Mild therapeutic hypothermia has proved beneficial after out-of-hospital cardiac arrest in the adult population, when the initial rhythm is ventricular fibrillation (VF). In this study, data from 110 consecutive patients with out-of-hospital cardiac arrest due to VF (n = 86) or to non-VF rhythm (n = 24), admitted to an intensive cardiac care unit with restoration of spontaneous circulation and who remained unconscious on admission, were analyzed. Patients were cooled using an external cooling system. Of the patients with VF, 66% had favorable outcomes (Glasgow-Pittsburgh Cerebral Performance Category 1 or 2), and 30% died. Of the patients with non-VF, 8% had favorable outcomes (p <0.001 vs VF), and 63% died (p = 0.004 vs VF). In patients with VF, those with poor outcomes were older than those with favorable outcomes (odds ratio [OR] 1.61, 95% confidence interval [CI] 1.03 to 2.7, p = 0.001) and had previous ejection fractions <35% (OR 7.72, 95% CI 1.8 to 33, p = 0.002). Outcomes were also worse when patients presented to the emergency room with seizures (OR 20.96, 95% CI 2.48 to 177.42, p = 0.003) or hemodynamic instability (OR 14.4, 95% CI 3.47 to 60, p <0.0001). In the non-VF group, the 2 patients with good outcomes were younger than those with unfavorable outcomes (39 ± 16 vs 65 ± 12 years, respectively, p = 0.04), with good left ventricular function on presentation (100% vs 4.5%, p = 0.0001) and with short asystole and/or short time from collapse to restoration of spontaneous circulation. In conclusion, mild therapeutic hypothermia in the adult population is more effective in patients with VF compared to those with non-VF. Good prognostic factors for patients with non-VF could be young age, good left ventricular function, and short anoxic time.
Our group and others have shown that mild therapeutic hypothermia (MTH) is feasible and safe in different patient cohorts presenting with shockable rhythm: ventricular fibrillation (VF) or ventricular tachycardia after out-of-hospital cardiac arrest (OHCA), including patients in shock and those with acute ST-segment elevation myocardial infarction. Despite increasing experience and evidence regarding the benefit of MTH in treating patients with OHCA due to VF, only limited data exist regarding patients with other initial rhythms, for example, asystole and pulseless electrical activity (PEA): nonshockable rhythms. Recently, the European Resuscitation Council and the American Heart Association published new resuscitation guidelines, and although they acknowledge the low level of evidence for the use of hypothermia after nonshockable rhythms, both recommend using therapeutic hypothermia for comatose survivors of cardiac arrest with nonshockable rhythms as well, stating that “it might also benefit” such patients. In this study, we examined the outcomes of 110 consecutive patients treated with MTH after OHCA due to VF and to non-VF rhythms.
Methods
From February 2002 to May 2009, 110 resuscitated comatose patients admitted to our intensive cardiac care unit after cardiac arrest (presumed to be of cardiac origin) were treated with MTH. Of these patients, the first rhythm recorded was VF in 86, asystole in 18, and PEA in 6. Exclusion criteria for MTH were pregnancy, terminal illness, cardiogenic shock, and known primary coagulopathy. The families of all 110 patients were informed about MTH, and 3 physicians gave signed approval for the treatment. All patients received standard treatment according to intensive cardiac care unit protocols and were mechanically ventilated. Patients were sedated with induction and intravenous infusion of midazolam and were paralyzed with atracurium besylate to prevent shivering. Patients were undressed and cooled using a CritiCool external cooling system (MTRE Advanced Technologies Ltd., Rehovot, Israel), which enables simultaneous water flow through numerous channels in a body-shaped heat exchange garment (°CureWrap; MTRE Advanced Technologies Ltd.). On the basis of previous publications advocating cooling with cold intravenous fluids, we began adding cold intravenous fluids to the external cooling method protocol in August 2005 to achieve faster cooling. Starting in the emergency room, 65 patients (48 with VF and 17 with non-VF) received an infusion of 4°C intravenous normal saline (30 ml/kg, 100 ml/min). There was no prehospital cooling phase. All patients with acute ST-segment elevation myocardial infarction were immediately transferred to the catheterization laboratory. In patients who underwent primary percutaneous coronary intervention, MTH was started before and maintained throughout the procedure, either by cold intravenous normal saline alone or together with the CritiCool external cooling system. The goal MTH temperature was 32°C to 34°C. MTH duration was 24 hours at the goal temperature. The goal rewarming rate was 0.5°C/hour. We monitored blood pressure, oxygen saturation, electrocardiography, end-tidal carbon dioxide, respiratory rate, and core temperature. Antibiotics were added when aspiration was suspected clinically or radiographically.
Data on all consecutive patients who underwent therapeutic hypothermia because of VF and non-VF initial rhythms were collected prospectively using our electronic record charts. Patients were analyzed according to their initial rhythms. Neurologic outcomes on discharge were determined using Glasgow-Pittsburgh Cerebral Performance Category (CPC) on a 5-point scale : favorable outcomes included CPC 1 (good cerebral function) and CPC 2 (moderate neurologic disability), and unfavorable outcomes included CPC 3 (severe neurologic disability), CPC 4 (coma), or CPC 5 (death).
Restoration of spontaneous circulation (ROSC) was estimated according to the recorded time of patient evacuation, the number of advanced cardiac life support cycles, and the reports of the paramedics and the teams who admitted the patients to the intensive cardiac care unit. Severe co-morbidities were defined as ≥1 of the following: (1) ≥3 risk factors for ischemic heart disease (i.e., diabetes mellitus, hypertension, smoking, and hyperlipidemia), (2) ischemic heart disease or its equivalent (i.e., cerebrovascular event or peripheral vascular disease), (3) chronic renal failure, and (4) moderate to severe left ventricular dysfunction.
When comparing patients with favorable and unfavorable outcomes on the basis of their initial rhythms, categorical variables are presented as percentages and continuous variables as mean ± SD. We used nonparametric statistics (Wilcoxon-Mann-Whitney test) for comparisons. We used multiple logistic regression analysis to define the potential independent predictors for worse outcome in the VF group.
Results
From February 2002 to May 2009, data from a total of 110 consecutive patients treated with MTH were analyzed. Cardiac arrest was due to VF in 86 patients (78%) and to non-VF initial rhythms in 24 patients (22%). Baseline clinical and physiologic characteristics were comparable between the VF and non-VF groups ( Table 1 ). Treatment was discontinued prematurely in 5 patients because of hemodynamic instability in 3 and death during cooling in 2. Goal temperature was achieved in 4 of them. All these patients were included in the analysis.
| Variable | VF | p Value | |
|---|---|---|---|
| Yes (n = 86) | No (n = 24) | ||
| Men | 71 (83%) | 19 (79%) | 0.70 |
| Age (years) | 58 ± 15 | 63 ± 14 | 0.15 |
| Severe co-morbidities ⁎ | 49 (57%) | 14 (58%) | 0.90 |
| Previously known ejection fraction <35% | 26 (30%) | 5 (21%) | 0.37 |
| Coronary heart disease | 40 (46%) | 10 (42%) | 0.67 |
| Diabetes mellitus | 16 (19%) | 7 (29%) | 0.26 |
| Chronic renal failure | 7 (8%) | 1 (4%) | 0.51 |
| Hypertension | 42 (49%) | 14 (58%) | 0.41 |
| Smoking | 29 (34%) | 10 (42%) | 0.50 |
| Dyslipidemia | 37 (43%) | 10 (42%) | 0.91 |
| Cerebrovascular accident | 16 (19%) | 5 (21%) | 0.81 |
| Chronic obstructive pulmonary disease/asthma | 4 (5%) | 2 (8%) | 0.48 |
| Peripheral vascular disease | 2 (2%) | 3 (13%) | 0.03 |
| Previous malignancy | 1 (1%) | 1 (4%) | 0.33 |
| Basic life support by bystander | 51 (59%) | 15 (63%) | 0.78 |
| Ischemic causes (STEMI and ACS) | 48 (56%) | 11 (46%) | 0.39 |
| STEMI | 41 (48%) | 8 (33%) | 0.21 |
| Cardiomyopathy/heart failure–induced cardiac arrhythmia | 21 (25%) | 3 (13%) | 0.20 |
| Other causes | 16 (19%) ‡ | 10 (42%) † | 0.02 |
| Primary percutaneous cardiac intervention | 34 (40%) § | 8 (33%) | 0.58 |
| Ejection fraction on admission ∥ | |||
| Good (>45%) | 20 (23%) | 3 (13%) | 0.25 |
| Moderate (35%–45%) | 12 (14%) | 3 (13%) | 0.85 |
| Poor (<35%) | 47 (55%) | 16 (67%) | 0.29 |
⁎ Three or more risk factors for ischemic heart disease, ischemic heart disease or cerebrovascular accident or peripheral vascular disease, chronic renal failure, or left ventricular ejection fraction ≤35%. Hypertension and hyperlipidemia were reported by family members or per previous medical reports of the patients. Coronary artery disease refers to previous coronary bypass surgery or myocardial infarction or coronary angiography showing significant coronary disease.
† Due to massive pulmonary embolism, hyperkalemia, complete heart block, myocarditis, chronic obstructive lung disease exacerbation, electrocution, unknown cause, or deteriorated primary ventricular fibrillation.
‡ Due to arrhythmogenic right ventricular dysplasia, infective endocarditis, primary ventricular fibrillation, atrial septal defect, mitral valve prolapse, hypokalemia, long-QT syndrome, catecholaminergic polymorphous ventricular tachycardia, Brugada syndrome, or sepsis.
§ Two patients were not catheterized emergently, because of ST-segment resolution >70% on electrocardiography on admission, reflecting spontaneous reperfusion. They underwent percutaneous cardiac intervention later in this hospitalization.
∥ Ejection fraction on admission was not available in 7 patients with VF and 2 patients with non-VF.
Patients in the non-VF group had a longer time interval from collapse to ROSC compared to the VF group (p = 0.005) and were more likely to have seizures on arrival to the hospital before MTH treatment was applied (p = 0.02). Otherwise, there were no differences between the groups, including use of catecholamines, first measured lactate, pH, and cooling parameters ( Table 2 ). Although effective MTH was equally provided for patients who presented with VF and non-VF rhythms, patients with non-VF rhythms had significantly worse prognoses with significantly higher mortality and worse neurologic outcomes, reflected by a significantly lower incidence of patients in CPC 1 or 2 ( Table 3 ). Of the 86 patients with VF, 57 (66%) had favorable outcomes (CPC 1 or 2), and 26 (30%) died. Seventy-seven percent of the favorable outcome group were in CPC 1.
| Variable | VF | p Value | |
|---|---|---|---|
| Yes (n = 86) | No (n = 24) | ||
| Collapse to advanced cardiac life support ≥10 minutes | 29 (35%) | 13 (54%) | 0.08 |
| ROSC <25 minutes from collapse | 46 (54%) | 5 (21%) | 0.005 |
| Use of catecholamines or systolic blood pressure <90 mm Hg on arrival | 32 (37%) | 9 (39%) | 0.87 |
| Seizures on arrival (before MTH) | 11 (13%) | 8 (33%) | 0.02 |
| First measured lactate (mg/dl) | 52 ± 37 ⁎ | 67 ± 42 ⁎ | 0.10 |
| First measured arterial pH | 7.33 ± 0.14 † | 7.28 ± 0.15 † | 0.19 |
| Temperature on admission (°C) | 36.6 ± 0.8 | 36.7 ± 1.2 | 0.92 |
| Time from emergency room to hypothermia (minutes) ‡ | 147 ± 113 | 154 ± 103 | 0.68 |
| Time from hypothermia to goal temperature (minutes) | 284 ± 181 | 234 ± 158 § | 0.15 |
| Total duration of therapy (from initiation to end) (hours) | 23 ± 7 | 22 ± 6 | 0.68 |
| Duration of target temperature 32°C–34°C (hours) | 18.5 ± 7.6 | 18.1 ± 8.2 | 0.74 |
⁎ First measured lactate: n = 70 in VF group and n = 20 in non-VF group.
† First measured arterial pH: n = 74 in VF group and n = 20 in non-VF group.
‡ Emergency room or intensive cardiac care unit (wherever the patient was seen first).
| Variable | VF | p Value | Asystole (n = 18) | p Value vs VF | PEA (n = 6) | p Value vs VF | |
|---|---|---|---|---|---|---|---|
| Yes (n = 86) | No (n = 24) | ||||||
| Death | 26 (30%) | 15 (63%) | 0.004 | 11 (61%) | 0.01 | 4 (67%) | 0.07 |
| CPC 1 or 2 | 57 (66%) | 2 (8%) | <0.0001 | 2 (11%) | <0.0001 | 0 | 0.001 |
| Pneumonia | 53 (62%) | 13 (54%) | 0.51 | 11 (61%) | 0.97 | 2 (33%) | 0.17 |
| Sepsis | 13 (15%) | 8 (33%) | 0.05 | 6 (33%) | 0.07 | 2 (33%) | 0.25 |
| Any bleeding | 11 (13%) | 6 (25%) | 0.14 | 5 (28%) | 0.11 | 1 (17%) | 0.79 |
| Seizures (clinical/electroencephalography) | 21 (24%) | 17 (71%) | <0.0001 | 12 (67%) | 0.0005 | 5 (83%) | 0.002 |
| Cardiogenic shock | 13 (15%) | 0 | 0.04 | 0 | 0.08 | 0 | 0.31 |
| Arrhythmia | 16 (19%) | 2 (8%) | 0.23 | 1 (5.5%) | 0.18 | 1 (17%) | 0.91 |
| Pulmonary congestion | 14 (16%) | 8 (33%) | 0.07 | 5 (28%) | 0.25 | 3 (50%) | 0.04 |
| Need for catecholamines during MTH | 15 (17%) | 5 (21%) | 0.70 | 3 (17%) | 0.94 | 2 (33%) | 0.33 |
Of the 24 patients with non-VF, only 2 (8%) had favorable outcomes (p <0.0001 vs the VF group), and 15 patients (63%) died (p = 0.004 compared to the VF group). The 2 patients with favorable outcomes were younger (39 ± 16 vs 65 ± 12 years, p = 0.04),had good left ventricular function on admission (100% vs 4.5%, p = 0.0001), and had no evidence of obstructive coronary artery and/or structural heart disease ( Table 4 ). These 2 patients presented with asystole; 1 had a very short time from collapse to ROSC, and the other had only short asystole as the initial rhythm, with rapid convergence from asystole to VF after the initiation of cardiopulmonary resuscitation (CPR). He then underwent long CPR for recurrent episodes of VF. Among the non-VF group presenting with asystole (n = 18), 11 patients (61%) died. None of the 6 patients with PEA had good prognoses: 4 of 6 (67%) died, and none had CPC 1 or 2 ( Table 3 ). In the group with poor outcomes, 10 patients had longer CPR of several minutes before converting to VF. Most of the unfavorable non-VF group had prolonged time from collapse to ROSC. Although not statistically significant, the 2 patients in the non-VF group with favorable outcomes had a longer total duration of MTH and duration of MTH at target temperature compared to the patients with unfavorable outcomes ( Table 5 ).
| Variable | VF (n = 86) | Non-VF (n = 24) | ||||
|---|---|---|---|---|---|---|
| Favorable Outcome (n = 57) | Unfavorable Outcome (n = 29) | p Value | Favorable Outcome (n = 2) | Unfavorable Outcome (n = 22) | p Value | |
| Men | 48 (84%) | 23 (79%) | 0.57 | 2 (100%) | 17 (77%) | 0.46 |
| Age (years) | 54 ± 15 | 66 ± 13 | 0.001 | 39 ± 16 | 65 ± 12 | 0.04 |
| Severe co-morbidities ⁎ | 25 (44%) | 24 (83%) | 0.0006 | 0 | 14 (64%) | 0.09 |
| Previously known ejection fraction <35% | 11 (19%) | 15 (52%) | 0.002 | 0 | 5 (23%) | 0.46 |
| Ischemic heart disease | 20 (35%) | 20 (69%) | 0.003 | 0 | 10 (46%) | 0.22 |
| Diabetes mellitus | 9 (16%) | 7 (24%) | 0.35 | 0 | 7 (32%) | 0.35 |
| Chronic renal failure | 3 (5%) | 4 (14%) | 0.17 | 0 | 1 (5%) | 0.76 |
| Hypertension | 24 (42%) | 18 (62%) | 0.08 | 0 | 14 (64%) | 0.09 |
| Smoking | 20 (36%) | 9 (31%) | 0.67 | 0 | 10 (45%) | 0.22 |
| Dyslipidemia | 21 (37%) | 16 (55%) | 0.11 | 0 | 10 (45%) | 0.22 |
| Cerebrovascular accident | 9 (16%) | 7 (24%) | 0.35 | 0 | 5 (23%) | 0.46 |
| Chronic obstructive pulmonary disease/asthma | 2 (4%) | 2 (7%) | 0.48 | 0 | 2 (9%) | 0.66 |
| Peripheral vascular disease | 1 (2%) | 1 (3%) | 0.62 | 0 | 3 (14%) | 0.58 |
| History of malignancy | 1 (2%) | 0 | 0.48 | 0 | 1 (4.5%) | 0.76 |
| Basic life support by bystander | 35 (61%) | 16 (55%) | 0.58 | 1 | 14 (64%) | 0.71 |
| Ischemic cause (acute myocardial infarction and ACS) | 33 (58%) | 15 (52%) | 0.59 | 0 | 11 (50%) | 0.18 |
| STEMI | 30 (53%) | 11 (38%) | 0.2 | 0 | 8 (36%) | 0.31 |
| Cardiomyopathy/CHF-induced cardiac arrhythmia | 13 (23%) | 8 (28%) | 0.66 | 0 | 3 (14%) | 0.58 |
| Other causes † | 11 (20%) | 5 (17%) | 0.79 | 2 | 8 (36%) | 0.09 |
| Primary percutaneous cardiac intervention | 25 (44%) | 9 (31%) | 0.25 | 0 | 8 (36%) | 0.31 |
| Ejection fraction on admission † | ||||||
| Good (>45%) | 15 (26%) | 5 (17%) | 0.35 | 2 | 1 (4.5%) | 0.0001 |
| Moderate (35%–45%) | 10 (18%) | 2 (7%) | 0.18 | 0 | 3 (14%) | 0.58 |
| Poor (<35%) | 30 (53%) | 17 (59%) | 0.6 | 0 | 16 (73%) | 0.04 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
