Assignment: Quantitative Research Analysis Web Page

www.aana.com/aanajournalonline AANA Journal February 2018 Vol. 86, No. 1 19

Temperature monitoring is a standard of anesthesia care as listed in Standard V of the American Asso- ciation of Nurse Anesthetists Standards of Nurse Anesthesia Practice. The purpose of this quantitative correlational study was to examine which temperature modality (tympanic vs temporal) best correlates with pediatric surgical patients’ core rectal temperature. Data were from a sample of 106 intraoperative pedi- atric surgical patients with ASA physical classifica- tion 1 or 2 who were scheduled for elective surgical procedures.

Findings from this study support that tympanic tem- perature correlates more to core rectal temperature both before (Pearson r = 0.36 vs 0.16) and after surgery (Pearson r = 0.57 vs 0.33) and had less bias with core

rectal temperature (r = 0.37 vs 0.55) than temporal temperature. Multiple regression analyses further sup- ported tympanic temperature as the best predictor of core rectal temperature both before surgery (R2 = 0.17, R2adj = 0.13, F(5, 100) = 4.18, P = .0007) and after sur- gery (R2 = 0.34, R2adj =0 .30, F(7, 99) = 7.47, P = .001). Although generalizations are limited beyond this study population, the findings add support to recommend tympanic temperature as the temperature modality of choice in the pediatric surgical population.

Keywords: Core temperature, pediatric surgical patients, temperature, temporal temperature, tym- panic temperature.

Relationship of Tympanic and Temporal

Temperature Modalities to Core Temperature

in Pediatric Surgical Patients

Debra J. Minzola, PhD, MSN, CRNA

Rebecca Keele, PhD, PHCNS-BC

T emperature is one of the oldest monitored vital signs used to measure human physiology and the body’s response to stress and illness. One responsibility of Certified Registered Nurse Anesthetists (CRNAs) in the operating room

(OR) is patient temperature monitoring. The historical timeline of thermometers began in the 16th century by Galileo, but the method changed in 1709 when Daniel Fahrenheit developed an alcohol-filled thermometer and a mercury-filled thermometer.1,2 William Squire, a medi- cal student in London in 1846, was the first to use these thermometers to monitor his patients under anesthesia.1 Squire concluded through his seminal research that activity increases body temperature and rest lowers body temperature.3

Developments in temperature monitoring and anes- thesia during the 1800s and early 1900s made it possible to observe the effects of anesthesia on patients’ body temperature.1,2,4 In 1868, Carl Wunderlich developed the clinical thermometer used in medicine for the next 130 years.2 Throughout the 1900s, an ongoing rise in morbidity and mortality occurred in surgical patients. Hypothermia led to coagulopathy, surgical wound infec- tions, delayed recovery, and death in surgical patients.1,2,4 The hazards of hypothermia and hyperthermia led to rec- ognizing the importance of routine temperature monitor- ing and early identification of temperature extremes in

the OR.5-7 The development of the thermometers in use in the 21st century occurred in the 1800s by measuring the heat conducted from mucous membranes or skin by direct contact.8 Seminal research indicated concerns regarding the imbalance of patient thermoregulation due to clinical conditions in the OR.

One measure that may reduce the risk of surgical com- plications is temperature monitoring and accurate main- tenance of normothermia in surgical patients. General an- esthesia impairs thermoregulation; both inhalation agents and nitrous oxide inhibit thermoregulatory vasoconstric- tion in the surgical patient. This measure signifies the importance of the study in monitoring pediatric patients’ intraoperative temperature with the best temperature modality correlating with the pediatric patients’ core temperature as they transition through the surgical suite. The Surgical Care Improvement Project (SCIP), formed in 2003, consists of numerous nationally recognized organi- zations: the American Hospital Association, the Agency for Healthcare Research and Quality, the Centers for Disease Control and Prevention, the Institute for Healthcare Improvement, the Centers for Medicare and Medicaid Services (CMS), the Joint Commission, and the Veterans Health Administration.9 The primary objective of the SCIP is to reduce the risk of surgical complications.9

The goal of this national campaign and partnership is a 25% reduction in surgical complications. The SCIP

20 AANA Journal February 2018 Vol. 86, No. 1 www.aana.com/aanajournalonline

is an extension of CMS, which funded the previous initiative: the Surgical Infection Prevention Project.9 To receive 100% of the annual payment, updated by CMS, hospital leaders must participate in the Medicare Hospital Inpatient Quality Reporting Program.10

Previous studies have focused on obtaining con- tinuous recordings of adult patients’ temperatures using esophageal and tympanic temperature monitors while patients were under anesthesia.11 This emphasis on the importance of temperature measurement intraoperatively supports the need for the current study with its focus on the pediatric surgical population, which is more vulnera- ble than adults.11 Pediatric patients are a population with an increased risk of complications related to extremes of temperature intraoperatively because of their proportion- ately large body surface area.12-14 These intraoperative and postoperative complications include coagulopa- thy, surgical wound infections, delayed recovery time, prolonged neuromuscular blockade, increased surgical wound infections, and death.1,2,4 Pediatric patients are considered a vulnerable patient population because of the potential risk of impaired body temperature regula- tion stemming from their large body surface area.12-14

Prevention of complications in pediatric surgical patients related to temperature variation is critical for patient safety and positive outcomes. Since the early 1990s, tech- nological innovations have helped healthcare providers detect and measure the infrared energy radiated from the human body.8 Both temporal and tympanic ther- mometries are standards of care in the OR suite, require minimum patient cooperation, and are quick and easy to use, but healthcare providers often question the accuracy of both methods.15

There is continued controversy over which non- invasive temperature monitoring method used in the preoperative and postoperative units most closely relates to core body temperature of pediatric surgical patients. Although rectal or pulmonary artery catheter tempera- ture measurement is an appropriate method to measure core temperature, it is often not feasible to use because of the level of invasiveness, type of surgical procedure, and potential risk of contamination. Thus, it is important to explore the relationship between noninvasive methods and core body temperature. The most widely used non- invasive temperature measuring methods during the surgical experience from preoperative to perioperative to recovery room are the tympanic and temporal tempera- ture measurements. Research has been saturated with focus on temperature measurement in the adult patient population in various settings. However, limited research exits on which temperature modality most closely relates to core body temperature in the pediatric surgical patient.

Therefore, current research has yielded mixed find- ings regarding which modality is best for the pediatric population in general. Intraoperatively, esophageal and

nasal temperature measurements are often performed. Esophageal temperature methods were not applicable to the present study because the esophagus is not always accessible in many surgical cases in this patient popula- tion. However, tonsillectomy and adenoidectomy were among the procedures evaluated in this study. The use of irrigation in the oral pharynx created the concern for unreliable temperature measurements; therefore, esopha- geal and nasopharyngeal temperature measurement was not considered for the dependent variable in this study. The purpose of the current study was to determine which method of temperature measurement most closely re- flects core temperature by comparing rectal temperature measurement to the less invasive tympanic and temporal temperature measurements in children during the peri- operative period.

Methods Because the focus of this study was on examining the relationship between rectal temperature and tympanic/ temporal temperature measurements, a quantitative cor- relational research design was chosen. Eligibility criteria for this study included ASA classes 1 and 2, age 3 to 17 years, and elective surgical procedure scheduled at Geisinger Medical Center or Geisinger Woodbine Outpatient Surgery in Danville, Pennsylvania. Healthy pediatric outpatients were the target sample for this study. Patients excluded from this study were any who required emergency surgery, sustained trauma, or had an ASA classification greater than class 2 because of con- genital defects or ear or rectal anomalies.

There were 2 primary research questions: (1) How do tympanic and temporal temperatures relate to rectal temperature in the pediatric surgical patient population? (2) Which temperature modality (tympanic or temporal) commonly used in the pediatric surgical patient popu- lation most accurately reflects rectal temperature after controlling for age, gender, type and length of surgery, and OR temperature?

Institutional review board approval was obtained at Geisinger Medical Center for all pediatric patients undergoing surgery at Geisinger Medical Center or Geisinger Woodbine Outpatient Surgery who satisfied the eligibility criteria. A researcher’s contact information was given to the patients and families if they wanted their study data to be removed from the study after their surgi- cal experience.

• Study Participants. Participation for this study re- quired written parental or guardian consent. Information on the consent form included the title of the study, an explanation of the voluntary involvement of the partici- pant, the lead researcher’s (D.J.M.) contact information, and the parental authorization form for the Geisinger Health System. Study description included the tempera- ture methods the researcher would use, a description of

www.aana.com/aanajournalonline AANA Journal February 2018 Vol. 86, No. 1 21

the data the researcher would collect, and the study’s purpose and procedure. Parental consent form included the presence of any risks, such as an alteration in an- esthetic technique, and the benefits, such as additional patient temperature methods. The permission form in- cluded provisions for patient anonymity. After receiving an informed parental consent form for participation, the researcher placed a coded alphanumeric index card on the head portion of the OR stretcher; this code indicated consent to participate in the study for the data collector. The code on the index card matched the alphanumeric code on the signed parental consent and child assent forms. In addition, the index card informed the data col- lector of the correct code to list on the data collection chart intraoperatively. The researcher shredded each participant’s index card and discarded it in the OR trash can immediately after completion of the surgery.

The current study had 7 variables, and assuming = .05 and power = .80, moderate effect size with R = 0.13, and 7 predictor variables, Tabachnick and Fidell16 sug- gested estimating sample size using the formula 50 + (8 × Number of predictors). Using this guideline with 7 predictors, the sample size should be at least (50 + [8 × 7]) = 106. One surgical procedure was canceled before data collection because the child vomited on induction of anesthesia, and the researcher recruited an additional participant to meet the requirement for 106 participants.

• Data Collection. Data collection involved taking tympanic temperature (Braun ThermoScan PRO 4000 ear thermometer, Welch-Allyn, Skaneateles Falls, NY) and temporal temperature (Exergen thermometer, Exergen Corp, Watertown, MA) at the start and the end of the patient’s surgery. Timing of the collection of the tym- panic and temporal temperature measurements was important because it best reflected the pediatric surgical patients’ temperature in the preoperative and postop- erative units where nursing staff use the noninvasive temperature monitors. For data collection purposes, the start of surgery was defined as after induction of general anesthesia and intravenous access placement; the end of surgery was defined as completion of surgical procedure and removal of surgical drapes. Because of variability in the length of surgeries, measuring tympanic and tempo- ral temperatures at the start and end of the case allowed for the best reflection of patients’ temperature in the preoperative and postoperative care units where these monitors are normally used during assessment for base- line measurements to guide the plan of care. The data collection for the study took place at the start and end of the surgical procedure in the OR by the anesthesia pro- vider assigned to the room; the researcher was present to ensure consistency in data collection among subjects.

The anesthesia provider received information on the data collection sequence before the procedure and had re- search study guidelines, including timing and procedural

approach of temperature measurements, available for ref- erence on the morning of the surgical procedures. Rectal temperature was documented on the data collection tool 10 minutes after insertion to account for adjustment of body temperature and measured continuously after induction of general anesthesia using a standard-sized 9F level 2 esophageal/rectal temperature probe (Smiths Medical, Dublin, OH). All temperature monitoring ceased at the end of the surgical case before extubating the patient to prevent patient awareness and discomfort.

The CRNAs collected temperature measurements for 106 pediatric patients on the day of their scheduled surgery. Data collection extended throughout an 11- week, 4-day period to obtain the required sample size of 106 pediatric surgical participants. The data included participants’ demographic data, which were age, gender, surgical procedure, and operating room temperature.

After the induction of general anesthesia and secured endotracheal tube, the anesthesia provider monitored a constant core temperature rectally in each individ- ual patient for the length of the surgical case using a Smiths Medical ER400-9, TAT-5000 temporal monitors by Exergen, and Braun PRO 4000 tympanic monitors by Welch-Allyn. Every patient received room-temperature intravenous fluids. According to the Geisinger Infoweb,17 the surgical OR temperature remained at 20.0° to 23.9°C (68°F to 75°F) in every OR according to the hospital in- fection control guidelines.

• Data Analysis. The study involved collecting 3 tem- perature measurements (tympanic, temporal, and rectal) from each pediatric surgical patient enrolled in the study during the data collection period at the start and end of the surgical case. Because there were no missing tempera- ture measurements, all temperature measurements for the 106 pediatric surgical patients were included in the analysis. Data analysis for this study was performed using SAS software version 9.2 (SAS Institute Inc, Cary, NC).

Results Of the 106 study participants, 48 (45.3%) were female and 58 (54.7%) were male. The ages of the participants ranged from 3 to 17 years, with the mean age being 7.67 years (SD = 3.95 years). Most cases were ear, nose, and throat procedures (45.28%), which lasted an average of 47.99 minutes with a mean OR room temperature of 21.94°C. Mean temperatures at the start of surgery were 36.93°C (rectal), 37.30°C (tympanic), and 37.31°C (tem- poral). At the end of surgery mean temperatures were 36.82°C (rectal), 37.05°C (tympanic), and 37.37°C (tem- poral). No significant differences were found in mean OR room temperature from start of surgery (21.67°C) to end of surgical procedure (22.2°C).

To answer the research questions, the investigators used Pearson r correlational analysis to examine the re- lationship between tympanic, temporal, and rectal tem-

22 AANA Journal February 2018 Vol. 86, No. 1 www.aana.com/aanajournalonline

peratures. Correlation between tympanic temperature and rectal temperature at the end of the surgical procedure was significant, with P < .0001 and a correlation coefficient of 0.57. Correlation between temporal temperature and rectal temperature at the end of the surgical procedure was also significant, with P = .0007 and a correlation coef- ficient of 0.33. Results of the correlation analysis are pre- sented in Tables 1 and 2. The coefficient of determination of the tympanic-rectal temperature (r = 0.57, R2 = 0.32) revealed that 32% of the total variation can be explained by the linear relationship between the tympanic and rectal temperature measurements. However, the coefficient of determination of the temporal-rectal temperature (r = 0.33, R2 = 0.10) indicated that only 10% of the total varia- tion in temperature can be explained by the linear relation- ship between the temporal and rectal temperature mea- surements. Therefore, the relationship between temporal temperature and rectal temperature was uniformly weaker than the relationship between tympanic temperature and rectal temperature (Tables 1 and 3).

Bland-Altman analysis indicated the bias between the mean differences and the agreement interval between tympanic, temporal, and rectal temperature methods. The differences between the tympanic and rectal tem- perature and the temporal and rectal temperature are plotted against the average measurement of the 2 tem-

perature devices. Bland-Altman plot analysis revealed a bias of only 0.37 with 95% limits of agreement ranging from to 1.35 to 0.62 when comparing the measurements of tympanic and rectal temperature at the start of surgery (Figure 1).

Plot analysis of the tympanic and rectal temperatures at the end of surgery revealed a bias of .37 with 95% limits of agreement ranging from 1.54 to 0.79 (Figure 2). The data results indicated tympanic temperature and rectal temperature were consistently similar, both at the start and end of the surgical case. In comparison, the Bland- Altman plot between temporal temperature and rectal temperature methods at the start of surgery revealed a bias of -0.37 with 95% limits of agreement ranging from 0.79 to 1.5 (Figure 3).

In addition, plot analysis of the temporal temperature and rectal temperature at the end of surgery revealed a bias of 0.55 with 95% limits of agreement ranging from 0.97 to 2.07. This finding indicated the 2 temperature

measures provided similar measures at the start of the surgical case but had a larger bias at the end of the surgi- cal case (Figure 4).

The difference in bias between temperature measure- ments at the start of the surgical case was 0.37, whereas the difference in bias between temperature measurements at the end of the surgical case was 0.55. Both tympanic

Table 1. Pearson Correlation of Tympanic, Temporal, and Rectal Temperatures at End of Surgerya aN = 106 for all analyses. bP = .0001.

Variable Tympanic Temporal Rectal

Tympanic 1.0

Temporal 0.63 1.0

Rectal 0.57b 0.33 1.0

Table 2. Temperature Measurements at Start and End of Surgerya aN = 106 for all analyses.

Percent Limits of agreement within limits Temperature method Bias SD Lower Upper of agreement

Tympanic–rectal, start of surgery – 0.37 0.49 – 0.62 1.35 95

Tympanic–rectal, end of surgery – 0.37 0.65 – 0.79 1.54 95

Temporal–rectal, start of surgery – 0.37 0.58 – 0.79 1.5 95

Temporal–rectal, end of surgery – 0.55 0.76 – 0.97 2.07 95

Table 3. Pearson Correlation of Tympanic, Temporal, and Rectal Temperatures at Start of Surgerya aN = 106 for all analyses. bP = .0002.

Variable Tympanic Temporal Rectal

Tympanic 1.0

Temporal 0.45 1.0

Rectal 0.36b 0.16 1.0

www.aana.com/aanajournalonline AANA Journal February 2018 Vol. 86, No. 1 23

and temporal temperature measurements are lower than the core rectal temperature both at the start and end of surgery. However, tympanic temperature measurements had smaller (less negative) bias than temporal temperature compared with the core rectal temperature (see Table 2).

Although both tympanic and temporal temperatures were positively correlated to core rectal temperature, tympanic temperature was more strongly correlated both at the start and end of surgery. Furthermore, results of

the Bland-Altman analysis supported less bias and greater levels of agreement between tympanic temperature and core rectal temperature vs temporal temperature and core rectal temperature. Although differences between both tympanic and temporal temperature measurements and core rectal temperature were small, a greater bias and wider limits of agreement existed between temporal and core rectal temperature measurements.

Two separate multiple regression analyses were per-

Ty m

p a

n ic

T e

m p

e ra

tu re

– R

e ct

a l T

e m

p e

ra tu

re

Mean

Mean of Tympanic and Rectal Temperatures

Figure 1. Bland-Altman Plot of Tympanic and Rectal Temperature Before Surgery

Ty m

p a

n ic

T e

m p

e ra

tu re

– R

e ct

a l T

e m

p e

ra tu

re

Mean

Mean of Tympanic and Rectal Temperatures

Figure 2. Bland-Altman Plot of Tympanic and Rectal Temperature at End of Surgery

24 AANA Journal February 2018 Vol. 86, No. 1 www.aana.com/aanajournalonline

formed using the general linear model because the dataset included both continuous and categorical variables. In the first analysis examining temperature measurements at the start of the surgical procedure, the predictor vari- ables were tympanic and temporal temperatures at the start of surgery and the criterion variable was core rectal temperature. Covariates include patient age, gender, and temperature of the OR. Results of regression analysis examining which temperature modality (tympanic vs

temporal) is the most accurate predictor of core rectal temperature at the start of the surgical procedure indi- cated an overall model of 1 independent variable (tym- panic temperature, P = .001) that significantly predicted core rectal temperature, R2 = 0.17, R2adj = 0.13, and F(5, 100) = 4.18, P = .0007. This model accounted for 17% of variance in core rectal temperature (Table 4).

To examine the influence of length of surgery and type of surgery on core rectal temperature, the researchers

Te m

p o

ra l T

e m

p e ra

tu re

– R

e ct

a l T

e m

p e ra

tu re

Mean

Mean of Temporal and Rectal Temperature

Figure 3. Bland-Altman Plot of Temporal and Rectal Temperature Before Surgery

Te m

p o

ra l T

e m

p e ra

tu re

– R

e ct

a l T

e m

p e ra

tu re

Mean

Mean of Temporal and Rectal Temperature

Figure 4. Bland-Altman Plot of Temporal and Rectal Temperature at End of Surgery

www.aana.com/aanajournalonline AANA Journal February 2018 Vol. 86, No. 1 25

conducted a separate regression analysis using end-of- surgery tympanic and temporal temperatures as predic- tor variables and core rectal temperature as the criterion variable. This model added the length of surgery and type of surgery covariates to the regression analysis because increased length of time in the OR environment while the patient was under general anesthesia could significantly affect the anesthetized pediatric patient’s core rectal tem- perature. Type of surgery can increase exposure of the patient to the OR environment and negatively affect core rectal temperature. Results indicated that tympanic tem- perature was the only variable that significantly predicted core rectal temperature, R2 = 0.34, R2adj = 0.30, F(7, 99) = 7.47, P = .001. This model accounted for 34% of vari- ance in core rectal temperature. A summary of regression coefficients appears in Table 5 and indicates that the tympanic temperature at the end of surgery (P = .00001) was the only variable that contributed significantly to the model after adjustment for age, gender, temperature of the OR, length of surgery, and type of surgery.

Discussion The results of this study showed that neither tympanic nor temporal temperature measurement were a perfect analog to core rectal temperature. In this study, tym- panic temperature measurement seems to be second best compared with rectal temperature measurement in the pediatric surgical patient but is not ideal.

The study had a few potential limitations. Several factors such as OR (room) temperature, body weight and size of the patient, and skin exposure during surgery may affect a surgical patient’s temperature. However, OR temperature was not a significant predictor of body temperature in this study. Furthermore, most cases were ear, nose, and throat cases, thus reducing the possible

limitation caused by skin exposure during surgery. Very young patients were excluded from the study, decreasing the impact of body size on temperature measurements. However, the impact of these potential confounding vari- ables could be examined in future research.

Results from this study may lead to the promotion of developing a standardized temperature protocol in pre- operative and postoperative areas of the surgical suite. Findings of this research study support recommending that the protocol for temperature measurement during the pediatric surgical patient experience include using the tympanic temperature modality throughout the OR suite. The most effective measure that practitioners can provide to protect surgical patients from undetected temperature change is accurate temperature monitor- ing. Although the results of this study’s findings indicate tympanic temperature measurement is inferior to core rectal temperature measurement, it remains the next best noninvasive measurement in the preoperative and post- operative settings.

Table 4. Regression Table With Five Predictor Variables at Start of Surgery

Variable 95% CI Partial correlation t P

Age .02 6.91-8.43 0.006 1.92 .058

Gender .02 – 0.15- – 0.19 0.592 0.25 .805

Temperature of operating room .02 71.01-71.96 0.00002 1.45 .151

Tympanic temperature .36 37.22-37.38 0.009 3.33 .001

Temporal temperature .01 37.22-37.39 0.006 0.13 .896

Table 5. Regression Table With Seven Predictor Variables at End of Surgery

Variable 95% CI Partial correlation t P

Age .002 6.91-8.43 0.006 0.12 .904

Gender – .09 – 0.09-0.25 0.592 – 0.89 .378

Temperature of operating room .01 71.01-71.96 0.00002 0.64 .526

Tympanic temperature .39 36.90-37.19 0.280 4.65 .00001

Temporal temperature – .03 37.23-37.51 0.002 – 0.39 .694

Length of surgery .002 41.2-54.8 0.149 1.72 .089

Type of surgery .009 – 0.26- – 0.05 0.021 0.13 .898

26 AANA Journal February 2018 Vol. 86, No. 1 www.aana.com/aanajournalonline

AUTHORS Debra J. Minzola, PhD, MSN, CRNA, is an assistant professor at Blooms- burg University of Pennsylvania and program director of the Geisinger Health System/Bloomsburg University of Pennsylvania Nurse Anesthesia Program. She completed this research study for her dissertation and graduate requirements from University of Phoenix. Email: djminzola@ geisinger.edu.

Rebecca Keele, PhD, PHCNS-BC, is a professor at Texas Woman’s University in the College of Nursing, Denton, Texas. She served as the dissertation chair for Dr Minzola during the completion of this research study. Email: [email protected].

DISCLOSURES The authors have declared no financial relationships with any commercial entity related to the content of this article. The authors did not discuss off-label use within the article.

Publishing

The 582-page second edition of the essential resource for nurse anesthesia

educators includes the latest information and best practices for:

Recommended for:

Understanding legal issues and reimbursement rules

Nurse anesthesia program administrators

Fulfi lling new educational requirements

Didactic and clinical faculty

Writing exam questions and evaluations

Students currently in leadership positions and those aspiring to an academic career

Incorporating simulation into your curriculum

Encouraging research and scholarship application

ORDER YOUR COPY TODAY! www.AANA.com/educatorsbook

#AANABOOKS

NOW AVAILABLE! A Resource for Nurse Anesthesia Educators SECOND EDITION

Also available as an e-book

Copyright of AANA Journal is the property of American Association of Nurse Anesthetists and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.