|
|
||||||||
| Abstract |
|---|
|
|
|---|
Although body temperature can be measured by using a variety of sites and devices, continuous measurement of core temperature, particularly on a long-term basis, has been problematic. Indwelling thermistor-tipped catheters or probes, such as those flow directed into the pulmonary artery, have been used mainly in intensive care units (ICUs), but only for selected patients who require hemodynamic monitoring. Esophageal probes have been used mainly in the operating room, but esophageal temperature is rarely monitored in critical care areas, and placement of the probe varies. Last, rectal probes have been used, mostly in the emergency department, for continuous monitoring of hypothermic or hyperthermic patients. The urinary bladder has become a more common and more widespread site for continuous monitoring of body temperature, particularly for patients who also require an indwelling catheter for urine drainage.
A database search revealed that no systematic review of the literature on monitoring of urinary bladder temperature had been published. In this article, I provide a comprehensive and critical review of research undertaken in ICUs to compare body temperatures measured in the urinary bladder with temperatures measured at a core site, the pulmonary artery. The ICU is an ideal clinical area for this comparison. Many ICU patients require a pulmonary artery catheter for hemodynamic monitoring and an indwelling catheter for urine drainage. In critically ill patients, insertion of a urinary catheter is considered justified,1 and continued catheterization is justified when close monitoring of urine output is required.2 Further, critically ill patients often provide a wide range of temperature measurements, particularly immediately after open heart surgery.3,4
| Criteria for Selection of Articles |
|---|
|
|
|---|
Temperature-sensing bladder catheters were initially described in the literature approximately 2 decades ago, so only publications from 1980 and later were sought. Various databases were used to locate relevant published and unpublished studies. These included MEDLINE, Current Contents, Cumulative Index for Nursing and Allied Health (CINAHL), and Pro Quest Digital Dissertation Abstracts. The reference sections of relevant articles provided additional sources. Key words used in the search were bladder temperature, pulmonary artery temperature, body temperature, intensive care, and critical care. In total, 8 studies met the just-described criteria, and all were included in the review.
| Background |
|---|
|
|
|---|
| Urinary Bladder Temperature: Advantages and Disadvantages |
|---|
|
|
|---|
Although bladder temperature shows inertia during times of dynamic thermal changes, the bladder has distinct advantages and uses. Measurement of bladder temperature is convenient, reliable, and safe1315 and is a good estimate of core temperature.16 Measuring bladder temperature is useful during intensive care management9 and as a substitute for measuring pulmonary artery temperature.17 Because of the frequency of hypothermia in the operating room during anesthesia and surgery, measurement of bladder temperature is also useful as an indicator of rewarming, particularly in cardiac surgery patients.18
A distinct advantage of using the bladder for temperature measurement, when an indwelling catheter has been inserted for continuous urinary drainage, is that doing so causes no additional discomfort for the patient.15 Insertion of a temperature-sensitive indwelling catheter eliminates the need to use other sites for temperature measurement and allows both an exact measurement of urine output and a continuous measurement of body temperature.7 Measurement of bladder temperature is particularly useful in critically ill patients such as those with extensive burns19 or severe head injuries and in patients who have undergone neurosurgery or surgery for head and neck cancer.
Disadvantages include the need to use a special catheter and monitoring system, which often are not available outside the ICU or operating room.16,17The additional cost of these special catheters18 also is a drawback. The decreased need for alternative temperature-taking methods and the nursing time saved may outweigh this cost; however, these factors have not been investigated.
| Bladder Temperature Compared With Pulmonary Artery Temperature |
|---|
|
|
|---|
| Bladder Temperatures in the ICU |
|---|
|
|
|---|
|
In another study, Moorthy et al9 evaluated bladder temperature as an indicator of core body temperature. The number of data collectors was not reported. Using repeated-measures analysis of variance, Moorthy et al noted a nonsignificant difference between mean pulmonary artery temperature and bladder temperature. Although they concluded that bladder temperature was representative of core temperature in the steady state, they did not report the range of patients body temperatures or, more importantly, the range of differences between the pulmonary artery and bladder temperatures. A larger sample size and the testing of catheters after their removal from patients would have enhanced the validity of the study findings.
Using pulmonary artery temperature as a reference, Mravinac et al25 compared temperatures at that site with bladder and rectal temperatures in hypothermic patients after cardiac surgery. Multiple data collectors recorded the temperatures. Repeated-measures analysis of variance indicated a significant difference between temperatures measured at the 3 sites over time (P<.001). The higher correlation at each time point between pulmonary artery and bladder temperatures (r = 0.78 to 0.94, P<.01) than between pulmonary artery and rectal temperatures (r = 0.49 to 0.82, P<.01) led to the conclusion that bladder temperature provided a better representation of core temperature than did rectal temperature. Further, bladder and pulmonary artery temperatures varied similarly. The sample size is a strong point of this study. Additional statistical testing to determine which temperature site in particular differed from the pulmonary artery site would have been beneficial. Whether patients were shivering when cool was not reported. Shivering, if it was present, may have accounted for the change in the relationship between pulmonary artery and bladder temperatures observed once normothermia was reached.23 Omission of tests of interrater reliability and catheter accuracy are limitations of this study.
Nierman7 compared pulmonary artery and bladder temperatures in patients admitted to a medical ICU. Using Bland and Altmans27 technique for assessing agreement between methods of measurement, he found good agreement between temperatures measured at these 2 sites throughout the monitoring period. Nierman concluded that pulmonary artery and bladder thermometer catheters provided consistent and highly reliable measures of core temperature in ICU patients. A larger sample size and measures of interrater reliability would have enhanced the validity of the findings.
The preliminary and completed findings of a study of the relationship between pulmonary artery and bladder temperatures during the rewarming phase in patients in the ICU after hypothermic cardiac surgery were reported in 2 separate publications.23,26 In the preliminary study by Earp and Finlayson,26 data were collected by one data collector, and participants hemodynamic status was stable. Mean bladder temperatures were generally higher than pulmonary artery temperatures except for 3 measurement times when bladder temperature was lower by 0.03°C to 0.04°C. Significant differences (P<.05) occurred at 11 time points; 24 of the 25 measures had differences of 0.17°C or less. The authors found a strong positive relationship, as indicated by correlation coefficients of 0.94 to 0.99, between the measurements at these 2 sites. Both pulmonary artery and urinary bladder temperatures were recommended as effective measures for evaluating body temperature during rewarming of patients after cardiac surgery, and it was suggested that both be used in order to provide a better index of core temperature. Unfortunately, the omission of catheter testing after removal makes it impossible to determine whether the differences in temperature detected in the study are true differences or are due to catheter error.
In the final results presented by Earp and Finlayson,23 the thermal gradient between pulmonary artery and urinary bladder temperatures in shivering and nonshivering patients was assessed. Shivering was recorded by using both an electromyogram and a 4-point observational scale. Interestingly, for the non-shivering group (24%), bladder temperature generally stayed higher than pulmonary artery temperature. This relationship was reversed in the shivering group (76%), resulting in pulmonary artery temperatures that were significantly higher than bladder temperatures (P<.03). Shivering occurred from 75 minutes to 5 hours after admission, with between 1 and 13 patients shivering at each measurement time. The authors concluded that urinary bladder temperature is a reliable index of core temperature for hypothermic patients and suggested that it also could serve as an adjunct to pulmonary artery temperature. Unfortunately, catheters were not tested for accuracy after removal; rather, the investigators relied on manufacturers reports. Further, although 1 of the studys purposes was to determine whether urinary bladder and pulmonary artery temperature gradients were related to shivering, the ratio was reported, but the gradients were not reported. Moreover, it is unclear which shivering scale was used in the analysis of the shivering data. The use of a single data collector in this and the authors earlier study decreases the variation in measurement and procedure that may occur with multiple data collectors.
Bladder temperature generally corresponded well to reference pulmonary artery temperature in the 2 final studies16,17 reviewed. Using 2 experienced critical care nurses as data collectors, Erickson and Kirklin16 compared temperatures measured at 4 routine sites with pulmonary artery temperature. When compared with mean pulmonary artery temperature, bladder temperature was significantly higher (P<.01). Much like Mravinac et al,25 Erickson and Kirklin16 noted that when patients were normothermic (>37°C), mean bladder temperature exceeded pulmonary artery temperature. However, this relationship was reversed when patients were cool (<36.5°C), with mean bladder temperature lower than pulmonary artery temperature. The presence or absence of shivering was not reported. The possible effect of time on measurements of bladder and pulmonary artery temperatures was not examined.
Erickson and Meyer,17 in a complex study using a cross-sectional repeated-measures design, compared multiple ear-based temperatures with temperatures measured at various sites, including the pulmonary artery and the bladder. Three critical care nurses collected data. Bladder temperature was nearly identical to pulmonary artery temperature at each measurement time, and high correlations (r = 0.99) were noted. Patients in the coolest range of temperature (<36°C) had pulmonary artery temperatures that were higher than bladder temperatures. In comparison, patients with temperatures between 36.0°C and 37.9°C had pulmonary artery temperatures that were lower than bladder temperature (P<.01). For patients in the 38.0°C to 38.8°C range, temperatures measured at the 2 sites did not differ. The prevalence of shivering was not reported. Although randomized order of measurement was used to eliminate order effect, the multiple sequences possible when 10 measurements are taken but only 50 subjects are used does not provide equal representation to all the orders. Testing of catheters after removal and reporting of interrater reliability would have enhanced the validity of the findings in both this study17 and the study undertaken earlier by Erickson and Kirklin.16
| Discussion |
|---|
|
|
|---|
In general, the instruments (catheters and monitors) used to measure pulmonary artery and bladder temperatures were not well described. In 2 studies,23,26 the manufacturers reports of the instruments accuracy was relied upon, and although this practice is common in the clinical area, it is generally not acceptable in research. In order to determine if the temperature differences between measurement sites reflect true differences or are due to error, at a minimum, information about the instruments accuracy and resolution, the smallest increment in a quantity that can be measured with certainty, is required. It is not feasible to pretest disposable pulmonary artery and bladder catheters, which must be sterile when inserted. However, accuracy of catheters can be assessed after they are removed. In only 1 study24 were catheters tested after removal. When instrument validity is not assessed, the validity of the studys findings is open to question.
Although sites that both accurately reflect core temperature and have minimal variability are the ideal, Erickson and Kirklin16 note that some variability is to be expected with clinical measurements. This variability may be due to the patient or the environment, or it may be a result of factors related to the measurement technique. Further, the accuracy of temperature measurements does not have to be absolute but must be sufficient for detection of clinically important changes in core temperature.16 Although empirical studies to determine what is a clinically significant change in temperature have not been done, temperature differences between ±0.2°C and ±0.5°C have been suggested.3,28,29 With temperature differences of that size, the mean differences (including the SDs) between pulmonary artery and urinary bladder temperatures reported in all 8 studies, except for the second study by Earp and Finlayson,23 would not be considered clinically significant. The second study by Earp and Finlayson23 was the only study in which shivering of patients was reported.
Overall, bladder temperature was similar to or slightly higher than pulmonary artery temperature when patients were normothermic. This relationship, however, generally was reversed when patients were shivering or cool, resulting in pulmonary artery temperatures that exceeded bladder temperatures. The most likely explanation of this disparity is that shivering and its subsequent production of thermal energy11 resulted in a transient rapid increase in core temperature that had yet to be reflected in the urinary bladder temperature. Shivering in postoperative patients can produce dramatic increases in oxygen consumption, from 65% to 220%.30
Only 1 study23 was specifically designed to assess shivering. However, shivering may have occurred in other studies16,17,25,26 as well, in which the sample or subgroups of the sample were reported to be cool. Shivering would account for the change in relationship between pulmonary artery and bladder temperature that occurred when temperatures increased and neared normothermia. A lag of bladder temperature behind pulmonary artery temperature in cardiac surgery patients during rapidly induced temperature changes has been reported810 and may have occurred in these studies16,17,25,26 when pulmonary artery temperature was increasing quickly because of the thermal energy produced by shivering. Shivering also may have accounted for the range of differences between pulmonary artery and bladder temperatures noted by Erickson and Kirklin,16 Erickson and Meyer,17 and Earp and Finlayson.23 The smaller gradient noted in Earp and Finlaysons earlier study26 may have been due to the smaller sample size in that study.
In the studies reviewed, urinary flow rate was not reported. The gradient between bladder and pulmonary artery temperatures may vary with rates of urine flow. Horrow and Rosenberg,8 in a study of patients undergoing cardiac surgery, found that temperature differences between the 2 sites were significantly greater when the urine flow rate was low than when the urine flow rate was high. The occurrence of this phenomenon has not been investigated in ICU patients.
Last, although differences between bladder and core site measurements of body temperature and the variability associated with these measurements are important, the degree of association between pulmonary artery and bladder temperatures also is noteworthy. Moderate to high correlations of 0.78 to 0.99 between pulmonary artery and urinary bladder temperatures were found in several studies.7,17,23,25,26 Correlation coefficients, however, do not provide information about the gradient or temperature difference between variables, but only about the strength and direction of the association. Further, the association depends on the range of measurements. The wider the range of the true quantity in the sample, the greater will be the correlation coefficient.27
| Conclusion |
|---|
|
|
|---|
Development of a temperature-sensing indwelling catheter has enabled measurement of temperature in the urinary bladder.23 In critical care areas, measuring bladder temperature has become a common method of estimating core temperature.17 Results of the studies reviewed here support the use of bladder temperature as a reliable index of core temperature in patients who are normothermic (>36.5°C) and in patients with elevated body temperatures (up to 38.8°C). Further, although the relationship between pulmonary artery and urinary bladder temperature reversed when patients were cool (<36.5°C), mean temperature differences between these 2 sites remained clinically insignificant.16,17,23,25
Because many ICU patients have indwelling bladder catheters to assist in management of their acute illness,7 the ability to use this device to measure both urine output and continuous body temperature is a distinct advantage. For patients in the ICU who are already under considerable stress and do not have a pulmonary artery catheter in place, using a temperature-sensing bladder catheter eliminates the need to measure temperature at other sites, reducing disturbance of the patient and freeing nursing time as well. Therefore, use of this device for continuous measurement of temperature or for assessing a patients response to treatment is suggested for ICU patients who require an indwelling urinary catheter.
Further research is needed to determine what roles shivering and urine flow rate play in the accuracy of measuring urinary bladder temperature in the ICU. Investigation into the cost-effectiveness of temperature-sensing bladder catheters also is required.
| ACKNOWLEDGMENTS |
|---|
To purchase reprints, contact The InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 809-2273 or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail, reprints{at}aacn.org.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. Kimberger, D. Cohen, U. Illievich, and R. Lenhardt Temporal Artery Versus Bladder Thermometry During Perioperative and Intensive Care Unit Monitoring Anesth. Analg., October 1, 2007; 105(4): 1042 - 1047. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. B. Hunter and M. S. Taljanovic Medical Devices of the Abdomen and Pelvis RadioGraphics, March 1, 2005; 25(2): 503 - 523. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |