Mild perioperative hypothermia is common and occurs in around half of all surgical patients . It is associated with three major adverse outcomes and specifically increases the risks of (1) cardiac complications [2,3], (2) bleeding  and (3) infectious complications [5,6], especially in high-risk patients (ASA Class III and IV) . Furthermore, hypothermia is also associated with two other concerns: (1) prolonged stay on the post-anaesthesia care unit (PACU)  and (2) thermal discomfort , outcomes which impact patient satisfaction and increase costs .
From the perspective of patient safety, body temperature is, simply stated, a vital sign . Therefore, in order to prevent perioperative hypothermia, core temperature has to be monitored in the first place. Secondly, intraoperative heat loss should be minimized and, thirdly, patients at risk should be actively warmed throughout the surgical procedure .
Despite evidence of the relevant positive effects of preventing perioperative hypothermia in most patient groups, to date no European-wide practice guidelines have been published. Moreover, information detailing European approaches regarding perioperative patient temperature management is limited [1,13,14]. In order to acquire more information on temperature monitoring and patient warming practices in Europe, we conducted this prospective survey on a sample of European anaesthesia departments.
According to the Walnut Medical European Hospital Register , we randomly selected a sample of 801 hospitals representative of the total of around 4000 European hospitals with an anaesthesia department. The number of hospitals to survey in each country was calculated by the ratio of each country's hospital population and the total European hospital population. Additionally, hospitals were a priori classified into three categories: small-sized hospitals (200–400 beds), medium-sized hospitals (400–800 beds) and large-sized hospitals (more than 800 beds). The power analysis to determine the sample size of the study sought to achieve a 90% precision in determining the incidence of temperature monitoring and patient warming in the selected hospitals with an estimated incidence around 20%, and a response rate ranging between 40% and 50% .
In each European country, a member of the Thermoregulation in Europe, Monitoring and Managing Patient Temperature (TEMMP) group was selected as the national delegate of the study group. Each delegate's mission was to liaise with the selected anaesthesia departments in order to provide them with any assistance they might require, and to monitor the adequacy of data collection.
The questionnaire (Appendix 1) was developed in a nominal group process by the TEMMP members. It was agreed to include information on the type of hospital (private or state funded), number of surgical beds and whether there was an anaesthesia training programme. Furthermore, the questionnaire also included questions about the anaesthesia technique performed, the intraoperative patient temperature monitoring and patient warming methods.
A first announcement letter was mailed to the directors of the anaesthesia departments of the selected hospitals explaining the plan of the study and the scheduled survey date. Then, a second mail was sent out 30 days before the survey date containing the questionnaire and instructions for its completion as well as a form requesting survey results once the study was published. To facilitate reply of completed questionnaires, a prepaid self-addressed envelope was enclosed. The questionnaire had to be completed within 2 days and returned to the collection centre within a week after the survey date.
A call centre verified that each selected hospital had received the questionnaire, answered any questions about the survey and confirmed that the hospital would participate. The call centre covered the following languages: English, French, Italian, Greek, Spanish, German and Dutch. Nine days before the survey, a reminder postcard was sent out to each selected hospital containing the survey date.
The survey was conducted on 13 October 2004 in each hospital, in every country. A designated member of each department collected the information according to the questionnaire (Appendix 1).
All data were summarized in a data file (Excel 6.0; Microsoft® Corporation, Redmond, WA 98052-6399, USA) and were stored on a personal computer. Statistical analysis was performed using a statistical software package (Systat® 7.01; Systat Software Inc., Point Richmond, CA 94804-2028, USA). For continuous variables, mean and standard error of the mean, or count and percentage for dichotomous variables were calculated. The t-test or contingency table analysis with Fisher's exact test were used as appropriate. For post hoc analysis, Tukey's test was performed.
Initially, 306 questionnaires were returned (response rate of 38.2%). In order to increase the response, all the anaesthesia departments that did not respond within 2 weeks were contacted again by mail and by phone. By this means, ten additional questionnaires were collected, achieving a final response rate of 39.4%. The response rate varied in the different European countries, ranging from zero (0/2) in Luxembourg to 90% in Switzerland (9/10, Table 1).
Of the 316 responding departments, 159 (50.3%) came from small hospitals, 109 (34.5%) from medium-sized hospitals and 48 (15.2%) from large hospitals. Expectedly, the number of surgical beds reported correlated well with the hospital size as well as with the annual volume of surgical procedures (Table 2).
The survey assessed 8083 surgical procedures performed in the selected hospitals on 1 day. One thousand and seventy-five patients (19.4%) had their body temperature monitored, and 3116 patients were actively warmed (38.5%). A total of 5818 patients (72%) received general anaesthesia, while regional anaesthesia was applied in 2265 procedures (28%). Regional anaesthetic techniques were used between 20% and 30% in most countries; however, this rate was higher in Finland, Italy, The Netherlands, Norway and Spain, namely between 37% and 49% of cases (Fig. 1). When stratifying for the anaesthetic technique, body temperature was monitored in 1449 patients (25%) during general anaesthesia, but in only 126 patients (6%) during regional anaesthesia (P = 0.0005). For body temperature monitoring, the nasopharyngeal cavity was used most often in patients under general anaesthesia, while in patients receiving regional anaesthesia the preferred monitoring site was the tympanic membrane (Fig. 2).
Two thousand four hundred and ninety-one patients (43%) were actively warmed under general anaesthesia and 625 patients (28%) under regional anaesthesia (P = 0.0005). Among active warming measures, forced-air warming was chosen most often, irrespective of the anaesthetic technique (Fig. 3). When considering supportive methods such as fluid warming or passive insulation, patients under general anaesthesia received these in 98% of cases as compared to 79% of regional anaesthesia cases (P = 0.005, Fig. 4).
The majority of the responding hospitals were state-run (77%), however, in Belgium, private and public hospitals were equally distributed. All of the large hospitals had an anaesthesia training programme, whereas the rate was around 40% for medium-sized hospitals and around 20% for small hospitals. In the UK, all of the small-sized hospitals reported having an anaesthesia training programme.
Inadvertent intraoperative hypothermia occurs frequently in anaesthetized patients, irrespective of the anaesthesia technique used [9,17]. Even mild intraoperative hypothermia is known to have a negative impact on patient safety and outcome [3-5]. Hypothermia increases mortality in surgery of ruptured aortic aneurysm  and is a bedside predictor of imminent death in patients with congestive heart failure . Furthermore, specific risk factors for the development of inadvertent perioperative hypothermia have been identified: patients of ASA Classes III and IV , major body cavities exposed over time , longer duration of surgery , high intraoperative blood loss and volume substitution [4,21] and operating room temperature less than 23°C . Not withstanding its importance, little is known about the routine practice of European anaesthetists concerning temperature monitoring and patient warming during general and regional anaesthesia. Although this survey has several limitations (specifically the low response rate, which probably was due to the high clinical work load of the participants or/and their general assumption that hypothermia prevention was already established in their institutions), it is the first Europe-wide study evaluating anaesthesia practice in terms of patients' intraoperative thermal management: Firstly, we could demonstrate that patient body temperature was only rarely monitored intraoperatively, although it has been recognized as a vital sign for long . We may speculate about the reasons; however, probably in some cases measurement was inconvenient or was regarded to be unreliable or there was no simply appropriate equipment available. Secondly, nevertheless, since active patient warming was applied in around 40% of all surgical procedures, we may speculate again that, to some extent, awareness of potential hypothermia harms was present and, consequently, warming of patients was considered important even without monitoring of temperature effects.
The present investigation was powered to detect a frequency of temperature monitoring and patient warming of 20% with a 90% precision. Accordingly, the resulting sample size is small, making it difficult to extrapolate the observed data to the entire population of European hospitals. Nonetheless, if we project the proportion of the final responding hospitals to the entire population, we would produce a number of about 125 000 000 surgical procedures performed in Europe per year, which corresponds with a surgical rate of 17/100 population per year. However, even nationwide, it is difficult to estimate the annually performed surgical procedures accurately. For example, Clergue and colleagues  reported a total number of approximately 8 000 000 anaesthetic procedures performed in 1996 in France resulting in a surgical rate of more than 13/100 population. In another multicentre survey conducted in Italy in 1999, the reported number of surgical procedures performed per year was around 5 000 000 . The differences reported may be partly due to the different study designs; for example in the French survey, all hospitals were evaluated during a 3-day period. In our survey, we considered only hospitals with an anaesthesia department, which potentially overestimated the proportion of surgical procedures per year. Additionally, this effect might have been heightened by the small sample size surveyed.
It is well known that inadvertent hypothermia can easily develop not only during general anaesthesia, but also during regional anaesthesia [25,26] Patients undergoing central neuraxial nerve blocks are often hypothermic when they enter the operating room after the first routine crystalloid infusion . Nonetheless, in the present survey, both temperature monitoring and active patient warming were used significantly less often during regional anaesthesia compared with general anaesthesia. However, our findings are comparable to the temperature monitoring practice during regional anaesthesia reported by US anaesthetists . Even though the majority of anaesthetists felt the prevention of perioperative hypothermia to be important, body temperature was only rarely monitored during regional anaesthesia, significantly increasing the chance that hypothermia was undetected and untreated in these patients.
In the present investigation, we observed a difference in the monitoring site between patients undergoing general or regional anaesthesia. The most frequently used monitoring site during general anaesthesia was the nasopharynx and the tympanic membrane being preferred during regional anaesthesia. Cattaneo and colleagues  evaluated the accuracy of different monitoring sites during general and regional anaesthesia, including tympanic, skin, axillary and rectal temperature monitoring. They reported that the sensitivity and specificity of each method was comparable either with general or spinal anaesthesia techniques. However, in major abdominal surgery, in which neither rectal nor bladder temperature monitoring is reliable, tympanic temperature measurement may provide a closer estimate of core body temperature than equilibrated rectal temperature [29,30].
It is noteworthy that, in many cases, the anaesthetist decided to actively warm the patients even without temperature monitoring. We may speculate that there might be some awareness of the specific effects anaesthesia exerts on patients' heat balance and redistribution of body heat  resulting in hypothermia. Forced-air warming systems were the most common method used for active patient warming. To date, forced-air warming is probably the most effective measure to prevent patient hypothermia [31,32], especially when only a limited skin surface is available . However, in major surgery, a combination with fluid warming may be necessary .
The focus of the survey was to take a snapshot of patients' intraoperative temperature-management practice throughout Europe and since we did not find principal differences related to hospital size (data not shown), we may assume that the apparent lack of attention of anaesthetists to a very simple but crucial monitoring, namely measurement of body temperature, holds true for the whole European hospital population. This is of special concern if we consider that most recent clinical research has been devoted to improve patient outcome, shorten hospital stay and advance fast-track surgery . Preventing inadvertent intraoperative hypothermia is a key point in a multimodal concept to reduce healthcare related costs and optimize patient recovery after elective surgery. Unfortunately, virtually none of the countries surveyed has specific evidence-based recommendations on perioperative temperature management guiding the anaesthetist. The Royal College of Anaesthetists and The Association of Anaesthetists of Great Britain and Ireland simply state that ‘a means of temperature measurement' should be available intraoperatively , while the Italian Society of Anaesthetists published recommendations of a consensus conference held in 1999 to prevent and treat mild perioperative hypothermia . However, to date, the European Society of Anaesthesiology has no guidelines for perioperative temperature monitoring and patient warming. Therefore, we aim to convene a European Consensus Conference, in the near future, which will promulgate the necessary guidelines and recommendations. Finally, an implementation strategy involving anaesthesia training programmes combined with local audit and re-audit of perioperative patient temperature-management practices will be necessary.
We thank Michele Debain (Arizant Healthcare Europe, Paris, France) for her valuable help in organizing and managing the coordination of the survey. The TEMMP Study Group for consensus-assisted development of thermoregulatory guidelines on prevention of inadvertent perioperative hypothermia comprised, in addition to the corresponding author (A Torossian, Marburg, Germany) the following members: Chairman: G Fanelli (Parma, Italy), R Alexander (Worcester, UK), P Alfonsi (Boulogne Billancourt, France), JM Campos (Barcelona, Spain), A Casati (Parma, Italy), L Foubert (Aalst, Belgium), A Kurz (Bern, Switzerland), L Lindgren (Tampere, Finland), R Lindwall (Stockholm, Sweden), M-T Mäkinen (Helsinki, Finland), E Scott (Durham, UK), W Weyland (Essen, Germany), P Wouters (Leuven, Belgium), H Wulf (Marburg, Germany) and JM Zaballos (San Sebastian, Spain).
Funding: Arizant Healthcare Europe generously sponsored the TEMMP group meetings.
1. Abelha FJ, Castro MA, Neves AM, Landeiro NM, Santos CC. Hypothermia
in a surgical intensive care unit. BMC Anesthesiol
2. Frank SM, Beattie C, Christopherson R et al
. Unintentional hypothermia
is associated with postoperative myocardial ischemia. The Perioperative Ischemia Randomized Anesthesia Trial Study Group. Anesthesiology
3. Frank SM, Fleisher LA, Breslow MJ et al
. Perioperative maintenance of normothermia reduces the incidence of morbid cardiac events. A randomized clinical trial. JAMA
4. Schmied H, Kurz A, Sessler DI, Kozek S, Reiter A. Mild hypothermia
increases blood loss and transfusion requirements during total hip arthroplasty. Lancet
5. Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. Study of Wound Infection and Temperature Group. N Engl J Med
6. Torossian A, Ruehlmann S, Middeke M et al
. Mild preseptic hypothermia
is detrimental in rats. Crit Care Med
7. Kongsayreepong S, Chaibundit C, Chadpaibool J et al
. Predictor of core hypothermia
and the surgical intensive care unit. Anesth Analg
8. Panagiotis K, Maria P, Argiri P, Panagiotis S. Is postanesthesia care unit length of stay increased in hypothermic patients? AORN J
9. Doufas AG. Consequences of inadvertent perioperative hypothermia
. Best Pract Res Clin Anaesthesiol
10. Mahoney CB, Odom J. Maintaining intraoperative normothermia: a meta-analysis of outcomes with costs. AANA J
11. Smith JJ, Bland SA, Mullett S. Temperature – the forgotten vital sign. Accid Emerg Nurs
12. Sessler DI. Complications and treatment of mild hypothermia
13. Gallagher GA, McLintock T, Booth MG. Closing the audit loop – prevention of perioperative hypothermia
: audit and reaudit of perioperative hypothermia
. Eur J Anaesthesiol
14. Kean M. A patient temperature audit within a theatre recovery unit. Br J Nurs
15. Crosby K, Tulloch B, Crosby R. Walnut Medical Hospital Register. 2006. http://www.walnutmedical.info
16. Browner WS, Newman TB. Sample size and power based on the population attributable fraction. Am J Public Health
17. Leslie K, Sessler DI. Perioperative hypothermia
in the high-risk surgical patient. Best Pract Res Clin Anaesthesiol
18. Janczyk RJ, Howells GA, Bair HA, Huang R, Bendick PJ, Zelenock GB. Hypothermia
is an independent predictor of mortality in ruptured abdominal aortic aneurysms. Vasc Endovascular Surg
19. Casscells W, Vasseghi MF, Siadaty MS et al
is a bedside predictor of imminent death in patients with congestive heart failure. Am Heart J
20. Bush Jr HL, Hydo LJ, Fischer E, Fantini GA, Silane MF, Barie PS. Hypothermia
during elective abdominal aortic aneurysm repair: the high price of avoidable morbidity. J Vasc Surg
21. Guest JD, Vanni S, Silbert L. Mild hypothermia
, blood loss and complications in elective spinal surgery. Spine J
22. El Gamal N, El Kassabany N, Frank SM et al
. Age-related thermoregulatory differences in a warm operating room environment (approximately 26 degrees C). Anesth Analg
23. Clergue F, Auroy Y, Pequignot F, Jougla E, Lienhart A, Laxenaire MC. French survey of anesthesia in 1996. Anesthesiology
24. Peduto VA, Chevallier P, Casati A. A multicenter survey on anaesthesia practice in Italy. Minerva Anestesiol
25. Arkilic CF, Akca O, Taguchi A, Sessler DI, Kurz A. Temperature monitoring and management during neuraxial anesthesia: an observational study. Anesth Analg
26. Cattaneo CG, Frank SM, Hesel TW, El Rahmany HK, Kim LJ, Tran KM. The accuracy and precision of body temperature monitoring methods during regional and general anesthesia. Anesth Analg
27. Frank SM, El Rahmany HK, Cattaneo CG, Barnes RA. Predictors of hypothermia
during spinal anesthesia. Anesthesiology
28. Frank SM, Nguyen JM, Garcia CM, Barnes RA. Temperature monitoring practices during regional anesthesia. Anesth Analg
29. Kamada Y, Miyamoto N, Yamakage M, Tsujiguchi N, Namiki A. Utility of an infrared ear thermometer as an intraoperative core temperature monitor. Masui
30. Rotello LC, Crawford L, Terndrup TE. Comparison of infrared ear thermometer derived and equilibrated rectal temperatures in estimating pulmonary artery temperatures. Crit Care Med
31. Brauer A, Weyland W, Kazmaier S et al
. Efficacy of postoperative rewarming after cardiac surgery. Ann Thorac Cardiovasc Surg
32. Kabbara A, Goldlust SA, Smith CE, Hagen JF, Pinchak AC. Randomized prospective comparison of forced air warming using hospital blankets versus commercial blankets in surgical patients. Anesthesiology
33. Casati A, Baroncini S, Pattono R et al
. Effects of sympathetic blockade on the efficiency of forced-air warming during combined spinal-epidural anesthesia for total hip arthroplasty. J Clin Anesth
34. Zhao J, Luo AL, Xu L, Huang YG. Forced-air warming and fluid warming minimize core hypothermia
during abdominal surgery. Chin Med Sci J
35. Kehlet H, Wilmore DW. Fast-track surgery. Br J Surg
36. The Association of Anaesthetists of Great Britain and Ireland. Recommendations for standards of monitoring during anaesthesia and recovery. 2000. http://www.aagbi.org/PDF/Absolute.pdf
37. Montanini S, Martinelli G, Torri G et al
. Recommendations on perioperative normothermia. Working Group on Perioperative Hypothermia
, Italian Society for Anesthesia, Analgesia, Resuscitation, and Intensive Care. Minerva Anestesiol
Appendix 1: The questionnaire sent to the selected anaesthesia departments