Infrared Sensor Measurement of Neck Carotid Skin Temperature in The Emergency Department: A Safe Alternative to the Tympanic Method

Introduction

Body temperature is essential in infection screening, evaluation of hemodynamic stabilization, and triage processes in emergency departments. The COVID-19 pandemic has increased the demand for rapid, non-contact, and reliable measurement methods, and the importance of these methods in clinical practice has been emphasized.¹

Tympanic thermometers are considered the gold standard because they reflect the hypothalamic temperature. However, a bulb in the ear canal, measurement errors (15-30% error rate), patient non-compliance, and technical difficulties correctly positioning the device limit its use.^2,3

The neck carotid artery region offers an ideal site for thermal measurements due to its superficial anatomy, high blood flow, and proximity of skin temperature to central body temperature.⁴ Infrared sensors can be used in this region with the advantage of non-contact measurement; however, the sensitivity of standard sensors can be affected by environmental factors (e.g., ambient temperature, skin humidity).⁵

Studies focusing on carotid artery skin temperature measurement in the literature are limited and mainly include the pediatric population.⁶ Comprehensive studies evaluating the clinical safety of this method in adult patients are lacking, and this study aims to fill this gap.

Materials and Methods

Study Population 

This prospective cross-sectional study included 520 patients admitted to Tokat Gaziosmanpaşa University Emergency Department between July 2023 and March 2024. Inclusion criteria: clarity of consciousness (Glasgow Coma Scale ≥14), written informed consent, and absence of eardrum pathology. Exclusion criteria: neck trauma, carotid artery pathology, skin lesions, use of antipyretics (last 4 hours), and severe hemodynamic instability (systolic blood pressure <90 mmHg).

Measurement Protocol 

In each patient, a reference measurement was taken with a Braun ThermoScan 7 tympanic thermometer. Three measurements were taken with a FLIR T530 infrared sensor at a distance of 5 cm from the right carotid artery, and the mean was recorded. The same trained operator performed the measurements with the patient’s neck exposed and skin dry. The measurement time was recorded with a stopwatch.

The ambient temperature was kept constant at 20-30±1°C, and sensor performance was evaluated in control experiments at 20°C, 25°C, and 30°C. In control experiments (n=60), the effect of temperature variations on measurement accuracy was limited to 1.1% variation. The influence of external factors was minimized by controlling ambient humidity (40-60%) and airflow.

Statistical Analysis 

Data were analyzed using SPSS 26.0 and MedCalc 20.0 programs. Agreement between methods was evaluated by Bland-Altman analysis (mean difference and 95% limits of agreement), Intraclass Correlation Coefficient (ICC, 2-way mixed model, absolute agreement), and Deming regression analysis. An independent sample t-test compared measurement times. Significance level was accepted as p<0.05. Power analysis showed that a minimum of 500 patients was required for power and a 0.2°C clinical difference (α=0.05).

Findings

Demographic Data 

The mean age of 520 patients was 39.1±13.8 years, 63.5% were female (n=330) and 36.5% were male (n=190). The mean body mass index (BMI) was 25.9±4.1 kg/m². Age distribution was grouped as  18-30 (40%), 31-50 (43%), and 51-65 (17%) years (Table 1).  

Table 1: Demographic and Clinical Characteristics of the Study Population (N=520)

Note: SD = Standard Deviation. Percentages were calculated according to the total number of participants (N=520).

The table shows demographic characteristics (mean age: 39.1 ± 13.8 years, gender: 63.5% female, BMI: 25.9 ± 4.1 kg/m², age groups: 18-30 years, 40%, 31-50 years, 43%, 51-65 years, 17%).

Thermal Measurement Performance 

The mean temperature was 37.28±0.47°C in tympanic measurements and 37.14±0.43°C in carotid artery infrared measurements (p=0.07). The measurement time was 8.3±1.4 seconds for tympanic and 3.0±0.6 seconds for carotid infrared measurements (p<0.001). Carotid measurement time was 64% shorter than the tympanic method. The effect of ambient temperature changes (20-30°C) on measurement accuracy was minimal (1.1% variation).

Table 2: Measurement Performance Comparison of Tympanic thermometer and RDA IR methods.

In this Table, parameters such as mean temperature values, measurement times, and the effect of ambient temperature variations are presented to assess the clinical applicability of the methods.

Tympanic and Forehead Method Comparison

The agreement between the tympanic and forehead methods was also analyzed for comparison. Bland-Altman analysis showed that the tympanic measurement was on average 0.63°C higher than the forehead measurement, and the limits of agreement (-0.71°C to +1.97°C) were broader than in the RCA comparison. The ICC value was 0.726 (95% CI) (p<0.001), indicating a “moderate” level of agreement. These results support that the RDA measurement is closer to the tympanic measurement and more reliable than the forehead measurement (Table 3).

Table 3: Comparative Fit Metrics of the Tympanic Method with the Carotid and Forehead Methods

Distribution of differences between tympanic and forehead temperature measurements. The Bland-Altman plot (Fig.1) shows the distribution of disagreements and means between tympanic and forehead temperature measurements. The mean difference was 0.633°C, which is statistically significant. The lower and upper limits of agreement were -0.454°C and 1.719°C, respectively. The distribution observed in the graph shows a moderate level of agreement between the two measurement methods. This result suggests that non-contact infrared measurements from the forehead region systematically give slightly lower values than tympanic measurements and should be considered carefully in clinical practice (Figure 1).

Tympanic and Cervical Carotid Artery Skin Temperature Measurements. The Bland-Altman plot shows the distribution of differences and means between tympanic and cervical carotid artery skin temperature measurements. The mean difference is 0.309°C, which is statistically significant. The lower and upper limits of agreement were -0.606°C and 1.223°C, respectively. The distribution in the graph shows a good agreement between the two measurement methods. This finding supports that cervical carotid artery skin temperature measurements provide closer and more reliable results than tympanic measurements and can be used as an alternative measurement site in the emergency department setting (Figure 2).

Figure 1: Bland-Altman analysis between tympanic and forehead temperature measurements.Click here to View Figure

Figure 2: Bland-Altman analysis between tympanic and neck carotid artery temperature measurements.Click here to View Figure

Concordance between Methods 

Bland-Altman analysis showed a mean difference of 0.14°C (±0.21°C) and limits of agreement between -0.28°C and +0.56°C. The ICC value was calculated as  0.93 (95% CI: 0.90-0.95, p<0.001).

DeMing regression analysis showed a significant linear relationship between the two methods (R²=0.89, p<0.001).

In subgroup analysis, ICC was found to be 0.90 in patients with BMI >30 kg/m² and 0.94 in patients with BMI ≤30 kg/m², but the difference was not statistically significant (p=0.15).

Additional Technical Information and Sensor Performance 

Sensor Calibration and Performance Tests 

The FLIR T530 infrared sensor was calibrated every 50 measurements using a standard blackbody source (Fluke 4180, ±0.05°C). Pilot measurements (n=30) showed a 10% increase in thermal sensitivity and a 7% reduction in measurement variation of the coated sensor compared to the uncoated sensor.

Impact of Environmental Conditions 

Sensor performance was tested at ambient temperatures of 20°C, 25°C, and 30°C, with measurement variation limited to 1.1%. Humidity was kept between 40% and 60%, and airflow was minimized.

Anatomical and Physiological Factors 

The depth of the carotid artery and subcutaneous adipose tissue thickness were measured by ultrasound, and correlation analysis was performed with the measurement results. A slight increase in measurement variation was observed with increasing depth (r=0.28, p=0.04), but it did not create a clinically significant difference.

Discussion

In this study, infrared sensors provided a high degree of agreement and reliability with tympanic thermometry in measuring neck carotid artery skin temperature. The mean temperature difference of 0.14°C and the Intraclass Correlation Coefficient (ICC) of 0.93 demonstrate clinically meaningful and acceptable proximity between the two methods. Measurement accuracy is vital for detecting small changes in body temperature, especially in critical clinical situations such as infection diagnosis and sepsis. In this context, the high sensitivity provided by infrared sensors supports early diagnosis and rapid intervention processes in emergency services.

The carotid artery region is ideal for accurate and reliable body temperature measurement due to its central blood flow and superficial anatomical location. Non-contact infrared measurements from this region give results close to tympanic measurement, which has a high capacity to reflect hypothalamic temperature, but offers the advantage of non-contact and faster measurement. The choice of the measurement site is a critical factor in the clinical validity and ease of application of the sensor.

In clinical practice, tympanic thermometers’ measurement time and technical difficulties can lead to time loss and measurement errors, especially during intensive patient triage. The 64% shorter measurement time of infrared sensors in the carotid region speeds up the workflow in emergency departments and increases the efficiency of healthcare personnel. In addition, the non-contact measurement method provides a significant advantage, especially in pandemic and infectious disease control by reducing the risk of infection transmission.

The minimal influence of environmental factors such as ambient temperature and humidity on the accuracy of the measurement shows that the sensor can be used reliably in different clinical conditions.

Limitations of the study include the lack of validation in elderly and pediatric patient groups and the lack of evaluation of the effect of different skin types on sensor performance. Further studies in these areas will expand the method’s general clinical validity and application area.

Conclusion 

Infrared sensors offer high sensitivity and reliability for neck carotid artery skin temperature measurement compared to tympanic thermometry, significantly reducing measurement time. Non-contact measurements from the carotid region provide accuracy and speed in clinical decision-making processes thanks to their capacity to reflect central body temperature. This technology stands out as an ideal alternative for infection screening, early diagnosis of sepsis, and effective management of triage processes in emergency departments.

The advantage of non-contact and rapid measurement is that it plays a critical role in preventing the spread of infectious diseases and improving the safety of healthcare workers and patients. In the future, comprehensive evaluation of sensor performance in different age groups, skin types, and environmental conditions will increase the prevalence and acceptance of this method in clinical practice.

Acknowledgment

No funding was received for this study. There are no additional individuals or organizations to acknowledge for their contributions to this work.

Funding Sources

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Conflict of Interest

The author(s) do not have any conflict of interest.

Data Availability Statement-

This statement does not apply to this article.

Ethics Statement

Tokat Gaziosmanpaşa University Ethics Committee approved the study with the number 20-KAEK-323. All participants gave written informed consent before the study.

Authors’ Contributions

All authors contributed equally to the study design, data collection, analysis, and manuscript preparation. All authors reviewed and approved the final version of the manuscript.

References

1. Tapar E. Kulak zarı, alın ve boyun vücut sıcaklığı ölçümlerinin değerlendirilmesi [yüksek lisans tezi]. Tokat, Türkiye: Tokat Gaziosmanpaşa Üniversitesi, Acil Tıp Hemşireliği Anabilim Dalı; 2022.
2. Zhen C, Xia Z, Ya Jun Z. Çocuklarda ateş tanısında kullanılan kızılötesi kulak termometrisinin doğruluğu: Sistematik bir inceleme ve meta-analiz. Clinical Pediatrics. 2014;54(2):114-126. doi:10.1177/0009922814545492
   CrossRef
3. Ogoina D. Ateş, ateş paternleri ve ‘ateş’ olarak adlandırılan hastalıklar – Bir derleme. Journal of Infection and Public Health. 2011;4(3):108-124. doi:10.1016/j.jiph.2011.05.002
   CrossRef
4. Bayhan C, Özsürekçi Y, Tekçam N, Güloğlu A, Ehliz G, Ceyhan M, Kara A. Temassız kızılötesi termometre ile kulak kızılötesi termometrenin karşılaştırılması. Çocuk Enfeksiyon Dergisi. 2014;8:52-55.
   CrossRef
5. Dzien C, Halder W, Winner H, Lechleitner M. Covid-19 taramasında soğuk dış ortam sıcaklıklarında alın sıcaklığı ölçümleri gerçekten faydalı mı? Wiener Klinische Wochenschrift. 2021;133(7-8):331-335. doi:10.1007/s00508-020-01794-8
   CrossRef
6. Imani, F., Rouzbahani, H. R. K., Goudarzi, M., Tarrahi, M. J., & Soltani, A. E. (2016). Skin temperature over the carotid artery is an accurate, non-invasive estimation of near-core temperature: Anesthesiology and Pain Medicine, 6(1), e32444.
   CrossRef