Article Text
PDF
Abstract
Background and objective Mild therapeutic hypothermia (MTH) is used to limit neurological injury and improve survival after cardiac arrest (CA) and cardiopulmonary resuscitation, but the optimal mode of cooling is controversial. We therefore compared the effectiveness of MTH using invasive intravascular or non-invasive surface cooling with temperature feedback control.
Methods This retrospective study in post-CA patients studied the effects of intravascular cooling (CoolGard, Zoll, n=97), applied on the intensive care unit (ICU) in one university hospital compared with those of surface cooling (Medi-Therm, Gaymar, n=76) applied in another university hospital.
Results Time to reach target temperature and cooling speeds did not differ between groups. During the maintenance phase, mean core temperature was 33.1°C (range 32.7–33.7°C) versus 32.5°C (range 31.7–33.4°C) at targets of 33.0 and 32.5°C in intravascularly versus surface cooled patients, respectively. The variation coefficient for temperature during maintenance was higher in the surface than the intravascular cooling group (mean 0.85% vs 0.35%, p<0.0001). ICU survival was 60% and 50% in the intravascularly and surface cooled groups, respectively (NS). Lower age (OR 0.95; 95% CI 0.93 to 0.98; p<0.0001), ventricular fibrillation/ventricular tachycardia as presenting rhythm (OR 7.6; 95% CI 1.8 to 8.9; p<0.0001) and lower mean temperature during the maintenance phase (OR 0.52; 95% CI 0.25 to 1.08; p=0.081) might be independent determinants of ICU survival, while cooling technique and temperature variability did not contribute.
Conclusions In post-CA patients, intravascular cooling systems result in equal cooling speed, but less variation in temperature during the maintenance phase, as surface cooling. This may not affect the outcome.
- resuscitation, research
- resuscitation, effectiveness
Statistics from Altmetric.com
Key messages
What is already known on this subject?
Previous studies comparing intravenous with external cooling methods to achieve therapeutic hypothermia in postcardiac arrest patients are limited in size and controversial in effect measures like time to achieve target temperature and temperature stability during maintenance.
What this study adds?
In this retrospective study at two university hospitals using two methods of cooling in the Netherlands, intravascular cooling resulted in equal cooling speed, but less variable temperature during the maintenance phase, in comparison to surface cooling. Cooling method was not associated with survival.
Introduction
Survival from cardiac arrest (CA) with complete neurological recovery is rare.1 Mild therapeutic hypothermia (MTH) limits neurological injury and improves outcome in CA patients after successful cardiopulmonary resuscitation.2 ,3 The American Heart Association (AHA) and the International Liaison Committee on Resuscitation (ILCoR) have recommended that all unconscious adult patients with spontaneous circulation after out-of-hospital CA as a result of ventricular fibrillation (VF) should be cooled to 32–34°C for 12–24 h.4 Many different methods and technical devices for inducing and maintaining hypothermia and for controlled rewarming have been described and are currently applied.5–8 However, the optimal target temperature and thus cooling mode are still controversial.9 Conventional cooling methods include the use of cooling blankets, cold fluid, gel-pads and ice packs,2 ,3 ,10 and each method has merits and detriments. International guidelines recommend that temperature maintenance is best achieved with devices that incorporate continuous temperature feedback control. However, they identify a critical knowledge gap in the optimal cooling techniques, mainly with respect to internal11 versus external12 cooling methods. Until now, only a few and relatively small studies reported on differences between invasive and non-invasive cooling modalities using an automatic temperature control module guided by the core temperature of the patient. In a small study, Hoedemaekers et al7 showed that cooling with water-circulating blankets, gel-pads and intravascular cooling was equally effective in inducing hypothermia. The intravascular cooling system best maintained target temperature. In another small study, no difference in time to target temperature between intravascular cooling and water-circulating cooling blankets was shown, but better maintenance of target temperatures using the former.13 Pittl et al14 recently showed in an randomised controlled trial (RCT) that invasive intravascular cooling was more effective in maintaining target temperature in comparison with non-invasive surface cooling. Using the same devices, in a larger non-randomised trial by Tømte et al15 focusing on outcome did not result in time differences for achieving target temperature. Effects on the temperature stability during the maintenance phase were not studied.
Because the studies on intravenous versus external cooling cited above are limited in size and controversial in effect measures like time to achieve target temperature and temperature stability during maintenance, we compared in a large population the effectiveness of inducing and maintaining MTH by non-invasive water-circulating body wraps (Gaymar, New York, USA) versus invasive intravascular (CoolGard) cooling devices, both systems using automatic temperature feedback control.
Patients and methods
Patients
According to Dutch legislation, no approval of the research ethics committee was required for this retrospective study as long as data were anonymised. This study was done at the intensive care unit (ICU) of two university hospitals in the Netherlands, between 2009 and 2010. Consecutive post-CA patients were treated using invasive intravascular cooling (CoolGard, Zoll, Massachusetts, USA, n=97) at Erasmus University Medical Center Rotterdam or non-invasive surface cooling (Medi-Therm, Gaymar, n=76) at VU University Medical Center Amsterdam. Both cooling systems are commercially available and widely used in clinical practice. Both systems use an oesophageal temperature probe to measure patients’ core temperature.
Therapeutic protocol
In both hospitals, patients with CA were treated in accordance with a standard operating procedure that includes MTH for all comatose patients, irrespective of initial heart rhythm or presumed cause of CA. They were all intubated and mechanically ventilated. Contraindications for MTH were uncontrolled active bleeding or severe bradycardia. When MTH was indicated, cooling was initiated in the emergency department on hospital arrival with infusion of refrigerated NaCl 0.9% (4°C) with a maximum of 1 L for the surface cooled patients in one institution. In this group, body wraps were placed around the patient's body and legs upon arrival in the ICU, while an external temperature control unit adjusted the temperature of the water circulating through the wraps (Medi-Therm, Gaymar). Target temperature was set at 32.5°C. The surface cooled patients were deeply sedated with propofol in combination with fentanyl and shivering was treated by increasing fentanyl. Infusion of a maximum of 300 mL refrigerated NaCl 0.9% in the intravascularly cooled patients was started during percutaneous intervention or after admission to the ICU. For the invasive intravascular group, a catheter (CoolGard, Quattro, Zoll) was inserted in the femoral vein. Temperature-adjusted sterile saline flows within heat-exchange balloons, located near the distal end of the catheter, and exchanges heat with the blood as it passes by. The target temperature in this group was set at 33.0°C. The intravascularly cooled patients were deeply sedated with propofol and midazolam in combination with fentanyl and shivering was treated by increasing fentanyl. The purpose was to achieve target temperature as soon as possible after admission to the ICU and maintained for 24 h after which rewarming starts with a maximum of 0.65°C per hour. Patients were mechanically ventilated, monitored and treated by vasoactive agents according to institutional and international guidelines, by board-certified intensivists in the ICU. To maintain arterial BP and rhythm, dobutamine was the preferred agent in one unit and norepinephrine in the other. Somatosensory evoked potential (SSEP) was tested 48 h after MTH treatment when normothermia was achieved. SSEP results were tested positive when bilateral N20 response was present.
Data collection
Prospectively collected data were retrospectively retrieved from charts and electronic patient data management systems, according to a predefined check list. Patient characteristics (gender, age, body weight), arrest conditions (out of hospital CA, cardiogenic shock upon admission, presenting rhythm VF/ventricular tachycardia (VT), ST segment elevation myocardial infarction on ECG, coronary angiogram, percutaneous coronary intervention, outcome (GCS), ICU survival, length of stay, whether SSEP was performed and tested positive and whether palliative care was initiated, were collected. Laboratory parameters upon admission (pH, lactate, bicarbonate, creatine kinase-myoglobin band (CK-MB), troponin-T), temperature parameters (first temperature at the ICU, administered cooled saline infusion, time from arrival at the ICU to the time when target temperature was reached ±<0.5°C, cooling speed, and the mean, maximum and minimum temperature and coefficient of variation in temperature during the maintenance phase, from start of cooling), and presence of sinus rhythm and use of isoprenaline, norepinephrine or dobutamine were recorded. Haemodynamics (heart rate and mean arterial pressure) and complications (device-related skin complications, shivering, pneumonia, sepsis, bleeding at site of catheter insertion and thrombosis) were retrieved. Patients who left the ICU alive were labelled as survivors and patients who did not survive their ICU stay but died after MTH was initiated were labelled as non-survivors throughout the manuscript.
Statistical analysis
Calculations and statistical analysis were performed using SPSS V.20 (IBM, New York, USA). Graphs were made using GraphPad Prism V.5.0 (GraphPad Software, San Diego, USA). Continuous variables are reported as mean±SD or when not normally distributed (Kolmogorov–Smirnov test, p>0.05) reported as median (IQR). The coefficient of variation is SD/mean. Categorical variables are reported as numbers and percentages. To test for significant differences between the two groups, Student's t test or Mann–Whitney U test and Fisher exact test for categorical variables were used. The association between clinically relevant variables and ICU survival/non-survival was tested through multiple logistic regression analysis using the backward method. ORs for survival (95% CIs) were calculated for age, VF/non-VF and mean temperature. A two-sided p≤0.05 was considered as indicating statistical significance. Exact p values are given unless <0.0001.
Results
Patient characteristics and outcome
Intravascular cooling in 97 patients and surface cooling in 76 patients were applied (table 1). No patient was excluded from this study.
Patient characteristics and outcome
In both groups, the majority of the patients were men, with comparable age and length of stay at the ICU. The surface cooled patients had a body weight of 75 kg, which was lower than of patients in the intravascularly cooled group with a body weight of 80 kg. The percentage of admitted patients suffering out-of-hospital CA, experiencing cardiogenic shock, presenting with VF as first rhythm, showing ST segment elevation myocardial infarction on ECG and who underwent coronary angiography upon hospital admission was lower in the surface cooled patients compared with the intravascularly cooled patients. The pH, lactate, bicarbonate, CK-MB and troponin-T upon ICU admission did not show differences between groups (table 2).
Laboratory values at ICU admission
The percentage of patients in which palliative care was initiated, SSEP was performed and SSEP results were positive did not statistically significantly differ between groups. However, patients in the intravascular and surface cooling groups had a median GCS of respectively 15 and 10 (p=0.008) at discharge from the ICU. No complications related to any of the cooling modalities were observed.
Induction and maintenance phase of MTH
The first temperature measured at the ICU was not different between the intravascular and surface cooling groups (figure 1 and table 3). Time from arrival at the ICU to reach target temperature and cooling speed did not differ between the groups. Due to different target temperatures, mean core temperatures differed between groups during the 24 h maintenance phase. Maintenance temperatures showed less variation and the coefficient of variation for temperature was lower in the intravascularly than in the surface cooled patients. There was a difference in the type of inotropes used by the two university hospitals. More norepinephrine and less dobutamine were used in the surface than in the intravascular cooling group. Heart rate decreased to the same extent in the groups during the induction phase of MTH (figure 2). During the 24-h maintenance phase, no differences in heart rate, and mean systolic and diastolic arterial pressure between the two cooling modalities were observed. In both groups, 88% of the patients showed sinus rhythm during the maintenance phase (table 3).
Induction and maintenance phase of mild therapeutic hypothermia
Mean body temperature during the induction and maintenance phase of mild therapeutic hypothermia using intravascular (n=97) or surface cooling (n=76). Time point zero indicates start of cooling. Data are mean±SD. The slight rise in temperature at the end of the 24 h cooling phase in both groups was due to termination of the cooling.
Heart rate and mean arterial blood pressure (MAP) in postcardiac arrest patients during mild therapeutic hypothermia using intravascular or surface cooling. Time point zero indicates start of cooling. Data are mean±SD.
Patient characteristics and laboratory values of survivors and non-survivors
Patient data were grouped for outcome (table 4).
Patient characteristics and laboratory values of ICU survivors and non-survivors
Patients who survived their ICU stay were younger compared with the non-survivors. The percentage of patients presenting with VF as first rhythm and who underwent coronary angiography upon hospital admission was higher in the ICU survivors compared with the non-survivors. Upon ICU admission, lactate levels were lower and bicarbonate levels were slightly higher in survivors compared with non-survivors. The percentage of patients in whom palliative care was initiated, SSEP was performed and SSEP results were positive was higher in the survivors compared with the non-survivors. ICU survivors had a higher median GCS at discharge from the ICU and longer length of stay at the ICU compared with non-survivors. The factors of age and presenting rhythm VF/VT together with the cooling technique, Gaymar or CoolGard, mean temperature and coefficient of variation in temperature during the maintenance phase were considered in a multiple logistic regression model with survival as the outcome. The model showed no statistical significance for the used cooling technique and the coefficient of variation in temperature during the maintenance phase. The main determinants of survival were lower age (OR 0.95, 95% CI 0.93 to 0.98; p<0.0001), having VF/VT as presenting rhythm (OR 7.6, 95% CI 1.8 to 8.9; p<0.0001) and, perhaps, lower mean temperature during the maintenance phase (OR 0.52, 95% CI 0.25 to 1.08; p=0.081).
Discussion
In this study, we retrospectively evaluated invasive intravascular cooling and non-invasive surface cooling systems for achieving MTH in post-CA patients. Both methods resulted in equal time to reach target temperature, but the intravascular method offered a more controlled and precise cooling in terms of less temperature variation during the maintenance phase. However, the clinical significance of this finding remains unclear, although intravascular cooling was associated with a higher GCS at ICU discharge. Moreover, age and VF/VT as presenting rhythm but not the cooling technique used or temperature variability were independently associated with survival.
The optimal cooling technique to deliver therapeutic hypothermia after CA still remains debated. The AHA and ILCoR recommend that temperature maintenance is best achieved with devices that incorporate continuous temperature feedback control. However, they identify a critical knowledge gap in the optimal cooling techniques, mainly with respect to internal versus external cooling methods. External cooling methods like air- or water-circulating cooling blankets, fluid-packs or ice packs and precooled metal plates can be applied around the head, torso and limbs. For hospital staff, external cooling methods are logistically difficult to administer, require significant nursing attention and reduce overall access to critically ill patients. When applying external cooling methods, freezing-induced skin damage and shivering need continuous attention.16 Invasive cooling can be achieved by peritoneal17 or gastric lavage with cold sterile fluids or intravenous cooling via rapid infusion of iced isotonic fluid. The latter is safe, easy and inexpensive to induce cooling. However, this method cannot be used to maintain MTH, and so it must be combined with a follow-up method for maintenance of MTH. Intravenous cooling can also be accomplished by means of an intravascular catheter inserted in the femoral vein. The catheters are made of metal or contain a balloon which is filled with fluid. A bedside heat exchanger maintains the fluid at a controlled temperature. This limits the use to the hospital setting as well as the need for a trained physician to insert the catheter into the large vein. Both invasive intravascular (CoolGard, Zoll) and non-invasive water-circulating surface (Medi-Therm, Gaymar) cooling devices used in this study were equipped with a continuous temperature feedback control.
Our results obtained in a fairly large but retrospective study suggest that the mean temperature during the maintenance (p=0.081) as well as less advanced age and VF/VT as presenting rhythm may be independently associated with survival and that cooling method and temperature variability may not contribute. In a recent RCT by Nielsen et al,9 therapeutic hypothermia at a target temperature of 33°C seemed to have no benefit compared with a target temperature of 36°C. Our observations are also in agreement with the relatively small studies by Gillies et al18 and Finley Caulfield et al13 who found that intravascular cooling (n=26) is superior to surface cooling by circulating cooling blankets (n=15) in maintaining target temperature within small limits, but it is not faster in reaching target temperature. Hoedemaekers et al7 compared five cooling modalities and demonstrated, in only 10 patients per group, that water-circulating blankets, gel-pads and intravascular cooling were more effective in reaching target temperature compared with conventional cooling and air-circulating blankets. However, for maintaining target temperature, intravascular cooling was superior to all other cooling methods, showing less time spent out of target range. Our results obtained in a larger number of patients only partly agree with theirs.7 Pittl et al14 recently showed in a randomised comparison of invasive intravascular (CoolGard, n=40) and non-invasive gel-pad surface cooling (Arctic Sun, n=40) that both devices were similarly safe and effective, so that the time to reach target temperature was equal in both groups, but maintaining target temperature appeared easier using the intravascular cooling method. Outcome did not differ. Although their methods differ from ours, the results are similar. A non-randomised trial by Tømte et al,15 comparing invasive intravascular (CoolGard) and surface cooling (Arctic Sun), showed no difference with respect to time for achieving target temperature and survival with good neurological outcome, which may agree with our findings.
Limitations of the study
Due to the retrospective design of this study, some group and institutional differences were observed. The intravascularly cooled patients had a higher body weight compared with the surface cooled patients, which did not result in longer time to reach target temperature. Also, outcome did not differ between groups even though more patients in the surface cooled group had asystole as the first observed electrocardiographic rhythm, which may result in a worse outcome.19 Indeed, when multiple logistic regression was used to determine the association between clinically relevant variables and ICU survival, lower age and VF/VT as presenting rhythm were the main, independent determinants of ICU survival, but cooling method and temperature variables did not significantly contribute. No difference in first temperature at the ICU, cooling speed or time to reach target temperature from ICU admission (where both cooling devices were started) was observed despite intravenous infusion of cooled saline already at the emergency department in the surface cooled group. Furthermore, there was a difference in the type of inotropes used by the two university hospitals. More norepinephrine and less dobutamine were used in the surface cooled compared with the intravascularly cooled group. However, different use of inotropes was not reflected in differences in heart rate or BP between both groups, which might be associated with a diminished effect of inotropes during hypothermia. Detailed knowledge of temperature-dependent changes in pharmacodynamics and pharmacokinetics of cardioactive drugs is not available. The study was apparently sufficiently powered to demonstrate a difference in temperature variability between techniques.
Conclusions
Our non-randomised study comparing different cooling techniques after CA and cardiopulmonary resuscitation suggests that use of intravascular or surface cooling systems for achieving MTH results in similar cooling speed irrespective of body weight. The intravascular method offers a more controlled cooling during the maintenance phase. However, the variation in temperature during the maintenance phase and thus the cooling system used may not affect outcome.
Acknowledgments
We thank the staff of the participating intensive care units.
References
Footnotes
Contributors MCdW: design of the work, acquisition, analysis and interpretation of data for the work, and submitted the study, revised it critically for important intellectual content and approved the final version to be published. RPB and LSDJ: acquisition and analysis of data for the work, revised it critically for important intellectual content, and approved the final version to be published. AS and ARJG: revised it critically for important intellectual content and approved the final version to be published. ABJG: design of the work, interpretation of data for the work, revised it critically for important intellectual content and approved the final version to be published.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.