 1997, 2781: 145 patients, non-diabetic, with witnessed VF arrest. Negative association between high median blood glucose levels over the first 24 hours after ROSC and poor functional outcome (P = 0.015).
• Borquist 2017, 2091: 939 patients included in TTM trial and retrospectively (post-hoc) analyzed for BG impact on outcome. Median BG levels were higher in patients with poor neurologic outcomes (12 mmol/L [IQR, 8.9–15.9 mmol/L]) compared to those with FNO (10 mmol/L [IQR, 7.71–13.5 mmol/L], P<0.0001). Only considering the first two BG concentrations after admission, the study found a 22% increase in the probability of a poor neurologic outcome per 1 mmol/L increase in glucose concentration (P=0.010).
• Minha 2017, 2773: Small retrospective, single centre study including 47 initial survivors of cardiac arrest and treated with TTM for 24 hours: investigators found a temporary association between higher BG and poor neurological outcome early after ROSC, not at other times (likely underpowered study).
• Losert 2008, 2778: Post-hoc analysis including 234 OHCA patients undergoing TTM. Individuals were categorized according to quartiles of median glucose concentrations at 12 hours post ROSC: QI 5.6 (3.7—6.4 mmol/L), QII 7.2 (6.4—7.9 mmol/L), QIII 9.0 (7.9—10.7 mmol/L) and QIV 14.7 (10.7—25.8 mmol/L). In multivariate analysis adjusting findings for relevant confounders including down-time and duration of CPR, those in QI (OR, 4.55; 95% CI, 1.28—16.12), and QII (OR, 3.02; 95% CI, 3.29—49.9) were more likely to survive with good FNO compared to those in QIV. Although this was not the objective of this study, the authors conclude that a more liberal BG target might still be effective, and that tight glycemic control might not be necessary.
• Niemann 2011, 2776: Experimental swine study, n=22; OHCA VF model, 7 minutes no circulation time followed by CPR for up to 15 minutes; BG determined at 15, 30 and 60 minutes post-ROSC. No association between animals that had a BG >12.5 mmol/L versus <12.5 mmol/L and neurological function at 72 hours post ROSC (20 of 22 survived to 72 hours).

Glucose levels and their role with duration of CPR/CPA:

• Steingrub 1996, 2782: Humans, IHCA and OHCA mix, patients with arrest time of more than or equal 5 mins (n equals 17) had a higher mean post-CPR blood glucose concentration (mean equals 372 mg/dL [20.6 mmol/L]) compared with those patients with duration of arrest less than 5 mins (n equals 63; mean equals 245 mg/dL [13.6 mmol/L]).

• Bleske 2001, 1313: Experimental study; Study using 26 dogs to evaluated glucose and other metabolic parameters during and after CPR. 3 min of untreated VF, followed by 10 min of CPR; PCA samples obtained 10 minutes after ROSC. BG at baseline was 6.0 ± 1.9 mmol/L and increased to 12.2 ± 3.7 mmol/L 10 minutes after ROSC (P<0.05). None of the relevant outcomes reported