- Email: [email protected]
Femoro-iliacal artery versus pulmonary artery core temperature measurement during therapeutic hypothermia: An observational study
Resuscitation 84 (2013) 805–809
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
Femoro-iliacal artery versus pulmonary artery core temperature measurement during therapeutic hypothermia: An observational study夽 Danica Krizanac a , Peter Stratil a , David Hoerburger a , Christoph Testori a , Christian Wallmueller a , Andreas Schober a , Moritz Haugk a , Maria Haller b , Wilhelm Behringer a , Harald Herkner a , Fritz Sterz a , Michael Holzer a,∗ a b
Department of Emergency Medicine, Medical University of Vienna, Austria Department of Nephrology, Rheumatology, Hypertension and Transplantation, Hospital Elisabethinen, Fadingerstraße 1, 4020 Linz, Austria
a r t i c l e
i n f o
Article history: Received 12 October 2012 Received in revised form 19 November 2012 Accepted 21 November 2012 Keywords: Body temperature Induced hypothermia Cardiopulmonary resuscitation Catheterization Swan Ganz Pulse contour cardiac output Temperature measurement
a b s t r a c t Aim of the study: Therapeutic hypothermia after cardiac arrest improves neurologic outcome. The temperature measured in the pulmonary artery is considered to best reflect core temperature, yet is limited by invasiveness. Recently a femoro-arterial thermodilution catheter (PiCCO-Pulse Contour Cardiac Output) has been introduced in clinical practice as a safe and accurate haemodynamic monitoring system, which is also able to measure blood temperature. The aim of the study was to investigate, if the temperature measured with the PiCCO catheter reflects pulmonary artery temperature better than other sites during therapeutic hypothermia. Methods: In this observational study twenty patients after cardiac arrest and successful resuscitation were cooled with various cooling methods to 33 ± 1 ◦ C for 24 h, followed by rewarming. Temperatures were recorded continuously in the pulmonary artery (Tpa), femoro-iliacal artery (Tpicco), ear canal (Tear), oesophagus (Toeso) and urinary bladder (Tbla). We assessed agreement of methods using the Bland Altman approach including bias and limits of agreement (LA). Results: All other sites differed significantly from Tpa with the bias varying from 0.4 ◦ C (Tbla) to −0.6 ◦ C (Tear). Standard deviations varied from 0.1 ◦ C (Tpicco, Toeso) to 0.5 ◦ C (Tear). For all sites bias was closer to zero with increasing average temperatures. Bias tended to be larger in the cooling phase compared to overall measurements. Conclusions: Temperature measurement in the femoro-iliacal artery (Tpicco) reflects the gold standard of pulmonary artery temperature most accurately, especially during the cooling phase. Tpicco is easily accessible and might be used for monitoring core temperature without the need for additional temperature probes. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Therapeutic hypothermia (32–34 ◦ C) after cardiac arrest has been shown to improve neurologic outcome and survival1–4 and is being recommended by current guidelines.5,6 Despite the widespread use of therapeutic hypothermia after cardiac arrest, the question of the optimal site for core temperature measurement is still unanswered. Measuring brain temperature would be desirable, but is highly invasive and therefore not available in patients resuscitated from cardiac arrest. Other possible monitoring sites of central temperature as a surrogate of brain temperature are tympanum e.g.
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at doi:10.1016/j.resuscitation.2012.11.022. ∗ Corresponding author. Tel.: +43 1 40400 1964; fax: +43 1 40400 1965. E-mail address: [email protected] (M. Holzer). 0300-9572/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.resuscitation.2012.11.022
ear-canal, oesophagus, bladder, rectum, or pulmonary artery via a Swan-Ganz catheter. Of these, the temperature in the pulmonary artery might reflect best brain temperature7 and has advanced to the gold standard of core temperature measurement.8–11 Nevertheless, the routine use of the pulmonary artery catheter in intensive care patients is still a matter of debate.12,13 When evaluating possible surrogates for pulmonary artery temperature, it was generally agreed that it would be desirable to have an easy-to-use and accurate monitoring. Yet, from literature review it was evident that there are still a lot of conflicting opinions about the best surrogate for measuring core temperature.10,14–18 Recently, an arterial catheter with the capability of temperature measurement was developed (Pulsiocath Arterial Thermodilution Catheter 5F, Pulsion Medical Systems AG, Munich, Germany). It can be inserted via the femoral, radial, brachial or axillar artery and measures arterial pressure and temperature at the respective site. The aim of this study was to investigate, if the temperature measured in the iliacal artery
806
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
reflects core temperature measured in the pulmonary artery during induction and maintenance of therapeutic hypothermia, as well as during rewarming, in patients resuscitated from cardiac arrest.
iliacal artery temperatures were recorded by the PiCCO Monitor (Pulsion Medical Systems AG, Munich, Germany). 2.2. Statistics
2. Materials and methods In this observational cohort study, twenty patients who have been successfully resuscitated from cardiac arrest were included at an emergency department of a tertiary care university hospital. The period of recruitment was from October 2008 to July 2010. The study protocol was approved by the institutional ethics committee (Ethikkommission der Medizinischen Universität Wien, http://ethikkommission.meduniwien.ac.at) which waived the need for informed consent. The inclusion criteria were a cardiac arrest of any rhythm, an age over 18 years and a cardiogenic shock as clinical indication for placement of a pulmonary artery catheter. Exclusion criteria were a cardiac arrest due to trauma, intracranial bleeding, pregnancy, patients obeying verbal commands on admission, terminal illness before cardiac arrest, known pre-existing coagulopathy (except therapeutically induced) and oesophageal temperature below 34 ◦ C on admission.
Descriptive data are presented as absolute and relative frequency or median and interquartile range (IQR). To assess agreement between the measured temperatures we used the Bland Altman approach.20 This plot of difference versus average values was modified to allow for repeated measurements by connecting data points patient-wise. For calculations we used random effects regression models to allow for repeated measurements. This enabled us to directly calculate within and between standard deviations, appropriate estimates of bias, and appropriate 95% confidence intervals of the estimates. We used a linear random effects model to estimate mean bias. The difference of bias between two respective methods was estimated by entering measurement type as covariate into the model for each two measurements. To model the effect of average temperature on bias we entered average temperature as covariate into our linear random effects models. The Wald test was used for hypothesis testing. For data management and analysis we used Excel for Windows 2011 and Stata 11 (Stata Corp, College Station, USA). Generally a two-sided p-value less than 0.05 was considered statistically significant.
2.1. Study procedures The patients were cooled to an oesophageal temperature of 33 ◦ C for 24 h followed by rewarming. They were sedated with midazolam 0.125 mg/kg/h, and for analgesia fentanyl 0.002 mg/kg/h was used. To avoid shivering, a rocuronium bolus of 0.5 mg/kg and a continuous infusion of 0.5 mg/kg/h were used. During rewarming, sedation, analgesia, and paralysis were discontinued at an oesophageal temperature of 35 ◦ C as feasible. The following cooling methods were used: either circulating water filled cooling pads (Arctic Sun, Medivance, Louisville, USA), or precooled cooling pads (EmcoolsPads, Emcools, Vienna, Austria), or intravascular cooling catheters (Icy catheter and CoolGard, Zoll, Chelmsford, MA, USA or Accutrol and InnerCool RTx Endovascular System, Philips, Eindhoven, The Netherlands), and i.v. infusions of cold saline. When using intravascular cooling methods, we placed the PiCCO catheter on the other side to avoid influencing the temperature sensor. Oesophageal temperature (Mon-a-therm, General Purpose, 12 Fr, Mallinckrodt Medical Inc., St. Louis, USA) was used as temperature monitoring site for treatment decisions. To avoid kinking of the oesophageal temperature probe and to ensure proper placement, the temperature probe was advanced via a tracheal tube (size 6.5) placed into the lower third of the oesophagus. For measurement of ear canal temperature a thermistor based tympanic temperature probe (Mon-a-Therm, Tyco Healthcare, Gosport, UK) was positioned in the ear canal. A Foley catheter with a urine bladder temperature probe (Foley catheter, Medtronic Electronics Inc., Parker, CO, USA) was placed to measure bladder temperature. A pulmonary artery catheter (Swan-Ganz Standard Thermodilution Pulmonary Artery Catheter, Edwards Lifescience LLC, Irvine, CA, USA) was used to measure pulmonary artery temperature and to measure femoro-iliacal artery temperature a femoral artery catheter (Pulsiocath Arterial Thermodilution Catheter 5F, 20 cm, Pulsion Medical Systems AG, Munich, Germany) was inserted via standard Seldinger technique. Clinical data were recorded according to the recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest.19 Additional data recordings included all temperatures and haemodynamic parameters, which were recorded continuously via an automated data acquisition system (Philips IntelliVue MP70, Philips Healthcare, DA Best, The Netherlands). Femoro-
3. Results Twenty patients after cardiac arrest were included. Demographic and resuscitation specific data as well as the used cooling methods are presented in Table 1. The following cooling methods were used: either circulating water filled cooling pads (Arctic SunTM , Medivance, Louisville, USA; n = 3), or i.v. infusions of cold saline in combination with precooled cooling pads (Emcools PadsTM , Emcools, Pfaffstätten, Austria; n = 15), or intravascular cooling catheters (IcyTM catheter and CoolGardTM , Zoll, Chelmsford, MA, USA; n = 1; and AccutrolTM and InnerCool RTxTM Endovascular System, Philips, Eindhoven, The Netherlands; n = 1). The mean cooling rate was 0.9 ± 0.4 ◦ C/h and the mean rewarming rate 0.4 ◦ C ± 0.1 ◦ C/h. There were no significant differences in anaesthesia doses between patients and there was no shivering observed. In one patient intracerebral bleeding was diagnosed after the cooling procedure was finished. Fig. 1 shows the temperature course over time in one typical patient where each temperature site is plotted every minute. Mean differences from pulmonary artery temperature (Tpa) to various temperatures were significantly different and are shown in Table 2. Overall (during cooling, steady state and rewarming) femoro-iliacal artery temperature (Tpicco) showed the smallest mean difference, (SD between subject: 0.1 ◦ C and SD within subject: 0.1 ◦ C), oesophageal temperature (Toeso) showed the second Table 1 Demographic and resuscitation specific data and data on used cooling devices. Baseline parameters Median age – years (IQR)a Male sex – n, % Witnessed cardiac arrest – n, % Cardiac cause – n, % Ventricular fibrillation – n, % Asystole – n, % PEAa – n, % Time to return of spontaneous circulation – min (IQR)a Cooling with EmcoolsPads and cold fluid Cooling with Arctic Sun Cooling with Accutrol and RTx Endovascular System Cooling with Icy catheter and Coolgard a
Interquartile range (IQR); pulseless electric activity (PEA).
63.4 (43–88) 16 (80%) 17 (85%) 17 (85%) 14 (70%) 2 (10%) 4 (20%) 30 (7–89) 15 (75%) 3 (15%) 1 (5%) 1 (5%)
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
807
Fig. 1. Temperature course over time in one typical patient. Each temperature site is plotted every minute. Tpa, pulmonary artery temperature; Toeso, oesophageal temperature; Tpicco, femoro-iliacal artery temperature; Tbla, bladder temperature; Tear, ear canal temperature. Tear was constantly lower compared to Tpa.
Table 2 Mean differences from various temperature measurement sites compared to pulmonal artery temperature, overall (during cooling, steady state and rewarming) and during cooling phase only. Agreement overall
Tpicco-Tpa Toeso-Tpa Tbla-Tpa Tear-Tpa
Agreement during cooling phase only
Bias (95%CI) (◦ C)
Limits of agreement (◦ C) (±2SD overall)
SD within patients (◦ C)
Bias (95%CI) (◦ C)
Limits of agreement (◦ C) (±2SD overall)
SD within patients (◦ C)
0.0 (0.0–0.1) 0.1 (0.0–0.1) 0.1 (0.0–0.2) −0.6 (−0.8 to −0.3)
−0.2 to 0.2 −0.1 to 0.3 −0.3 to 0.5 −1.6 to 0.4
0.1 0.1 0.2 0.2
0.1 (0.0–0.1) 0.2 (0.1–0.3) 0.4 (0.1–0.6) −0.6 (−0.8 to −0.4)
−0.1 to 0.3 0.0 to 0.4 −0.2 to 1.0 −1.6 to 0.4
0.1 0.2 0.3 0.2
Tpa: pulmonary artery temperature; Toeso: oesophageal temperature; Tpicco: femoro-iliacal artery temperature; Tbla: bladder temperature; Tear: ear canal temperature; SD: standard deviation.
smallest mean difference (SD between subject: 0.1 ◦ C and SD within subject: 0.1 ◦ C). Bladder temperature (Tbla) showed the third smallest mean difference (SD between subject: 0.2 ◦ C and SD within subject: 0.2 ◦ C), and ear canal temperature (Tear) showed the biggest mean difference (SD between subject: 0.5 ◦ C and SD within subject: 0.2 ◦ C). Overall and during the cooling phase only Tpicco showed the smallest mean difference (Table 2). The Bland Altman plots are presented in Fig. 2. The individual bias showed to be significantly correlated with the average. With increasing average temperature the bias gets less, in the bladder temperature about 0.03 ◦ C, in the ear canal temperature about 0.05 ◦ C, in the iliacal artery temperature about 0.01 ◦ C and in the oesophageal temperature about 0.01 ◦ C (all p < 0.01, Fig. 2). When evaluating the response time of various temperature sites compared to pulmonary artery temperature we found a very long time lag for bladder temperature. Ear canal temperature was constantly lower compared to Tpa. Tpicco showed the smallest time lag (Table 3 and Fig. 1). One patient developed a device related pseudo aneurysm in the femoral artery and had to undergo surgery. After 6 months 8
Table 3 Time difference between temperature sites in achieving 33.5 ◦ C during the cooling phase. Time difference (min) during cooling phase, median (25th to 75th quartile) Tpicco-Tpa Toeso-Tpa Tbla-Tpa Tear-Tpa
4.5 (3–12.8) 9 (5–15) 30 (22.3–66.3) −38 (−65 to −23.5)
Tpa: pulmonary artery temperature; Toeso: oesophageal temperature; Tpicco: femoro-iliacal artery temperature; Tbla: bladder temperature, Tear: ear canal temperature.
patients had an overall performance category score (OPC) of 1 or 2, one patient had an OPC of 3 and 11 patients had died. 4. Discussion Our results showed that femoro-iliacal artery temperature reflects the gold standard of pulmonary artery temperature most accurately when compared to other temperature sites in measuring core temperature during therapeutic hypothermia in patients resuscitated from cardiac arrest. It has been shown in animal models of cardiac arrest, that hypothermia should be induced as early and rapidly as possible in order to improve outcome and survival.21 Thus, an accurate, fast and easy to use temperature monitoring would be needed to exactly measure temperature during therapeutic hypothermia treatment. In order to ensure the beneficial effects of hypothermia, the adherence to the target temperature is very important. Several studies documented that overcooling is associated with a risk of severe complications, like atrial or ventricular fibrillation, coagulopathy, an increased risk of infections, and might be associated with a worse outcome.22–26 Furthermore, at temperatures of less than 30 ◦ C electric and pharmaceutical anti-arrhythmic therapies may be ineffective.27 Hence, precise temperature control is an important factor for the successful use of therapeutic hypothermia. The best temperature measurement site and the respective reliability reflecting core temperature is still being discussed.10,14,15,18 There is no generally established definition for the degree of bias and variability which can be considered clinically reliable. Various methods for assessing agreement give conflicting results and make interpretation and direct comparison difficult. There are three important components to consider when deciding whether two sets of measurements are in agreement or not: (1) the degree of linear relationship between the two sets; (2) the amount of
808
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
Fig. 2. Experiment wise modified Bland–Altman plots of temperature differences against averages; Tpa, pulmonary artery temperature; Toeso, oesophageal temperature; Tpicco, femoro-iliacal artery temperature; Tbla, bladder temperature; Tear ear canal temperature. A value near zero implies concordance. Dotted lines represent the limits of agreement (bias ± 2 standard deviation).
bias as represented by the difference in the means; and (3) the difference between the two variances.20 The better an individual measure addresses these three components, the better it indicates agreement.28 When judging the clinical acceptability of a given bias or variance, it has to be distinguished between steady state conditions – where most sites reflect core temperature10,14 – and induction of cooling with fast changes in temperatures. It has to be considered, that a faster cooling rate would lead to an increase in bias, variance and response time.14,15 Referring to these facts, we evaluated bias, variance and the response time of various temperature probes compared to pulmonary artery temperature (Tpa) (Tables 2 and 3). Overall (during cooling, steady state and rewarming) we found smaller differences in bias and variability between the temperature probes (compared to Tpa) then during the cooling phase (Table 2). Bladder temperature (Tbla) also showed a very long response time to temperature changes (Table 3) which could lead to misinterpretation of true patient temperature during rapid temperature changes. Therefore, Tbla is not recommended during the cooling phase, but can be used for steady state measurements.10,14,29,30 Oesophageal temperature (Toeso), which is increasingly used as standard measurement site during rapid induction of therapeutic hypothermia, showed similar bias and variability like the femoro-iliacal temperature when compared to pulmonary artery temperature. Compared to femoro-iliacal temperature it showed a significant grater bias and variance. Ear canal temperature (Tear) continuously and substantially underestimated Tpa during cooling as well as during steady state. The big variance of this measurement site makes its use unattractive for cooling as well as for steady state conditions (Fig. 2). There have been contrary results of the usefulness of ear canal
temperature,8,9,11,14,31 which might be due to the difficulty to find the correct position of the probe,32 which in most of the published studies was not verified. Using the femoro-iliacal artery thermometry could facilitate the implementation of therapeutic hypothermia and minimize the risk of overcooling as it is an accurate and easy to use temperature monitoring site. Since the placement of an arterial catheter is a necessary part of intensive care of post-cardiac-arrest-patients anyhow, use of femoro-iliacal temperature and blood pressure monitoring with the PiCCO catheter would make an additional placement of a temperature probe redundant. There are also downsides of the femoro-iliacal artery thermometry to mention. Though fairly easy to place it is still an invasive procedure and possible complications like pseudo aneurysms, infections or thrombosis33 have to be considered. We have several limitations to mention. The PiCCO catheter could also be placed in the radial, brachial or axillar artery. Those application sites were not tested. In this study we preferred a central measurement site because we assume that peripheral measurement sites, due to peripheral vasoconstriction during hypothermia, may not be as accurate as central sites. Furthermore we did not investigate possible local effects of external cooling methods on nearby sensors which could probably influence measurements. This has to be evaluated in further studies. When evaluating data of only twenty patients, the presumably small explanatory power of statistical significance has to be considered. 5. Conclusion Femoro-iliacal temperature measurement with PiCCO (Pulse Contour Cardiac Output) catheter reflects the core temperature
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
gold standard (pulmonary artery temperature) more accurate than other temperature measurement sites, showing a very small bias, variance and time lag even during the cooling phase. This temperature site is easily accessible and the pressure port could be used for arterial pressure monitoring, which is needed anyhow in post-cardiac-arrest-patients. Conflict of interest There are no conflicts of interests. Acknowledgments The study was supported by funds of the Oesterreichische Nationalbank (Anniversary Fund, project number: AP12934ONB). The sponsor was not involved in the study design, collection, analysis and interpretation of data, writing of the manuscript and in the decision to submit the manuscript for publication. With the money received, amongst other things, David Hoerburger was employed for patient recruitment and data collection. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.resuscitation. 2012.11.022. References 1. Sterz F, Holzer M, Roine R, et al. Hypothermia after cardiac arrest: a treatment that works. Curr Opin Crit Care 2003;9:205–10. 2. Nolan JP, Deakin CD, Soar J, et al. European resuscitation council guidelines for resuscitation. Resuscitation 2005;67:S39–86. 3. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002;346:557–63. 4. Hypothermia after cardiac arrest study group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549–56. 5. Committee ECC, Subcommittees and Task Forces of the American Heart Association. American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2005;112:IV1–203. 6. Deakin CD, Nolan JP, Soar J, et al. European resuscitation council guidelines for resuscitation 2010. Resuscitation 2010;81:1305–52. 7. Ao H, Moon JK, Tanimoto H, et al. Jugular vein temperature reflects brain temperature during hypothermia. Resuscitation 2000;45:111–8. 8. Rotello LC, Crawford L, Terndrup TE. Comparison of infrared ear thermometer derived and equilibrated rectal temperatures in estimating pulmonary artery temperatures. Crit Care Med 1996;24:1501–6. 9. Moran JL, Peter JV, Solomon PJ, et al. Tympanic temperature measurements: are they reliable in the critically ill? A clinical study of measures of agreement. Crit Care Med 2007;35:155–64.
809
10. Lefrant JY, Muller L, de La Coussaye JE, et al. Temperature measurement in intensive care patients: comparison of urinary bladder, oesophageal, rectal, axillary, and inguinal methods versus pulmonary artery core method. Intensive Care Med 2003;29:414–8. 11. Giuliano KK, Scott SS, Elliot S, et al. Temperature measurement in critically ill orally intubated adults: a comparison of pulmonary artery core, tympanic, and oral methods. Crit Care Med 1999;27:2188–93. 12. Pinsky MR, Vincent JL. Let us use the pulmonary artery catheter correctly and only when we need it. Crit Care Med 2005;33:1119–22. 13. Levin PD, Sprung CL. Another point of view: no swan song for the pulmonary artery catheter. Crit Care Med 2005;33:1123–4. 14. Stone JG, Young WL, Smith CR, et al. Do standard monitoring sites reflect true brain temperature when profound hypothermia is rapidly induced and reversed? Anesthesiology 1995;82:344–51. 15. Krizanac D, Haugk M, Sterz, et al. Tracheal temperature for monitoring body temperature during mild hypothermia in pigs. Resuscitation 2010;81:87–92. 16. Haugk M, Stratil P, Sterz F, et al. Temperature monitored on the cuff surface of an endotracheal tube reflects body temperature. Crit Care Med 2010;38:1569–73. 17. Zeiner A, Klewer J, Sterz F, et al. Non-invasive continuous cerebral temperature monitoring in patients treated with mild therapeutic hypothermia: an observational pilot study. Resuscitation 2010;81:861–6. 18. Erickson RS. The continuing question of how best to measure body temperature. Crit Care Med 1999;27:2307–10. 19. Cummins RO. The Utstein style for uniform reporting of data from out-ofhospital cardiac arrest. Ann Emerg Med 1993;22:37–40. 20. Bland JM, Altman DG. Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet 1995;346: 1085–7. 21. Pomeranz S, Safar P, Radovsky A, et al. The effect of resuscitative moderate hypothermia following epidural brain compression on cerebral damage in a canine outcome model. J Neurosurg 1993;79:241–51. 22. Merchant RM, Abella BS, Peberdy MA, et al. Therapeutic hypothermia after cardiac arrest: unintentional overcooling is common using ice packs and conventional cooling blankets. Crit Care Med 2006;34:S490–4. 23. Polderman KH. Application of therapeutic hypothermia in the intensive care unit. Opportunities and pitfalls of a promising treatment modality—Part 2: practical aspects and side effects. Intensive Care Med 2004;30:757–69. 24. Rohrer MJ, Natale AM. Effect of hypothermia on the coagulation cascade. Crit Care Med 1992;20:1402–5. 25. Sessler DI. Complications and treatment of mild hypothermia. Anesthesiology 2001;95:531–43. 26. Danzl DF, Pozos RS. Accidental hypothermia. N Engl J Med 1994;331:1756–60. 27. Reuler JB, Hypothermia:. pathophysiology, clinical settings, and management. Ann Intern Med 1978;89:519–27. 28. Sanchez MM, Binkowitz BS. Guidelines for measurement validation in clinical trial design. J Biopharm Stat 1999;9:417–38. 29. Akata T, Yamaura K, Kandabashi T, et al. Changes in body temperature during profound hypothermic cardiopulmonary bypass in adult patients undergoing aortic arch reconstruction. J Anesth 2004;18:73–81. 30. Akata T, Setoguchi H, Shirozu K, et al. Reliability of temperatures measured at standard monitoring sites as an index of brain temperature during deep hypothermic cardiopulmonary bypass conducted for thoracic aortic reconstruction. J Thorac Cardiovasc Surg 2007;133:1559–65. 31. McCarthy PW, Heusch AI. The vagaries of ear temperature assessment. J Med Eng Technol 2006;30:242–51. 32. Robinson J, Charlton J, Seal R, et al. Oesophageal, rectal, axillary, tympanic and pulmonary artery temperatures during cardiac surgery. Can J Anaesth 1998;45:317–23. 33. Taylor EE, Carroll JP, Lovitt MA, et al. Active intravascular rewarming for hypothermia associated with traumatic injury: early experience with a new technique. Proc (Bayl Univ Med Cent) 2008;21:120–6.
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
Femoro-iliacal artery versus pulmonary artery core temperature measurement during therapeutic hypothermia: An observational study夽 Danica Krizanac a , Peter Stratil a , David Hoerburger a , Christoph Testori a , Christian Wallmueller a , Andreas Schober a , Moritz Haugk a , Maria Haller b , Wilhelm Behringer a , Harald Herkner a , Fritz Sterz a , Michael Holzer a,∗ a b
Department of Emergency Medicine, Medical University of Vienna, Austria Department of Nephrology, Rheumatology, Hypertension and Transplantation, Hospital Elisabethinen, Fadingerstraße 1, 4020 Linz, Austria
a r t i c l e
i n f o
Article history: Received 12 October 2012 Received in revised form 19 November 2012 Accepted 21 November 2012 Keywords: Body temperature Induced hypothermia Cardiopulmonary resuscitation Catheterization Swan Ganz Pulse contour cardiac output Temperature measurement
a b s t r a c t Aim of the study: Therapeutic hypothermia after cardiac arrest improves neurologic outcome. The temperature measured in the pulmonary artery is considered to best reflect core temperature, yet is limited by invasiveness. Recently a femoro-arterial thermodilution catheter (PiCCO-Pulse Contour Cardiac Output) has been introduced in clinical practice as a safe and accurate haemodynamic monitoring system, which is also able to measure blood temperature. The aim of the study was to investigate, if the temperature measured with the PiCCO catheter reflects pulmonary artery temperature better than other sites during therapeutic hypothermia. Methods: In this observational study twenty patients after cardiac arrest and successful resuscitation were cooled with various cooling methods to 33 ± 1 ◦ C for 24 h, followed by rewarming. Temperatures were recorded continuously in the pulmonary artery (Tpa), femoro-iliacal artery (Tpicco), ear canal (Tear), oesophagus (Toeso) and urinary bladder (Tbla). We assessed agreement of methods using the Bland Altman approach including bias and limits of agreement (LA). Results: All other sites differed significantly from Tpa with the bias varying from 0.4 ◦ C (Tbla) to −0.6 ◦ C (Tear). Standard deviations varied from 0.1 ◦ C (Tpicco, Toeso) to 0.5 ◦ C (Tear). For all sites bias was closer to zero with increasing average temperatures. Bias tended to be larger in the cooling phase compared to overall measurements. Conclusions: Temperature measurement in the femoro-iliacal artery (Tpicco) reflects the gold standard of pulmonary artery temperature most accurately, especially during the cooling phase. Tpicco is easily accessible and might be used for monitoring core temperature without the need for additional temperature probes. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Therapeutic hypothermia (32–34 ◦ C) after cardiac arrest has been shown to improve neurologic outcome and survival1–4 and is being recommended by current guidelines.5,6 Despite the widespread use of therapeutic hypothermia after cardiac arrest, the question of the optimal site for core temperature measurement is still unanswered. Measuring brain temperature would be desirable, but is highly invasive and therefore not available in patients resuscitated from cardiac arrest. Other possible monitoring sites of central temperature as a surrogate of brain temperature are tympanum e.g.
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at doi:10.1016/j.resuscitation.2012.11.022. ∗ Corresponding author. Tel.: +43 1 40400 1964; fax: +43 1 40400 1965. E-mail address: [email protected] (M. Holzer). 0300-9572/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.resuscitation.2012.11.022
ear-canal, oesophagus, bladder, rectum, or pulmonary artery via a Swan-Ganz catheter. Of these, the temperature in the pulmonary artery might reflect best brain temperature7 and has advanced to the gold standard of core temperature measurement.8–11 Nevertheless, the routine use of the pulmonary artery catheter in intensive care patients is still a matter of debate.12,13 When evaluating possible surrogates for pulmonary artery temperature, it was generally agreed that it would be desirable to have an easy-to-use and accurate monitoring. Yet, from literature review it was evident that there are still a lot of conflicting opinions about the best surrogate for measuring core temperature.10,14–18 Recently, an arterial catheter with the capability of temperature measurement was developed (Pulsiocath Arterial Thermodilution Catheter 5F, Pulsion Medical Systems AG, Munich, Germany). It can be inserted via the femoral, radial, brachial or axillar artery and measures arterial pressure and temperature at the respective site. The aim of this study was to investigate, if the temperature measured in the iliacal artery
806
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
reflects core temperature measured in the pulmonary artery during induction and maintenance of therapeutic hypothermia, as well as during rewarming, in patients resuscitated from cardiac arrest.
iliacal artery temperatures were recorded by the PiCCO Monitor (Pulsion Medical Systems AG, Munich, Germany). 2.2. Statistics
2. Materials and methods In this observational cohort study, twenty patients who have been successfully resuscitated from cardiac arrest were included at an emergency department of a tertiary care university hospital. The period of recruitment was from October 2008 to July 2010. The study protocol was approved by the institutional ethics committee (Ethikkommission der Medizinischen Universität Wien, http://ethikkommission.meduniwien.ac.at) which waived the need for informed consent. The inclusion criteria were a cardiac arrest of any rhythm, an age over 18 years and a cardiogenic shock as clinical indication for placement of a pulmonary artery catheter. Exclusion criteria were a cardiac arrest due to trauma, intracranial bleeding, pregnancy, patients obeying verbal commands on admission, terminal illness before cardiac arrest, known pre-existing coagulopathy (except therapeutically induced) and oesophageal temperature below 34 ◦ C on admission.
Descriptive data are presented as absolute and relative frequency or median and interquartile range (IQR). To assess agreement between the measured temperatures we used the Bland Altman approach.20 This plot of difference versus average values was modified to allow for repeated measurements by connecting data points patient-wise. For calculations we used random effects regression models to allow for repeated measurements. This enabled us to directly calculate within and between standard deviations, appropriate estimates of bias, and appropriate 95% confidence intervals of the estimates. We used a linear random effects model to estimate mean bias. The difference of bias between two respective methods was estimated by entering measurement type as covariate into the model for each two measurements. To model the effect of average temperature on bias we entered average temperature as covariate into our linear random effects models. The Wald test was used for hypothesis testing. For data management and analysis we used Excel for Windows 2011 and Stata 11 (Stata Corp, College Station, USA). Generally a two-sided p-value less than 0.05 was considered statistically significant.
2.1. Study procedures The patients were cooled to an oesophageal temperature of 33 ◦ C for 24 h followed by rewarming. They were sedated with midazolam 0.125 mg/kg/h, and for analgesia fentanyl 0.002 mg/kg/h was used. To avoid shivering, a rocuronium bolus of 0.5 mg/kg and a continuous infusion of 0.5 mg/kg/h were used. During rewarming, sedation, analgesia, and paralysis were discontinued at an oesophageal temperature of 35 ◦ C as feasible. The following cooling methods were used: either circulating water filled cooling pads (Arctic Sun, Medivance, Louisville, USA), or precooled cooling pads (EmcoolsPads, Emcools, Vienna, Austria), or intravascular cooling catheters (Icy catheter and CoolGard, Zoll, Chelmsford, MA, USA or Accutrol and InnerCool RTx Endovascular System, Philips, Eindhoven, The Netherlands), and i.v. infusions of cold saline. When using intravascular cooling methods, we placed the PiCCO catheter on the other side to avoid influencing the temperature sensor. Oesophageal temperature (Mon-a-therm, General Purpose, 12 Fr, Mallinckrodt Medical Inc., St. Louis, USA) was used as temperature monitoring site for treatment decisions. To avoid kinking of the oesophageal temperature probe and to ensure proper placement, the temperature probe was advanced via a tracheal tube (size 6.5) placed into the lower third of the oesophagus. For measurement of ear canal temperature a thermistor based tympanic temperature probe (Mon-a-Therm, Tyco Healthcare, Gosport, UK) was positioned in the ear canal. A Foley catheter with a urine bladder temperature probe (Foley catheter, Medtronic Electronics Inc., Parker, CO, USA) was placed to measure bladder temperature. A pulmonary artery catheter (Swan-Ganz Standard Thermodilution Pulmonary Artery Catheter, Edwards Lifescience LLC, Irvine, CA, USA) was used to measure pulmonary artery temperature and to measure femoro-iliacal artery temperature a femoral artery catheter (Pulsiocath Arterial Thermodilution Catheter 5F, 20 cm, Pulsion Medical Systems AG, Munich, Germany) was inserted via standard Seldinger technique. Clinical data were recorded according to the recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest.19 Additional data recordings included all temperatures and haemodynamic parameters, which were recorded continuously via an automated data acquisition system (Philips IntelliVue MP70, Philips Healthcare, DA Best, The Netherlands). Femoro-
3. Results Twenty patients after cardiac arrest were included. Demographic and resuscitation specific data as well as the used cooling methods are presented in Table 1. The following cooling methods were used: either circulating water filled cooling pads (Arctic SunTM , Medivance, Louisville, USA; n = 3), or i.v. infusions of cold saline in combination with precooled cooling pads (Emcools PadsTM , Emcools, Pfaffstätten, Austria; n = 15), or intravascular cooling catheters (IcyTM catheter and CoolGardTM , Zoll, Chelmsford, MA, USA; n = 1; and AccutrolTM and InnerCool RTxTM Endovascular System, Philips, Eindhoven, The Netherlands; n = 1). The mean cooling rate was 0.9 ± 0.4 ◦ C/h and the mean rewarming rate 0.4 ◦ C ± 0.1 ◦ C/h. There were no significant differences in anaesthesia doses between patients and there was no shivering observed. In one patient intracerebral bleeding was diagnosed after the cooling procedure was finished. Fig. 1 shows the temperature course over time in one typical patient where each temperature site is plotted every minute. Mean differences from pulmonary artery temperature (Tpa) to various temperatures were significantly different and are shown in Table 2. Overall (during cooling, steady state and rewarming) femoro-iliacal artery temperature (Tpicco) showed the smallest mean difference, (SD between subject: 0.1 ◦ C and SD within subject: 0.1 ◦ C), oesophageal temperature (Toeso) showed the second Table 1 Demographic and resuscitation specific data and data on used cooling devices. Baseline parameters Median age – years (IQR)a Male sex – n, % Witnessed cardiac arrest – n, % Cardiac cause – n, % Ventricular fibrillation – n, % Asystole – n, % PEAa – n, % Time to return of spontaneous circulation – min (IQR)a Cooling with EmcoolsPads and cold fluid Cooling with Arctic Sun Cooling with Accutrol and RTx Endovascular System Cooling with Icy catheter and Coolgard a
Interquartile range (IQR); pulseless electric activity (PEA).
63.4 (43–88) 16 (80%) 17 (85%) 17 (85%) 14 (70%) 2 (10%) 4 (20%) 30 (7–89) 15 (75%) 3 (15%) 1 (5%) 1 (5%)
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
807
Fig. 1. Temperature course over time in one typical patient. Each temperature site is plotted every minute. Tpa, pulmonary artery temperature; Toeso, oesophageal temperature; Tpicco, femoro-iliacal artery temperature; Tbla, bladder temperature; Tear, ear canal temperature. Tear was constantly lower compared to Tpa.
Table 2 Mean differences from various temperature measurement sites compared to pulmonal artery temperature, overall (during cooling, steady state and rewarming) and during cooling phase only. Agreement overall
Tpicco-Tpa Toeso-Tpa Tbla-Tpa Tear-Tpa
Agreement during cooling phase only
Bias (95%CI) (◦ C)
Limits of agreement (◦ C) (±2SD overall)
SD within patients (◦ C)
Bias (95%CI) (◦ C)
Limits of agreement (◦ C) (±2SD overall)
SD within patients (◦ C)
0.0 (0.0–0.1) 0.1 (0.0–0.1) 0.1 (0.0–0.2) −0.6 (−0.8 to −0.3)
−0.2 to 0.2 −0.1 to 0.3 −0.3 to 0.5 −1.6 to 0.4
0.1 0.1 0.2 0.2
0.1 (0.0–0.1) 0.2 (0.1–0.3) 0.4 (0.1–0.6) −0.6 (−0.8 to −0.4)
−0.1 to 0.3 0.0 to 0.4 −0.2 to 1.0 −1.6 to 0.4
0.1 0.2 0.3 0.2
Tpa: pulmonary artery temperature; Toeso: oesophageal temperature; Tpicco: femoro-iliacal artery temperature; Tbla: bladder temperature; Tear: ear canal temperature; SD: standard deviation.
smallest mean difference (SD between subject: 0.1 ◦ C and SD within subject: 0.1 ◦ C). Bladder temperature (Tbla) showed the third smallest mean difference (SD between subject: 0.2 ◦ C and SD within subject: 0.2 ◦ C), and ear canal temperature (Tear) showed the biggest mean difference (SD between subject: 0.5 ◦ C and SD within subject: 0.2 ◦ C). Overall and during the cooling phase only Tpicco showed the smallest mean difference (Table 2). The Bland Altman plots are presented in Fig. 2. The individual bias showed to be significantly correlated with the average. With increasing average temperature the bias gets less, in the bladder temperature about 0.03 ◦ C, in the ear canal temperature about 0.05 ◦ C, in the iliacal artery temperature about 0.01 ◦ C and in the oesophageal temperature about 0.01 ◦ C (all p < 0.01, Fig. 2). When evaluating the response time of various temperature sites compared to pulmonary artery temperature we found a very long time lag for bladder temperature. Ear canal temperature was constantly lower compared to Tpa. Tpicco showed the smallest time lag (Table 3 and Fig. 1). One patient developed a device related pseudo aneurysm in the femoral artery and had to undergo surgery. After 6 months 8
Table 3 Time difference between temperature sites in achieving 33.5 ◦ C during the cooling phase. Time difference (min) during cooling phase, median (25th to 75th quartile) Tpicco-Tpa Toeso-Tpa Tbla-Tpa Tear-Tpa
4.5 (3–12.8) 9 (5–15) 30 (22.3–66.3) −38 (−65 to −23.5)
Tpa: pulmonary artery temperature; Toeso: oesophageal temperature; Tpicco: femoro-iliacal artery temperature; Tbla: bladder temperature, Tear: ear canal temperature.
patients had an overall performance category score (OPC) of 1 or 2, one patient had an OPC of 3 and 11 patients had died. 4. Discussion Our results showed that femoro-iliacal artery temperature reflects the gold standard of pulmonary artery temperature most accurately when compared to other temperature sites in measuring core temperature during therapeutic hypothermia in patients resuscitated from cardiac arrest. It has been shown in animal models of cardiac arrest, that hypothermia should be induced as early and rapidly as possible in order to improve outcome and survival.21 Thus, an accurate, fast and easy to use temperature monitoring would be needed to exactly measure temperature during therapeutic hypothermia treatment. In order to ensure the beneficial effects of hypothermia, the adherence to the target temperature is very important. Several studies documented that overcooling is associated with a risk of severe complications, like atrial or ventricular fibrillation, coagulopathy, an increased risk of infections, and might be associated with a worse outcome.22–26 Furthermore, at temperatures of less than 30 ◦ C electric and pharmaceutical anti-arrhythmic therapies may be ineffective.27 Hence, precise temperature control is an important factor for the successful use of therapeutic hypothermia. The best temperature measurement site and the respective reliability reflecting core temperature is still being discussed.10,14,15,18 There is no generally established definition for the degree of bias and variability which can be considered clinically reliable. Various methods for assessing agreement give conflicting results and make interpretation and direct comparison difficult. There are three important components to consider when deciding whether two sets of measurements are in agreement or not: (1) the degree of linear relationship between the two sets; (2) the amount of
808
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
Fig. 2. Experiment wise modified Bland–Altman plots of temperature differences against averages; Tpa, pulmonary artery temperature; Toeso, oesophageal temperature; Tpicco, femoro-iliacal artery temperature; Tbla, bladder temperature; Tear ear canal temperature. A value near zero implies concordance. Dotted lines represent the limits of agreement (bias ± 2 standard deviation).
bias as represented by the difference in the means; and (3) the difference between the two variances.20 The better an individual measure addresses these three components, the better it indicates agreement.28 When judging the clinical acceptability of a given bias or variance, it has to be distinguished between steady state conditions – where most sites reflect core temperature10,14 – and induction of cooling with fast changes in temperatures. It has to be considered, that a faster cooling rate would lead to an increase in bias, variance and response time.14,15 Referring to these facts, we evaluated bias, variance and the response time of various temperature probes compared to pulmonary artery temperature (Tpa) (Tables 2 and 3). Overall (during cooling, steady state and rewarming) we found smaller differences in bias and variability between the temperature probes (compared to Tpa) then during the cooling phase (Table 2). Bladder temperature (Tbla) also showed a very long response time to temperature changes (Table 3) which could lead to misinterpretation of true patient temperature during rapid temperature changes. Therefore, Tbla is not recommended during the cooling phase, but can be used for steady state measurements.10,14,29,30 Oesophageal temperature (Toeso), which is increasingly used as standard measurement site during rapid induction of therapeutic hypothermia, showed similar bias and variability like the femoro-iliacal temperature when compared to pulmonary artery temperature. Compared to femoro-iliacal temperature it showed a significant grater bias and variance. Ear canal temperature (Tear) continuously and substantially underestimated Tpa during cooling as well as during steady state. The big variance of this measurement site makes its use unattractive for cooling as well as for steady state conditions (Fig. 2). There have been contrary results of the usefulness of ear canal
temperature,8,9,11,14,31 which might be due to the difficulty to find the correct position of the probe,32 which in most of the published studies was not verified. Using the femoro-iliacal artery thermometry could facilitate the implementation of therapeutic hypothermia and minimize the risk of overcooling as it is an accurate and easy to use temperature monitoring site. Since the placement of an arterial catheter is a necessary part of intensive care of post-cardiac-arrest-patients anyhow, use of femoro-iliacal temperature and blood pressure monitoring with the PiCCO catheter would make an additional placement of a temperature probe redundant. There are also downsides of the femoro-iliacal artery thermometry to mention. Though fairly easy to place it is still an invasive procedure and possible complications like pseudo aneurysms, infections or thrombosis33 have to be considered. We have several limitations to mention. The PiCCO catheter could also be placed in the radial, brachial or axillar artery. Those application sites were not tested. In this study we preferred a central measurement site because we assume that peripheral measurement sites, due to peripheral vasoconstriction during hypothermia, may not be as accurate as central sites. Furthermore we did not investigate possible local effects of external cooling methods on nearby sensors which could probably influence measurements. This has to be evaluated in further studies. When evaluating data of only twenty patients, the presumably small explanatory power of statistical significance has to be considered. 5. Conclusion Femoro-iliacal temperature measurement with PiCCO (Pulse Contour Cardiac Output) catheter reflects the core temperature
D. Krizanac et al. / Resuscitation 84 (2013) 805–809
gold standard (pulmonary artery temperature) more accurate than other temperature measurement sites, showing a very small bias, variance and time lag even during the cooling phase. This temperature site is easily accessible and the pressure port could be used for arterial pressure monitoring, which is needed anyhow in post-cardiac-arrest-patients. Conflict of interest There are no conflicts of interests. Acknowledgments The study was supported by funds of the Oesterreichische Nationalbank (Anniversary Fund, project number: AP12934ONB). The sponsor was not involved in the study design, collection, analysis and interpretation of data, writing of the manuscript and in the decision to submit the manuscript for publication. With the money received, amongst other things, David Hoerburger was employed for patient recruitment and data collection. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.resuscitation. 2012.11.022. References 1. Sterz F, Holzer M, Roine R, et al. Hypothermia after cardiac arrest: a treatment that works. Curr Opin Crit Care 2003;9:205–10. 2. Nolan JP, Deakin CD, Soar J, et al. European resuscitation council guidelines for resuscitation. Resuscitation 2005;67:S39–86. 3. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002;346:557–63. 4. Hypothermia after cardiac arrest study group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549–56. 5. Committee ECC, Subcommittees and Task Forces of the American Heart Association. American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2005;112:IV1–203. 6. Deakin CD, Nolan JP, Soar J, et al. European resuscitation council guidelines for resuscitation 2010. Resuscitation 2010;81:1305–52. 7. Ao H, Moon JK, Tanimoto H, et al. Jugular vein temperature reflects brain temperature during hypothermia. Resuscitation 2000;45:111–8. 8. Rotello LC, Crawford L, Terndrup TE. Comparison of infrared ear thermometer derived and equilibrated rectal temperatures in estimating pulmonary artery temperatures. Crit Care Med 1996;24:1501–6. 9. Moran JL, Peter JV, Solomon PJ, et al. Tympanic temperature measurements: are they reliable in the critically ill? A clinical study of measures of agreement. Crit Care Med 2007;35:155–64.
809
10. Lefrant JY, Muller L, de La Coussaye JE, et al. Temperature measurement in intensive care patients: comparison of urinary bladder, oesophageal, rectal, axillary, and inguinal methods versus pulmonary artery core method. Intensive Care Med 2003;29:414–8. 11. Giuliano KK, Scott SS, Elliot S, et al. Temperature measurement in critically ill orally intubated adults: a comparison of pulmonary artery core, tympanic, and oral methods. Crit Care Med 1999;27:2188–93. 12. Pinsky MR, Vincent JL. Let us use the pulmonary artery catheter correctly and only when we need it. Crit Care Med 2005;33:1119–22. 13. Levin PD, Sprung CL. Another point of view: no swan song for the pulmonary artery catheter. Crit Care Med 2005;33:1123–4. 14. Stone JG, Young WL, Smith CR, et al. Do standard monitoring sites reflect true brain temperature when profound hypothermia is rapidly induced and reversed? Anesthesiology 1995;82:344–51. 15. Krizanac D, Haugk M, Sterz, et al. Tracheal temperature for monitoring body temperature during mild hypothermia in pigs. Resuscitation 2010;81:87–92. 16. Haugk M, Stratil P, Sterz F, et al. Temperature monitored on the cuff surface of an endotracheal tube reflects body temperature. Crit Care Med 2010;38:1569–73. 17. Zeiner A, Klewer J, Sterz F, et al. Non-invasive continuous cerebral temperature monitoring in patients treated with mild therapeutic hypothermia: an observational pilot study. Resuscitation 2010;81:861–6. 18. Erickson RS. The continuing question of how best to measure body temperature. Crit Care Med 1999;27:2307–10. 19. Cummins RO. The Utstein style for uniform reporting of data from out-ofhospital cardiac arrest. Ann Emerg Med 1993;22:37–40. 20. Bland JM, Altman DG. Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet 1995;346: 1085–7. 21. Pomeranz S, Safar P, Radovsky A, et al. The effect of resuscitative moderate hypothermia following epidural brain compression on cerebral damage in a canine outcome model. J Neurosurg 1993;79:241–51. 22. Merchant RM, Abella BS, Peberdy MA, et al. Therapeutic hypothermia after cardiac arrest: unintentional overcooling is common using ice packs and conventional cooling blankets. Crit Care Med 2006;34:S490–4. 23. Polderman KH. Application of therapeutic hypothermia in the intensive care unit. Opportunities and pitfalls of a promising treatment modality—Part 2: practical aspects and side effects. Intensive Care Med 2004;30:757–69. 24. Rohrer MJ, Natale AM. Effect of hypothermia on the coagulation cascade. Crit Care Med 1992;20:1402–5. 25. Sessler DI. Complications and treatment of mild hypothermia. Anesthesiology 2001;95:531–43. 26. Danzl DF, Pozos RS. Accidental hypothermia. N Engl J Med 1994;331:1756–60. 27. Reuler JB, Hypothermia:. pathophysiology, clinical settings, and management. Ann Intern Med 1978;89:519–27. 28. Sanchez MM, Binkowitz BS. Guidelines for measurement validation in clinical trial design. J Biopharm Stat 1999;9:417–38. 29. Akata T, Yamaura K, Kandabashi T, et al. Changes in body temperature during profound hypothermic cardiopulmonary bypass in adult patients undergoing aortic arch reconstruction. J Anesth 2004;18:73–81. 30. Akata T, Setoguchi H, Shirozu K, et al. Reliability of temperatures measured at standard monitoring sites as an index of brain temperature during deep hypothermic cardiopulmonary bypass conducted for thoracic aortic reconstruction. J Thorac Cardiovasc Surg 2007;133:1559–65. 31. McCarthy PW, Heusch AI. The vagaries of ear temperature assessment. J Med Eng Technol 2006;30:242–51. 32. Robinson J, Charlton J, Seal R, et al. Oesophageal, rectal, axillary, tympanic and pulmonary artery temperatures during cardiac surgery. Can J Anaesth 1998;45:317–23. 33. Taylor EE, Carroll JP, Lovitt MA, et al. Active intravascular rewarming for hypothermia associated with traumatic injury: early experience with a new technique. Proc (Bayl Univ Med Cent) 2008;21:120–6.