Arthritis represents several painful joint pathologies. The two most common pathologies are osteoarthritis (OA), a degenerative joint disease with loss of cartilage in synovial joints as a main characteristic , and rheumatoid arthritis (RA), a systemic autoimmune disorder. According to a World Health Organisation (WHO) report in 2015, arthritis affects millions of people worldwide. In Canada, more than 4.4 million people are living with OA, and it is anticipated that more than 10 million people will show this pathology over the next 30 years .
Despite considerable progress in the clinical and radiological characterization of arthritis, it remains cumbersome to study arthritis-associated pain. Pain scales are influenced by gender, race, age, and even culture . OA and RA have in common an underlying inflammatory process responsible for pain, redness, heat, swelling, and joint stiffness . Therefore, it is intriguing to include biochemical, physiological, and inflammation-related parameters in the study of arthritic knee pain.
In arthritis, the chemical mediators of inflammation (histamine, serotonin, bradykinin, prostaglandins, and cytokines) are released into the joint by nerves, immune cells, synovial cells, and the vascular endothelium. Inflammation is characterized by four major signs: redness, heat, swelling, and pain. Swelling results from the accumulation of fluid outside the dilated and permeable blood vessels and infiltration of cells into the damaged area. Pain is due to the direct effects of the chemical mediators of inflammation, which increase the excitability of peripheral nociceptive sensory fibers. During the inflammatory process, there is a dilation of the blood vessels, leading to an increase in blood flow. This will cause redness and heat. The heat can be measured on the skin covering the sick joint by infrared thermography (IRT).
History of infrared thermography
The first report on body heat originated from Hippocrates (460–377 BC), who used a cloth soaked in clay to compare the temperatures between healthy and diseased parts of the body based on the drying time of the clay . In the 1800s, the German astronomer William Herschel discovered infrared radiation. Molecules of every object emit infrared radiation (electromagnetic waves) as a result of their rotational-vibrational movements. That radiation is reflected and refracted like light and can be detected. IRT is a method to capture this radiation and quantify the signal, i.e. temperature , . Thus, the first electronic sensors for the detection of infrared radiation were designed by the army and used as night vision systems. At that time, IRT technology was large and expensive, and the quality of the images was mediocre . The past few years have witnessed the evolution of IRT, which has been extended to several scientific and commercial areas, including medicine.
The first IRT instrument for medical use was based on contact thermography. The device was kept in contact with the skin, separated only by a sheet with liquid crystals of black cholesterol, which changed its color depending on the temperature of the skin with which it was in contact. Then arrived remote sensing thermography, which captured the infrared emission at a distance and produced an image of the temperature distribution in real time. Today, IRT uses computers and cameras that process pictures quickly and provide high-resolution thermal images. Furthermore, those devices are small and inexpensive compared to those using other techniques (e.g. radiography) and provide high-resolution images. Unlike radiography, scintigraphy, or magnetic resonance imaging, IRT provides noncontact, noninvasive, radiation-free, and dynamic measurement of joint inflammation.
Thermal data are acquired and interpreted using a camera and a thermal evaluation software. In 1974, Collins et al.  developed an index to quantify infrared imaging, namely the thermographic index (TI). It represents the difference between the measured temperature of several regions of interest and a predefined isotherm constant, divided by the total area measured. The isotherm constant is calculated from data from healthy subjects. In a very simplified way, TI represents the difference between the measured values and “normal” values. After realizing that the heat is not distributed in the same way depending on whether the joint is sick or healthy, Salisbury et al.  created few years later the heat distribution index (HDI). This index is more complex than TI and allows evaluation of the measured temperature while minimizing the influence of the environment. Accordingly, Salisbury et al. demonstrated in their study that HDI is more accurate than IT.
Presenting evidence of the relationship between skin temperature (measured by IRT) and arthritic knee pain and discussing the limitations of IRT in the clinical settings of arthritis are the objectives of this review. To achieve these goals, we performed a literature search using the following keywords on PubMed: infrared, thermography, arthritis. Among the 58 articles found, only 10 were relevant.
Infrared thermography in arthritis
IRT has been evaluated under several musculoskeletal conditions, such as complex regional pain syndrome , juvenile polyarthritis, rheumatoid arthritis, and osteoarthritis , , , , , , , , . Bhowmik et al. studied inflammation in human knees with OA and RA by IR . They found that patients with RA showed bilateral temperature increases, while in OA a unilateral temperature increase was observed (Figure 1).
Another study by Lerkvaleekul et al. evaluated the use of IRT and ultrasonography in the diagnosis of wrist arthritis . Three groups of patients with different degrees of arthritis were separated following physical examination: healthy controls, those with inactive arthritis, and those with active arthritis (including mild and severe arthritis as subgroups). The mean and maximum temperatures assessed by IRT in the region of interest were significantly higher in the active arthritis group than in the inactive group and healthy controls. Furthermore, temperatures in the severe arthritis subgroup were significantly higher than in the mild arthritis subgroup. HDI followed the same trend: in the moderate to severe arthritis group, this index was significantly higher than in healthy controls. The results clearly suggested a relationship between the severity of arthritis and absolute temperature changes.
Denoble et al. conducted a study in patients with knee OA using IR . The participants were selected according to Kellgren–Lawrence (KL) radiographic criteria: knee OA cases had a KL grade of 2 and 3. The knee temperature measured at the patella (region of interest) was higher for KL3 cases (30.5 °C) than for KL2 cases (30.1 °C). Even though the results were not statistically significant, an association between knee temperature and the severity of the pathology was found as in the previous study.
Ilowite et al.  studied the relationship between pain and the degree of joint inflammation measured by IRT in children from 4 to 16 years with juvenile RA (pauciarticular arthritis, polyarticular arthritis, and systemic onset disease). Pain was assessed using a visual analog scale by the children themselves, then by their parents, and finally by a pediatric rheumatologist immediately after a physical examination. There was a significant correlation between the pain score determined by the parents and physicians and the temperature of the diseased joint measured by IRT. The only significant correlation between the pain intensity assessed by children themselves and the joint temperature was in children under the age of seven.
All these studies confirm the feasibility of IRT as a reliable technique to detect inflammation related to arthritis (Table 1). However, except in one study, classical pain scores were neither included nor correlated.
Inflammatory pain and infrared thermography
Only a few studies used IRT as therapeutic monitoring tool for pain assessment. Hegedűs et al. evaluated the efficacy of low-level laser treatment for knee OA and employed IRT to verify the therapeutic effects . OA patients receiving low-level laser treatment were compared with untreated control patients. The efficacy of treatment to reduce pain in patients was evaluated by a subjective scale of pain from 0 to 10, sensitivity of the knee to pressure, and measurement of the skin temperature of the joint by IRT before, during, and after treatment. This study found a significant change of skin temperature of the knee measured by IRT in patients who had experienced pain relief.
Kwon et al. compared the analgesic effect of bee venom administered with traditional acupuncture in patients with OA of the knee . The comparison was based on heat (IRT) and pain relief (patients were asked to choose a category from 1 to 4 that best described the reduction of their pain: excellent “4”, good “3”, fair “2”, and poor “1”). After 4 weeks of treatment, patients receiving bee venom had a better reduction in knee pain and skin temperature than those given traditional acupuncture.
The ability of Hyalgan (1% hyaluronic acid solution) to improve pain reduction after knee surgery in patients with OA was observed in the study by Zazirnyĭ et al. . The intensity of pain (visual analog scale) and the temperature of the skin over the sick joint (measured by IRT) revealed that 5 weeks of treatment with Hyalgan improved the effectiveness of OA surgery. Overall, these studies suggest that IRT can be an interesting method to assess inflammatory pain (Table 2).
Limitations of infrared thermography
Some factors may impact the IRT signal, such as equipment specifics, room temperature, changes in the distance between the joint and the camera, angle of measurement, and internal factors such as variability of the blood flow due to blood composition (rheology) and/or mechanical disturbances of the local blood flow (e.g. deep vein thrombosis [DVT]) . These factors must be considered when using IRT. For better reproducibility of IRT, it is important to determine the region of interest on which the temperatures are to be measured. This region can be determined by the association of history of the disease and the apparent clinical signs of inflammation .
Another potential limitation for the use of IRT in clinical routine is the cost factor. The cost for IRT basically includes those of the IR camera and a laptop computer. Usually, no disposables or chemicals are needed. The cameras must be medically approved, and thermography costs are covered in many health care systems. In comparison to other noninvasive systems capable of recognizing changes of skin perfusion related to joint inflammation, such as laser Doppler flowmetry or laser speckle imaging, the IRT equipment is significantly less expensive.
IRT is a simple, accurate, inexpensive, noninvasive, and radiation-free technique to assess skin temperature. In the diagnosis and therapeutic follow-up of pathologies such as arthritis, numerous studies have demonstrated its effectiveness and its reproducibility to detect inflammation. Some authors have examined the relationship between pain and the degree of joint inflammation and have found a correlation between the temperature of the skin measured by IRT and the intensity of the pain. Despite some limitations such as the influence of the environment and blood flow variation in its measurements, IRT is a useful tool for assessing inflammatory pain.
Bombardier C, Hawker G, Mosher D. The impact of arthritis in Canada: today and over the next 30 years. Arthritis Alliance of Canada, Fall 2011. Google Scholar
Torres CA, Bartley EJ, Wandner LD, Alqudah AF, Hirsh AT, Robinson ME. The influence of sex, race, and age on pain assessment and treatment decisions using virtual human technology: a cross-national comparison. J Pain Res 2013;6:577–88. PubMedWeb of ScienceGoogle Scholar
Lasanen R. Infrared thermography in the evaluation of skin temperature: applications in musculoskeletal conditions. Publications of the University of Eastern Finland. Dissertations in Forestry and Natural Sciences No 186, 2015. Google Scholar
Swiderski W. Military applications of infrared thermography nondestructive testing in Poland. Zielonka, Poland: Military Institute of Armament Technology, 2015. Google Scholar
Collins AJ, Ring EFJ, Cosh JA, Bacon PA. Quantitation of thermography in arthritis using multi isothermal analysis. I. The thermographic index. Ann Rheum Dis 1974;33:113–5. CrossrefPubMedGoogle Scholar
Salisbury RS, Parr G, de Silva M, Hazleman BL, Page-Thomas DP. Heat distribution over normal and abnormal joints: Thermal pattern and quantification. Ann Rheum Dis 1983;42: 494–9. CrossrefPubMedGoogle Scholar
Devereaux MD, Parr GR, Thomas DP, Hazleman BL. Disease activity indexes in rheumatoid arthritis; a prospective, comparative study with thermography. Ann Rheum Dis 1985; 44:434–7. CrossrefPubMedGoogle Scholar
Paterson J, Watson WS, Teasdale E, Evans AL, Newman P, James WB, et al. Assessment of rheumatoid inflammation in the knee joint. A reappraisal. Ann Rheum Dis 1978;37:48–52. PubMedCrossrefGoogle Scholar
de Silva M, Kyle V, Hazleman B, Salisbury R, Page Thomas P, Wraight P. Assessment of inflammation in the rheumatoid knee joint: correlation between clinical, radioisotopic, and thermographic methods. Ann Rheum Dis 1986;45:277–80. PubMedCrossrefGoogle Scholar
Snekhalatha U, Anburajan M, Sowmiya V, Venkatraman B, Menaka M. Automated hand thermal image segmentation and feature extraction in the evaluation of rheumatoid arthritis. Proc Inst Mech Eng H 2015;229:319–31. Web of ScienceCrossrefPubMedGoogle Scholar
Capo A, Ismail E, Cardone D, Celletti E, Auriemma M, Sabatini E, et al. Joint functional impairment and thermal alterations in patients with psoriatic arthritis: a thermal imaging study. Microvasc Res 2015;102:86–91. Web of SciencePubMedCrossrefGoogle Scholar
Denoble AE, Hall N, Pieper CF, Kraus VB. Patellar skin surface temperature by thermography reflects knee osteoarthritis severity. Clin Med Insights Arthritis Musculoskelet Disord 2010;3:69–75. PubMedGoogle Scholar
Lerkvaleekul B, Jaovisidha S, Sungkarat W, Chitrapazt N, Fuangfa P, Ruangchaijatuporn T, et al. The comparisons between thermography and ultrasonography with physical examination for wrist joint assessment in juvenile idiopathic arthritis. Physiol Meas 2017;38:691–700. PubMedCrossrefWeb of ScienceGoogle Scholar
Ilowite NT, Walco GA, Pochaczevsky R. Assessment of pain in patients with juvenile rheumatoid arthritis: relation between pain intensity and degree of joint inflammation. Ann Rheum Dis 1992;51:343–6. CrossrefPubMedGoogle Scholar
Hegedűs B, Viharos L, Gervain M, Gálfi M. The effect of low-level laser in knee osteoarthritis: a double-blind, randomized, placebo-controlled trial. Photomed Laser Surg 2009;27:577–84. PubMedWeb of ScienceCrossrefGoogle Scholar
Kwon Y, Kim J, Yoon J, Lee J, Han H, Mar W, et al. The analgesic efficacy of bee venom acupuncture for knee osteoarthritis: a comparative study with needle acupuncture. Am J Chin Med 2001;29:1–10. Google Scholar
Zazirnyĭ IM, Mahomedov OM, Drobotun OV, Ievseienko VH. Effect of chondroprotectors in the treatment and prophylaxis of the knee joint osteoarthrosis. Likars’ka Sprav 2000;7–8:50–2. Google Scholar
About the article
Published Online: 2018-10-31
Author contributions: All the authors have accepted the responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: None declared.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.