If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Critical Care & Neuro-Critical Care Medicine, Program in Critical Care, Department of Medicine, London Health Sciences Centre, University of Western Ontario, LHSC-UC, 339 Windermere Road, London, Ontario N6A 5A5, Canada
Therapeutic hypothermia has been demonstrated to improve neurological outcome in comatose survivors of cardiac arrest. Current temperature control modalities however, have several limitations. Exploring innovative methods of temperature management has become a necessity.
We describe the first use of a novel esophageal cooling device as a sole modality for hypothermia induction, maintenance and rewarming in a series of four postcardiac arrest patients. The device was inserted in a manner similar to standard orogastric tubes and connected to an external heat exchange unit.
A mean cooling rate of 0.42 °C/hr (SD ± 0.26) was observed. An average of 4 hr 24 min (SD ± 2 hr 6 min) was required to reach target temperature, and this was maintained 90.25% (SD ± 16.20%) of the hypothermia protocol duration. No adverse events related to device use were encountered. Questionnaires administered to ICU nursing staff regarding ease-of-use of the device and its performance were rated as favorable.
When used as a sole modality, objective performance parameters of the esophageal-cooling device were found to be comparable to standard temperature control methods. More research is required to further quantify efficacy, safety, assess utility in other patient populations, and examine patient outcomes with device use in comparison to standard temperature control modalities.
Despite successful resuscitation, many patients develop neurologic deficits from post arrest cerebral reperfusion injury. Studies suggest that induced hypothermia, targeting a temperature between 32 and 34 °C for 12–24 hr, may improve survival and neurological prognosis.
Intravenous bolus administration of 4 °C normal saline, is usually effective for hypothermia induction but is not well suited for maintenance. Repetitive or continuous administration of cold saline during the maintenance phase may induce volume overload and pulmonary edema.
Recently, an internal cooling modality utilizing an esophageally-placed heat transfer device, has become available. Previous reports of use, in conjunction with other temperature reduction methods, suggest a synergistic effect in controlling temperature and reducing fevers.
This is the first report however, to describe esophageal cooling as a sole temperature control method in post cardiac arrest patients.
The Esophageal Cooling Device® (Advanced Cooling Therapy, Chicago, IL) is a triple-port, disposable, silicone tube, placed in the esophagus (Fig. 1). Two water ports connect to an external heat exchange unit and circulate cold or warm water in closed circuit within the outer lumen of the tube. The gastric port forms the central lumen, which opens via small apertures at the tube tip in the stomach. Currently, two esophageal cooling device (ECD) models are available: one compatible with the Blanketrol Hyper-Hypothermia System® (Cincinnati Sub-Zero, Cincinnati, OH), and the other compatible with the Medi-Therm Hyper/Hypothermia System® (Stryker, Kalamazoo, MI). The ECD is approved for clinical use as a temperature control modality in the US, Canada, UK, Australia, and the European Union.
Four patients admitted to the intensive care unit (ICU) at London Health Sciences Center, University Hospital, London, Canada between November 5, 2014 and February 5, 2015 requiring targeted temperature management after resuscitation from an out-of-hospital cardiac arrest, were screened for ECD use. Informed written consent was obtained from the patients' substitute decision makers prior to ECD insertion. In addition, consent to use patient information for publication was obtained. The Western University Research Ethics Board does not mandate a separate review process prior to the publication of case series if patient confidentiality is ensured.
All patients received a hypnotic (midazolam 2–3 mg IV) and a neuromuscular blocking agent (cisatracurium 10–20 mg IV) prior to ECD insertion. The technique of placement was similar to that of standard orogastric tubes. The appropriate depth of insertion was determined externally on the patient by extending the ECD from the angle of the mouth to the tragus, then from the tragus to the tip of the xiphoid process. The ECD was then connected to the external heat exchange unit and water flow was initiated.
This step (initiation of flow) stiffened the ECD facilitating its subsequent advancement. The tip and distal portion of the device was then lubricated with a water-soluble lubricant and advanced into the oropharynx/esophagus to the predetermined depth. A slight jaw thrust provided by an assistant aided the passage of the ECD beyond the hypopharynx. Adequate depth of placement was then confirmed by chest X-ray with visualization of the radio-opaque tip in the stomach (Fig. 2). The central (gastric) port was then connected to a low level of intermittent suction in the initial post-insertion time period. In all four patients however, this central port was subsequently used for the administration of enteral medications.
All patients were continuously monitored with temperature-sensing urinary catheters. A second temperature measurement was obtained at a different site (rectal or axillary), once every 12 hr, to verify measurement accuracy. The heat exchange unit (Blanketrol Hyper-Hypothermia System) was set on “Auto Control” mode. In this mode, the temperature of water circulating through the ECD is automatically adjusted to cool or warm the patient to the target temperature set by the operator. The patient target was set to 35 °C in all four patients post ECD insertion. A brief orientation to the temperature control system was provided to nurses at the bedside in addition to a bedside instruction sheet (supplementary appendix). Target and patient temperatures were recorded hourly in patient A. Patients B, C, and D however, had their temperatures recorded in an automated fashion using a laptop connected to the heat exchange unit. This laptop captured patient temperatures, target temperatures, and cooling water temperatures at 30-s to 1-min intervals using Blanketrol III Data Export Software (Cincinnati Sub-Zero, Cincinnati, OH).
As per our institutional protocol, comatose survivors of cardiac arrest are routinely cooled to a temperature of 35 °C (±1 °C) for a period of 24 hr. Our standard cooling modalities include the use of intravenous cold saline and the application of ice packs and cooling blankets (Maxi-Therm® Lite Blankets, Cincinnati Sub-Zero, Cincinnati, OH). In all four patients in this report however, the ECD was used as a sole temperature control modality for induction, maintenance and rewarming. All patients received intravenous infusions of both an opioid (fentanyl or hydromorphone) and a hypnotic agent (midazolam or propofol) during the 24 hr of hypothermia. Use of neuromuscular blocking agents to abolish the shivering response was left to the discretion of the attending physician.
After a 24-hr hypothermia maintenance phase, patients were rewarmed actively using the ECD. The patient target temperature was increased by 0.2 °C every hour until normothermia was achieved. At the end of each nursing shift, bedside staff were administered a brief questionnaire inquiring about their perception of the ease-of-use, ease of handling, overall performance, observed ease of insertion, and how it compares to our standard cooling modalities.
A 70 year-old male (80 kg) was admitted to the ICU after a witnessed out-of-hospital ventricular fibrillation (VF) arrest. Sustained return of spontaneous circulation was achieved after 30 min of cardio-pulmonary resuscitation (CPR). On admission, his temperature was noted to be 36.4 °C. Following insertion of the ECD, a temperature reduction to the set target (35 °C) was achieved within 7 hr. Once attained, the patient stayed within ±0.5 °C of target for 100% of the hypothermia maintenance phase. Twelve hours into the maintenance phase, the patient sustained a second VF arrest. To maximize his chances of neurological recovery, hypothermia was thereafter extended for another 24 hr (total maintenance: 36 hr). Gradual active rewarming using the ECD was subsequently achieved successfully over 7 hr.
Return of neurologic function was delayed. On post arrest day 8, a head CT scan demonstrated diffuse hypoxic ischemic injury. However, brainstem reflexes were present, and somato-sensory evoked potential testing demonstrated intact responses. With continued care, gradual neurological recovery was noted and the patient was extubated on post arrest day 12. Following intensive occupational and physiotherapy, he was able to have basic conversation, and transfer and walk with two-person assistance. Five weeks post arrest, the patient was repatriated back to home hospital for ongoing rehabilitation.
A 79 year-old male (54.5 kg) was admitted to the intensive care unit following successful resuscitation from an out of hospital pulseless electrical activity (PEA) arrest. Initial recorded temperature on ICU admission was 34.1 °C. The ECD was inserted, setting the target temperature to 35 °C. Attaining this target was accomplished within 1 hr and 12 min. Subsequently the ECD was used to maintain hypothermia. Patient temperature was noted to stay within ±0.5 °C of target for 98.8% of the maintenance phase (24 hr). Slow rewarming was then undertaken using the ECD successfully over 8 hr 18 min.
The etiology of this patient's arrest was thought be hypoxic, secondary to a COPD exacerbation. Post-admission coronary angiography showed no culpable coronary artery disease. Despite his ICU stay being complicated by delirium tremens and seizures, his trajectory was that of gradual improvement. At post arrest day 6, he was extubated, moving all four limbs, and obeying commands. He was discharged home with a favorable neurological outcome, on post arrest day 26.
A 58 year-old male (79.5 kg) presented after successful resuscitation from an out-of-hospital witnessed VF arrest. Inferior wall ST elevation myocardial infarction was diagnosed on presentation to a peripheral hospital, upon which he received tenecteplase. He was then transported to our center for coronary angiography. This revealed severe triple-vessel coronary artery disease and a bare metal stent was deployed to the right coronary artery (RCA). He was then transferred to the intensive care unit for post arrest care.
Patient temperature upon arrival was 36.1 °C. An ECD was inserted and patient target was set to 35 °C. Target temperature was attained after 5 hr and 11 min, and was maintained within ±0.5 °C from target for 62.2% of the maintenance phase (27.5 hr). Rewarming was achieved successfully utilizing the ECD over 4 hr 12 min.
The patient regained consciousness and was extubated 36 hr post arrest. After demonstrating a favorable neurological recovery, he was discharged home on post arrest day 11.
A 94-year-old male was admitted to the ICU after resuscitation from a witnessed out-of-hospital PEA cardiac arrest. Initial temperature upon arrival was 37.1 °C. Following insertion of the ECD, a target temperature of 35 °C was achieved within 4 hr 12 min. Due to the patient's poor quality of life prior to admission however, the patient's substitute decision makers decided to withdraw life-sustaining measures. This decision was made after 7 hr and 42 min of starting the maintenance phase. During that interval, patient temperature was maintained, within ±0.5 °C of the set target for 100% of the time. After a 3 hr and 42 min rewarming phase, life support was stopped and the patient expired.
Summary of results
The overall mean rate of cooling was found to be 0.42 °C/hr (SD ± 0.26). On average, patients reached target temperature within 4 hr and 24 min (SD ± 2 hr 6 min) of ECD insertion. Once target was obtained, patients stayed within 0.5 °C of the set temperature for 90.25% (SD ± 16.20%) of the hypothermia protocol duration (Table 1). No complications attributable to ECD insertion or utilization were observed.
Nursing staff responses to the questionnaire were consistently positive with respect to ease-of-use, observed ease-of-insertion, and handling. Furthermore, nursing staff rated the device's efficacy and overall performance as excellent. Commentary from staff included statements as “excellent device”, “much easier to regulate patient temperature”, “patient not in wet bed (as happens with ice packs)”, and “device very efficient.” In addition, nurses rated the ECD as superior to our standard cooling modalities. This favorable rating may have been secondary to a reduction in nurse workload associated with device use.
To our knowledge, this is the first report to describe the use of an ECD as a sole temperature control modality in out-of-hospital cardiac arrest survivors. Cooling rates during the induction phase averaged 0.42 °C/hr (SD ± 0.26). When compared to published cooling rates of endovascular cooling catheters (median: 0.39 °C/hr, 25th to 75th percentile: 0.25–0.49 °C/hr) and surface cooling methods (median: 0.27 °C/hr, 25th to 75th percentile: 0.19–0.37 °C/hr), the ECD was found to be comparable.
Subsequently, patient temperatures were maintained within ±0.5 °C of set target for a mean of 90.25% (SD ± 16.20) of the maintenance phase. Normothermia was then successfully achieved in all four patients using the ECD as a rewarming device. The results suggest device efficacy, when used in isolation, in all three phases of therapeutic hypothermia: induction, maintenance and rewarming.
The ECD induces body temperature change through heat transfer between the esophagus and heart/major vessels.
The vena cavae, aorta and left atrium lie in close proximity to the esophagus leading to efficient heat exchange with the central circulation. Kulstad et al, were the first to demonstrate proof of concept, and feasibility of esophageal cooling in swine models.
In these initial animal studies, esophageal cooling was found to be safe with no thermal injurious effects detected on histological exam of the esophageal mucosa after 36 hr of use. Since the publication of these initial studies, additional case series reported successful ECD use in conjunction with other cooling modalities in postcardiac arrest patients.
Previously reported cooling rates however (0.52 °C/hr and 1.12 °C/hr), were faster than the rates observed in this report. Concomitant use of other cooling modalities likely hastened the hypothermia induction phase in the previous series. This is therefore the first unconfounded report characterizing ECD efficacy.
Time to reach target temperature in this series, was observed to be widely variable (range: 1 hr 12 min to 7 hr). Factors that may have contributed to this variability include different baseline temperatures, body masses, and different temperature trajectories prior to insertion of the device (patients attempting to mount a fever likely being more difficult to cool). In addition, administering medications or flushes (at room temperature) via the central port may counter cooling efficiency and contribute to the variability in time to attaining target temperature.
The esophageal approach to temperature management may have several advantages.
When compared to surface cooling techniques, the device is not patient-encompassing and therefore not obstructive to patient care. In comparison to intravascular cooling devices, ECD insertion was found to be easy and rapid, therefore enabling early initiation of cooling. In addition, the classic risks of endovascular cooling catheter insertion (thrombosis, infection, needle-stick injuries) are avoided. The risks of volume overload associated with the intravenous administration of refrigerated cold saline may potentially be averted by utilizing the ECD for hypothermia induction. Lastly, ECD catheters are MRI-compatible and can be disconnected from the external heat exchange unit for MRI studies, then reconnected upon patient return.
This approach however, has several limitations. As with any invasive modality, the risk of trauma and insertion failure are present. In addition, exposure of the esophagus to intraluminal cold temperatures has been demonstrated to reduce esophageal motility.
Whether this may impact clinically relevant outcomes or lead to complications (e.g. aspiration pneumonia) is unknown. Currently, the esophageal cooling device central (gastric) port is not approved for enteral feeding and medication administration. Its use for enteral medication administration in this report, and for feeding in previous series, despite being without complication remains off-label.
In addition, use of the esophageal cooling device is contraindicated in patients with esophageal strictures, recent upper gastro-intestinal bleeding, and patients with uncontrolled hemorrhagic diathesis.
Further research on this device should follow a pre-defined health technology assessment framework. The step-wise approach proposed by the IDEAL Collaboration (Table 2) constitutes a structured and pragmatic framework to guide future investigation.
Initial reports of ECD use and our current case series contribute to the first phase of this approach (phase 1/the idea phase). Given these early successes, the next phase (phase 2a/the development phase) should focus on technique modifications, developing indications, and defining roles for ECD use in various patient populations. Questions to be addressed in this phase include applicability of use in the general critically-ill neurological population, and whether this technique can be used with continuous enteral feeding without deterring its effectiveness. Phase 2b or the exploration phase, further examines safety and efficacy in these predefined populations. Currently ongoing prospective observational trials ascribe to this phase of the framework (NCT02327871 and NCT02387775 in postcardiac arrest patients, and NCT02420639 in traumatic brain injury patients). If the results from these prospective studies indicate favorable safety and efficacy, progression to Phase 3 (assessment phase) of the IDEAL framework would be incumbent. This entails formal study through adequately powered randomized controlled trials comparing the ECD to current standard temperature control modalities. Long-term monitoring for rare and late adverse effects and changes in the device use constitute the final and ongoing phase of this step-wise approach.
In summary, our report of 4 patients provides initial evidence of efficacy of the esophageal cooling device as a sole temperature control modality in post cardiac arrest patients. Its ease-of-use and ability to effectively control temperature during induction, maintenance, and rewarming make it an appealing modality in this patient population. Before its place in therapy can be defined however, further studies are required to quantify its benefits and risks compared to other available modalities.
We thank Advanced Cooling Therapy (Chicago, IL) for providing:
(1) The esophageal cooling devices used in this report.
(2) A laptop computer with Blanketrol III Data Export Software installed. This was used for automated temperature data collection.
The researchers did not receive any cash or specific grants from Advanced Cooling Therapy nor other funding agencies in the public, commercial, or not-for-profit sectors.