| ||||||||||||||
|
|
|||||||||||||

*Department of Anesthesia, Stanford University, Stanford, California; and
Division of Management Consulting, Department of Anesthesia, University of Iowa, Iowa City, Iowa
Address correspondence and reprint requests to Alex Macario, MD, MBA, Department of Anesthesia, Stanford University School of Medicine, Stanford, CA 94305-5640. Address e-mail to amaca{at}stanford.edu
| Abstract |
|---|
|
|
|---|
IMPLICATIONS: Surveys of clinicians and physician researchers identified what they consider to be the most important risk factors for perioperative hypothermia (e.g., neonates, a low ambient operating room temperature, burn patients, and general anesthesia with neuraxial anesthesia).
| Introduction |
|---|
|
|
|---|
The rapid decrease in core temperature in the first hour after the induction of general anesthesia is caused by redistribution of heat within the body away from the core and toward the peripheral thermal compartment (9). However, currently, there are no algorithms (or clinical risk-stratifying tools) to help anesthesiologists determine the appropriate heat-conserving measures to use on a particular patient. For example, should a patient scheduled to undergo a hysterectomy receive warmed IV fluids prophylactically? Should a patient undergoing hip replacement surgery receive forced air warming starting at the induction of general anesthesia, or should a cotton blanket be placed on the patient initially (10)?
As a first step in developing a clinical risk-stratifying tool to help determine appropriate heat-conserving measures, we aimed to prioritize risk factors for a patient to develop intraoperative hypothermia. For example, is the size of the incision more or less predictive of whether the patient will develop intraoperative hypothermia than the temperature of the operating room (OR) during surgery?
We hypothesized, on the basis of review of the literature, that the most important risk factors for a patients developing intraoperative hypothermia were the thermal status of the patient before surgery (11,12), the size and age of the patient (13,14) (e.g., a newborn versus an obese adult), the ambient OR temperature (15), the size of the surgical incision (16,17), and the presence of neuropathy (18). The goal of this study was to prioritize risk factors for a patient developing intraoperative hypothermia during routine surgical cases, by surveying clinical practitioners. We also surveyed physician researchers who have published studies on temperature regulation to determine whether there were any differences in their expectations compared with the rankings provided by surveyed clinicians without published research in temperature regulation. The risk factors specifically identified as important can then be studied in clinical trials to quantify the relative importance of each risk factor and develop a multivariate tool to predict a patients risk for developing hypothermia.
| Methods |
|---|
|
|
|---|
The contents of all 81 abstracts and full-text articles obtained from our literature search were read in full (a sample of these studies is included in References) to identify validated factors associated with patient temperature. In addition, a manual search, which was performed by screening citation lists in review articles, yielded another 48 articles.
The validity of a survey instrument may be defined as the extent to which differences in observed scale scores reflect true differences among objects on the characteristic being measured, rather than systematic or random error. To assess content validity, sometimes called face validity, the researcher examines whether the scale items adequately cover the entire domain of the construct being measured. Because the comprehensive list of 41 risk factors in our survey instrument associated with the development of intraoperative hypothermia was constructed from the articles obtained from the literature search, we believe that the questionnaire has content validity.
During a pretest phase, this list was examined by four senior, board-certified anesthesiologists in the Department of Anesthesiology, Stanford University, as well as by a convenience sample of five local community practitioners. This pretest phase helped to ensure that the questions were properly interpreted, that significant risk factors were not excluded, and that the survey could be completed within a short period of time (i.e., 1012 min). The final complete survey used for this study is available from the authors.
A sample of the text on the first page of the questionnaire reads as follows. "The purpose of this study is to quantify the relative importance of risk factors for a patient developing hypothermia (core T < 36.0) in the OR, assuming no temperature preserving/increasing interventions are used. Some of the factors listed may be inter-related. For now assume that they are independent."
Although the exact definition of intraoperative normothermia has been the subject of debate, we choose to use <36°C as the definition of hypothermia (21). Normal body temperature averages 37°C with a normal circadian body temperature variation, equaling approximately 1°C (22,23).
We asked respondents to "estimate the relative importance (which risk factor is more important compared to others) on a scale of 110, with a 1 corresponding to not likely to be important in causing hypothermia and a 10 corresponding to most likely to be important in causing hypothermia. The more points the more important."
The survey was mailed to two groups: 1) 180 anesthesiologists randomly selected from the 1999 American Society of Anesthesiologists Members Directory and 2) 24 physician researchers in thermoregulation. The researchers were selected from the list of authors with "MD" after their name and one or more studies identified during the literature search described previously. A second mailing was performed for those subjects who did not respond to the first mailing.
"Reliability" refers to the extent to which a scale produces consistent results if repeated measurements are made (24). Systematic sources of error do not have an adverse effect on reliability because they affect the measurement in a constant way and do not lead to inconsistency. In contrast, random error produces inconsistency, leading to lower reliability. We used the test-retest approach for assessing reliability. In test-retest reliability, respondents are administered identical sets of scale items at two different times under as nearly equivalent conditions as possible. We asked 17 of the respondents to complete an identical questionnaire a second time, at least 3 days after the original questionnaire.
Because the data were not normally distributed, the results are presented on the basis of medians (quartile deviation). Confidence intervals for the shifts in the two groups statistical distributions of responses for each item were calculated by using the Hodges-Lehmann method (StatXact-4, Cytel Software Corp., Cambridge, MA).
| Results |
|---|
|
|
|---|
|
Researchers and clinicians rated the following to be the most important risk factors for hypothermia (in sequence): neonates, a low ambient OR temperature, burn patients, general anesthesia with neuraxial anesthesia, geriatric patients, low temperature of the patient before induction, a thin body type, and large blood loss (Table 2).
|
| Discussion |
|---|
|
|
|---|
The risk factors identified in this study could be used to establish clinical studies to quantify the relative importance of each risk factor. These clinical studies are continuing. As an example, in a study of only patients receiving spinal anesthesia, a high level of spinal blockade and increasing age were predictors of hypothermia (25). It is interesting that for this group receiving spinal anesthesia, the duration of surgery, the ambient OR temperature, and body habitus were not predictors of hypothermia.
A next step would be to quantify how much more or less important one risk factor is than another in a wider variety of cases. Then, a multivariate predictive tool could be developed to calculate a patients a priori risk for developing hypothermia, if preventive measures are not taken. Such tools have been developed to assess the risk of death from heart failure after coronary artery bypass graft surgery, for example (26). Finally, recommendations ranging in cost and efficacy for appropriate hypothermia prophylaxis can be provided to practitioners (e.g., via a pocket-sized card or an automated anesthesia information system). However, whether this type of clinical decision support improves perioperative outcomes also requires rigorous study.
The rankings presented here are based on researchers interpretation of studies findings. The fact that clinicians ratings were the same as those by researchers points to the validity of the studys findings, clinicians reading the results of scientific studies, clinical experience helping predict scientific findings, or a combination of these. That clinical experience has predicted scientific findings is likely to hold because temperature can be measured easily and routinely in patients undergoing general anesthesia.
Intraoperative core hypothermia develops in three phases: core-to-peripheral redistribution of body heat, a slow linear decrease in core temperature caused by heat loss exceeding metabolic heat production, and core temperature plateau as vasoconstriction constrains heat in the core compartment. The highest rated item was neonates. The head constitutes a larger fraction of the total surface area in neonates. Heat loss from the head by neonates may be greater than in adults because the skull and scalp are thin, allowing loss of heat delivered to the brain (27). Redistribution of body heat that is most prominent in the first hour of anesthesia contributes less to hypothermia in infants (than in adults) because their extremities are small compared with their trunk and head (also resulting in a large surface area to body weight ratio) (28).
The second highest rated item was a low ambient OR temperature. At OR temperatures <21°C, hypothermia often occurs, especially during general anesthesia (29). At 23°C, hypothermia occurs less often. In a study of 40 patients receiving general anesthesia for orthopedic surgery, an ambient OR temperature near 26°C (79°F) was effective in preventing core hypothermia during general anesthesia regardless of patient age (30).
Another highly rated item was general anesthesia with neuraxial anesthesia. Because both general and regional anesthetics impair the thermoregulatory system, anesthetized patients become relatively poikilothermic, whereby core temperature drifts toward ambient temperature (31). Patients receiving epidural (32,33) or spinal anesthesia (34,35) have impaired thermoregulation because of vasodilation in the lower extremities and an impaired hypothalamic response with decreased shivering and vasoconstriction thresholds (36,37). This effect is proportional to the number of dermatomes blocked (38). General anesthesia inhibits the shivering that might otherwise increase heat production during neuraxial anesthesia.
Three other risk factors deemed important by both researchers and clinicians were patient age (i.e., geriatric patients), low temperature of the patient before induction, and a thin body habitus. Geriatric patients thermoregulate via vasoconstriction and shivering less effectively than younger patients, both in the presence and absence of anesthesia (39,40). Patient temperature before induction refers to peripheral tissue heat content and core-to-peripheral tissue temperature gradient, which determines the magnitude of redistribution. The effects of body morphology have also been reasonably well evaluated. In 40 patients undergoing elective colon surgery, the amount of redistribution hypothermia was inversely proportional to the percentage of body fat and the ratio of weight to surface area (41).
There are several potential limitations to the study. As with any survey-based study, the results may be affected by a variety of biases. The overall response rate was 51%. We did not detect differences in responses from the group that returned the survey after the first mailing as compared with the responses obtained after the second mailing. In addition, respondents were asked to list other risk factors that should be included that were not on the original questionnaire. No one risk factor was suggested by more than two respondents. Thus, it seems that our study did not miss any important risk factors. The generalizability of these rankings to all North American anesthesia providers is supported because the subject population was composed of practitioners from 21 states in the United States. However, it is possible that the opinion in other parts of the world could differ.
We conclude that neonates, a low ambient OR temperature, burn patients, general anesthesia with neuraxial anesthesia, geriatric patients, low patient temperature before induction, a thin body type, and large blood loss (in that order) were rated as most likely to be important in causing hypothermia. No significant differences were detected in responses between researchers and clinicians.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. Silvay, J. G. Castillo, J. Chikwe, B. Flynn, and F. Filsoufi Cardiac Anesthesia and Surgery in Geriatric Patients Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2008; 12(1): 18 - 28. [Abstract] [PDF] |
||||
![]() |
Y. F. Bryan, T. W. Templeton, T. G. Nick, M. Szafran, and A. Tung Brain magnetic resonance imaging increases core body temperature in sedated children. Anesth. Analg., June 1, 2006; 102(6): 1674 - 1679. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Y. Kim, H. Shinn, Y. J. Oh, Y. W. Hong, H. J. Kwak, and Y. L. Kwak The effect of skin surface warming during anesthesia preparation on preventing redistribution hypothermia in the early operative period of off-pump coronary artery bypass surgery Eur. J. Cardiothorac. Surg., March 1, 2006; 29(3): 343 - 347. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. Horowitz, M. A. Delagarza, J. J. Pulaski, and R. A. Smith Flow Rates and Warming Efficacy with Hotline and Ranger Blood/Fluid Warmers Anesth. Analg., September 1, 2004; 99(3): 788 - 792. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. L. Joris, N. J. Poth, A. M. Djamadar, D. I. Sessler, E. E. Hamoir, T. R. Defechereux, M. R. Meurisse, and M. L. Lamy Supplemental oxygen does not reduce postoperative nausea and vomiting after thyroidectomy{dagger} Br. J. Anaesth., December 1, 2003; 91(6): 857 - 861. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|