Abstract
Background
Haemophilia, caused by deficiencies in factor VIII or IX, leads to deep tissue bleeding and haemophilic arthropathy. Bleeding, arthropathy, and the perioperative period of surgery for arthropathy are associated with severe pain. To avoid this pain, patients may compensate by overloading other parts of the body, potentially resulting in pain or secondary damage in areas distant from the primary site. This study investigates the use of infrared thermography (IRT) to monitor whole-body surface temperatures in people with haemophilia and to explore the potential of infrared thermography for early detection of compensatory overload.
Methods
A cross-sectional study involved 24 people with haemophilia aged 6–76 years old, experiencing pain after bleeding, post-orthopaedic surgery, or haemophilic arthropathy. Thermal images were captured with IRT and analyzed using software that utilises deep learning for whole-body temperature mapping.
Results
Increases in temperature indicative of overload were observed in areas distant from the affected site, either on the same side (vertical pattern) or on the opposite side (diagonal pattern) relative to the site of pain. These patterns were observed in 13 of 14 participants with haemophilic arthropathy, all post-orthopaedic surgery participants, and 5 of 6 after bleeding.
Conclusions
Temperature increases occurred in areas beyond the painful area, suggesting strain even in asymptomatic regions. People with haemophilia experiencing pain may require careful monitoring and treatment of surrounding areas. Our findings could aid in diagnosing haemorrhage and local inflammation while informing treatment decisions.
Clinical trial registration
Not applicable.
Keywords: Bleeding, Haemophilia, Infrared thermography, Pain, Whole-body mapping
Background
Haemophilia is an X-linked recessive bleeding disorder caused by a deficiency or abnormality in either factor VIII (haemophilia A) or factor IX (haemophilia B) [1]. People with haemophilia (PwH) experience bleeding in deep tissues, such as intramuscular and intra-articular tissues, not only during injury or bruising but also during exercise and daily living activities [2]. Intramuscular bleeding causes severe pain, muscle weakness and atrophy, and intra-articular bleeding causes severe pain; if repeated, intra-articular bleeding can damage articular cartilage and subchondral bone, resulting in haemophilic arthropathy, which causes chronic pain and loss of function [3, 4]. PwH with advanced haemophilic arthropathy may undergo surgical treatment, such as synovectomy [5] (excision or destruction of the hypertrophic synovium) or joint replacement therapy [6, 7], and severe pain is common after these surgeries. All these conditions can cause functional impairment, not only because of structural abnormalities but also because of pain. In clinical situations, we sometimes encounter people with haemophilic arthropathy who develop pain or new haemophilic arthropathy in surrounding joints, including the contralateral joint, but we have not been able to find any reports in which these phenomena have been investigated. We hypothesized that these secondary injuries were caused by overloading of unaffected parts of the body to compensate for the function of the affected part.
In recent years, infrared thermography (IRT) has been introduced into medical practice, allowing for the detection of abnormalities and tissue damage by measuring differences in skin temperature between the left and right sides of the body. In individuals without haemophilia, an asymmetry of 0.5 °C or more is not considered physiological [8, 9]. We have previously reported the usefulness of IRT-based local surface temperature analysis as an adjunctive method for the early detection of bleeding and the prompt initiation of treatment in the early stages of symptoms, demonstrating its potential to help prevent the progression of haemophilic arthropathy [10]. In the field of sports medicine, researchers are utilizing increases in body surface temperature to identify overused areas, enabling early rest and rehabilitation to prevent future injuries [11]. Similarly, in individuals with haemophilia, early detection of excessive mechanical load—alongside bleeding—may help prevent further joint or muscle damage. Currently, some professional soccer teams are using systems developed based on sports medicine research that automatically analyze skin temperature and detect asymmetrical body temperature patterns. These systems require only a handheld infrared camera and a PC connected to an Internet-linked image analysis system, making them easy to implement in clinical settings.
In this study, we tested the hypothesis that IRT-based whole-body surface temperature monitoring could be adapted for individuals with haemophilia who have joint or muscle damage, allowing for the visualization of areas under excessive load.
Methods
Study design and participants
This was a descriptive cross-sectional study in which IRT was used to image the whole-body surface temperature of PwH. The IRT method adopted in this study has been reported very rarely in the medical field, making it difficult to determine the number of cases required to establish statistical significance in advance. For this reason, this study did not set a target number of cases and analysed all cases that met the inclusion criteria among PwH who visited Nara Medical University Hospital between November 2019 and June 2024. The participants included twenty-four PwH, aged 6–76 years. The inclusion criteria were as follows: (i) had a recent bleeding event, (ii) had undergone orthopaedic surgery, or (iii) had haemophilic arthropathy. Participants were excluded if they had difficulty maintaining an upright position independently, including due to pain, or if they did not consent to participate in the study. The presence of haematoma was also examined using ultrasonography or MRI immediately after treatment with clotting factor concentrates. An orthopaedic surgeon assessed the presence and severity of haemophilic arthropathy in each joint with the Arnold–Hilgartner staging system. The participants with haemophilic arthropathy were categorized as having stage 4–5 disease, whereas those without haemophilic arthropathy were categorized as having stage 0–3 disease.
Ethics statement
All participants received a clear and comprehensive explanation of the study in accordance with the Declaration of Helsinki. The participants subsequently signed informed consent forms. Ethical approval for the studies was obtained from both the Nara Medical University Ethics Committee (ID: 2097) and the Chugai Ethics Committees (ID: E19086).
Data collection
The following data were collected from the participants’ electronic medical records: age; severity of haemophilia on the basis of plasma factor levels (severe: <1%; moderate: 1%−<5%; and mild: 5% to < 40%); and information on the treatment regimen, including the formulation name of coagulation factor concentrates and the infusion interval of the concentrates.
Thermal image acquisition via IRT
IRT is a nonradiating, noncontact, noninvasive technology that facilitates the monitoring of physiological variables through the measurement of skin temperature [12]. Thermal images were obtained with a FLIR E54-EST (Teledyne FLIR LLC, Wilsonville, OR, USA), which has a spectral sensitivity range of 7.5 μm to 14 μm and a geometric resolution of 1.75 mrad. Prior to thermal imaging, the participants disrobed to their undergarments to acclimate their skin to ambient temperature for at least 5 min. During the imaging process, precautions were taken to prevent the influence of environmental factors, such as air conditioner drafts and direct sunlight, from affecting the participants.
Thermal image analysis with ThermoHuman software
The absolute temperature is affected by various factors, such as age, hair density, and/or genetics [8, 12]. Therefore, while it is possible to compare the temperature of a region of interest with the corresponding area on the opposite side, evaluating temperature in absolute temperatures or comparing different body regions is challenging. For this reason, body surface temperature assessments using IRT have traditionally been conducted by comparing temperatures between symmetrical areas. Our method also follows this approach. The temperature difference between the region of interests (ROI) and its contralateral side was defined as the Δ skin temperature (Δ skin temp).
The obtained thermal images were analysed with ThermoHuman software version 2.21 (Pema Thermo Group, Madrid, Spain). The software quickly maps the body on the basis of anatomical knowledge by means of methods built on deep learning, calculates the average temperature of all subdivided regions, and generates a colour-coded avatar image (Fig. 1A). The higher side of the body surface temperature is coloured and displayed in the avatar as follows: Δ skin temp of 0.3 to < 0.6 °C: yellow; 0.6 to < 0.9 °C: orange; 0.9 to < 1.2 °C: red; 1.2 to < 1.5 °C: purple; and ≥ 1.5 °C: black (Fig. 1B). The estimates of clinically significant skin temperature asymmetry have been previously reported to exceed 0.3 ~ 0.5 °C [8, 10]. Therefore, in the present study, a Δ skin temp of “0.3 to < 0.6°C: yellow” or higher was considered a significant temperature difference between the right and left sides.
Fig. 1.
(A) Measurement areas provided by ThermoHuman software (front and back views) and segmentation. Panel (a) shows an example of a thermal image captured using IRT. Prior to thermal imaging, the participants disrobed to their undergarments to acclimate their skin to ambient temperature for at least 5 min. Panel (b) is the ROI image analysed from the thermal image using ThermoHuman® software. A Δ skin temp of more than 0.3 °C compared with the same area on the contralateral side is indicated in yellow and red. The relative temperature of the ROI and its contralateral side in the same subject was evaluated. The temperature difference between the ROI and its contralateral side was considered the Δ skin temp. (B) Colour coding by temperature range. Δ skin temp of 0.3 to < 0.6 °C: yellow; 0.6 to < 0.9 °C: orange; 0.9 to < 1.2 °C: red; 1.2 to < 1.5 °C: purple; and ≥ 1.5 °C: black
Results
Clinical characteristics of people with haemophilia
A summary of the clinical information is shown in Table 1 with detailed information shown in Table 2. Twenty-four participants were examined: 6 participants with haematoma, 4 participants after orthopaedic surgery, and 14 participants with haemophilic arthropathy. Among the participants, 20 had haemophilia A, and 4 had haemophilia B. The severity was classified as follows: 14 severe cases, 3 moderate cases, and 7 mild cases. Eleven participants received prophylaxis with clotting factor concentrates, 8 received emicizumab prophylaxis, and 5 participants underwent on-demand treatment.
Table 1.
Summary of the clinical information of the enrolled participants
| Total | Acute pain (After bleeding) |
Postoperative pain | Chronic pain (Haemophilic arthropathy) |
|
|---|---|---|---|---|
| Total participants (N) | 24 | 6 | 4 | 14 |
| Age range (year) | 6–76 | 6–45 | 17–68 | 11–76 |
| Median | 39 | 18 | 32.5 | 44 |
| Haemophilia type (N) | ||||
| A; mild | 6 | 3 | 1 | 2 |
| A; moderate | 1 | 0 | 0 | 1 |
| A; severe | 13 | 0 | 2 | 11 |
| B; mild | 1 | 1 | 0 | 0 |
| B; moderate | 2 | 1 | 1 | 0 |
| B; severe | 1 | 1 | 0 | 0 |
| Treatment type (N) | ||||
| CFC prophylaxis | 11 | 3 | 1 | 7 |
| Emi prophylaxis | 8 | 0 | 2 | 6 |
| On-demand | 5 | 3 | 1 | 1 |
| Details |
• Intramuscular: 4 • Intra-articular: 1 • Subcutaneous: 1 |
• Clavicle locking plate: 1 • Arthroscopic synovectomy: 2 • Total knee arthroplasty: 1 |
• All cases: no exacerbation of pain |
CFC, clotting factor concentrates; Emi, emicizumab
Table 2.
Details of the clinical information of the enrolled participants
| Participant | Age(y) | Haemophilia type; severity | Treatment | Symptom/Procedure | Type of pain |
|---|---|---|---|---|---|
| 1 | 45 | A; mild | On demand | After bleeding (intramuscular) | Acute pain due to bleeding |
| 2 | 12 | B; mild | On demand | After bleeding (intra-articular) | Acute pain due to bleeding |
| 3 | 6 | B; severe | CFC prophylaxis | After bleeding (subcutaneous) | Acute pain due to bleeding |
| 4 | 20 | A; mild | CFC prophylaxis | After bleeding (intramuscular) | Acute pain due to bleeding |
| 5 | 16 | A; mild | On demand | After bleeding (intramuscular) | Acute pain due to bleeding |
| 6 | 22 | B; moderate | CFC prophylaxis | After bleeding (intramuscular) | Acute pain due to bleeding |
| 7 | 17 | A; mild | On demand | Orthopaedic surgery (clavicle locking plate) | Postoperative pain |
| 8 | 22 | B; moderate | CFC prophylaxis | Orthopaedic surgery (arthroscopic synovectomy) | Postoperative pain |
| 9 | 43 | A; severe | Emi prophylaxis | Orthopaedic surgery (arthroscopic synovectomy) | Postoperative pain |
| 10 | 68 | A; severe | Emi prophylaxis | Orthopaedic surgery (total knee arthroplasty) | Postoperative pain |
| 11 | 46 | A; severe | CFC prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 12 | 25 | A; severe | CFC prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 13 | 53 | A; severe | CFC prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 14 | 42 | A; mild | Emi prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 15 | 51 | A; severe | On demand | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 16 | 30 | A; mild | CFC prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 17 | 11 | A; severe | Emi prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 18 | 41 | A; moderate | Emi prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 19 | 37 | A; severe | Emi prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 20 | 62 | A; severe | CFC prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 21 | 68 | A; severe | Emi prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 22 | 50 | A; severe | CFC prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 23 | 22 | A; severe | CFC prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
| 24 | 76 | A; severe | Emi prophylaxis | Haemophilic arthropathy | Chronic pain due to haemophilic arthropathy (no exacerbation of pain) |
CFC, clotting factor concentrates; Emi, emicizumab
Temperature increases in affected areas
The details of the analysis of Δ skin temp are shown in Table 3 (participants after bleeding), Table 4 (participants after orthopaedic surgery), and Table 5 (participants with haemophilic arthropathy).
Table 3.
Details of the analysis of the increase in skin temperature in participants after bleeding
| Participant | Affected area | Δ skin temp of the affected area | Δ skin temp of the surrounding area | Other haemophilic arthropathy and history of surgery | Days from the date of inception of the active bleeding to the date of the shooting | |
|---|---|---|---|---|---|---|
| Affected side (vertical pattern) |
Unaffected side (diagonal pattern) |
|||||
| 1 | L-triceps surae muscle |
+ 0.52–0.63 °C (L-inner back leg: +0.63 °C, L-outer back leg: +0.52 °C) |
L-ankle: +0.68 °C L-anterior foot: +1.05 °C |
R-front thigh: +0.53–0.62 °C R-knee: +1.14 °C R-popliteus: +0.42 °C R-front leg: +0.41 °C |
R-ankle | 1 |
| 2 | R-ankle | + 0.78 °C |
R-front thigh: +0.32–1.07 °C R-back thigh: +0.48 °C R-popliteus: +0.31 °C R-front leg: +0.41–0.48 °C R-anterior foot: +0.93 °C |
- | None | 1 |
| 3 | R-back femoral |
+ 0.55–1.00 °C (R-outer back thigh: + 0.61 °C, R-central back thigh: +1.00 °C, R-inner back thigh: +0.55 °C) |
R-gluteus: +0.63 °C R-popliteus: +0.59 °C R-back leg: +0.38 °C R-ankle: +1.09 °C R-achilleas: +1.15 °C R-anterior foot: +0.97 °C R-back foot: +1.43 °C |
L-gluteus: +0.48 °C L-front thigh: +0.36–1.76 °C L-knee: +0.56 °C L-back thigh: +0.30 °C L-front leg: +0.36 °C |
None | 7 |
| 4 | L-iliopsoas muscle | None (could not be photographed) | - |
R-front thigh: +0.35–0.48 °C R-back thigh: +0.30–0.54 °C R-knee: +0.30–0.68 °C R-back foot: +0.36 °C |
None | 25 |
| 5 | R-iliopsoas muscle | None (could not be photographed) | - |
L-front thigh: +0.51 °C L-knee: +0.61 °C L-front leg: +0.41 °C L-back thigh: +0.39–0.60 °C L-popliteus: +0.35 °C L-back leg: +0.35–1.04 °C L-achilleas: +0.49 °C L-back foot: +0.44 °C |
None | 30 |
| 6 | R-brachial muscle | -0.02 °C |
R- front trapezius: +0.31 °C R-front wrist: +0.87 °C R-back wrist: +0.80 °C |
L-back trapezius: +0.34 °C L-elbow: +1.92 °C L-back forearm: +0.34 °C L-back stem: +0.48–0.50 °C |
None | 21 |
R, right; L, left
Table 4.
Details of the analysis of the increase in skin temperature in participants after orthopaedic surgery
| Participant | Affected area | Δ skin temp of the affected area | Δ skin temp of the surrounding area | Other haemophilic arthropathy and history of surgery | Days from the date of admission for operation to the date of the shooting | |
|---|---|---|---|---|---|---|
| Affected side (vertical pattern) |
Unaffected side (diagonal pattern) |
|||||
| 7 | L-clavicle | + 0.37 °C | L-front shoulder: +0.79 °C |
R-triceps: +0.42 °C R-front forearm: +0.60–0.81 °C R-back forearm: +0.89 °C R-front wrist: +0.92 °C R-back wrist: +1.18 °C |
None | 7 |
| 8 | R-shoulder | + 0.61 °C |
R-back trapezius: +0.70 °C R-outer rotators: +0.45 °C R-back elbow: +0.53 °C |
L-front elbow: +0.67 °C L-front forearm: +1.61 °C L-back forearm: +0.82 °C L-front wrist: +0.43 °C |
None | 10 |
| 9 | R-elbow |
+ 0.80–2.80 °C (R-front elbow: +0.80 °C, R-back elbow: +2.8 °C) |
R-biceps: +0.31 °C R-triceps: +0.73 °C R-front forearm: +0.82–1.01 °C R-back forearm: +0.82–0.83 °C R-front wrist: +1.66 °C R-back wrist: +0.45 °C |
L-front neck: +0.71 °C L-front trapezius: +0.53 °C L-back stem: +0.45 °C |
R-elbow | 12 |
| 10 | L-knee | + 1.67 °C |
L-gluteus: +1.19–2.15 °C L-front thigh: +1.46–2.29 °C L-front leg: +1.59–1.66 °C L-back leg: +0.54–0.55 °C L-achilleas: +0.59 °C L-anterior foot: +0.53 °C L-back foot: +1.62 °C |
R-front thigh: +0.60–1.49 °C R-back thigh: +1.05–2.17 °C |
Bilateral elbow Bilateral ankle |
13 |
R, right; L, left
Table 5.
Details of the increase in skin temperature in people with haemophilic arthropathy
| Participant | Affected area | Δ skin temp of the affected area | Δ skin temp of the surrounding area | Other haemophilic arthropathy and history of surgery | |
|---|---|---|---|---|---|
| Affected side (vertical pattern) |
Unaffected side (diagonal pattern) |
||||
| 11 | R-knee | -0.27 °C |
R-front thigh: +0.47–1.05 °C R-back thigh: +0.49 °C R-achilleas: +0.64 °C R-front leg: +0.67 °C R-back foot: +0.80 °C |
L-back thigh: +0.65–0.77 °C L-popliteus: +0.37 °C L-back leg: +0.38 °C L-anterior foot: +0.31 °C |
R-elbow R-hip L-knee |
| 12 | R-elbow | + 1.17 °C |
R-front stem: +0.32 °C R-back wrist: +0.38 °C |
L-front neck: +0.84 °C L-front shoulder: +0.40 °C |
None |
| 13 | R-ankle | + 0.54 °C |
R-front leg: +0.56 °C R- back thigh: +0.30 °C |
L-outer gluteus: +0.62–1.01 °C L-front thigh: +0.31–0.76 °C L-knee: +0.94 °C L-front leg: +0.64 °C L-back leg: +0.44 °C L-anterior foot: +0.57 °C L-back foot: +0.54 °C |
Bilateral hip Bilateral TKA L-ankle |
| 14 | R-ankle | + 1.30 °C |
R-achilleas: +0.44 °C R-anterior foot: +2.43 °C |
L-front thigh: +0.62–0.82 °C L-back thigh: +0.37 °C L-knee: +0.72 °C L-popliteus: +0.57 °C L-back leg: +0.46–0.50 °C |
None |
| 15 | R-ankle | + 1.02 °C |
R-knee: +1.05 °C R-popliteus: +0.30 °C R-front leg: +0.88–0.92 °C R-back leg: +0.67–0.74 °C R-achilleas: +1.43 °C R-anterior foot: +1.50 °C R-back foot: +1.46 °C |
L-front thigh: +0.36 °C L-back thigh: +0.40–0.90 °C |
L-elbow L-ankle |
| 16 | R-knee | + 0.73 °C | - |
L-front thigh: +0.51 °C L-back thigh: +0.31 °C L-front leg: +0.38–0.57 °C L-back leg: +0.38–0.84 °C L-ankle: +1.57 °C L-achilleas: +1.27 °C L-back foot: +0.52 °C |
None |
| 17 | L-knee |
+ 0.43–1.76 °C (L-patella: +0.43 °C, L-lower vastus medialis: +0.60 °C, L-lower rectus femoris: +1.76 °C, L-lower vastus lateralis: +1.55 °C) |
L-front thigh: +0.52–1.27 °C |
R-ankle: +0.63 °C R-anterior foot: +0.93 °C |
None |
| 18 | R-ankle | + 0.09 °C | R-front thigh: +0.46 °C |
L-back thigh: +0.32 °C L-back leg: +0.49 °C |
L-ankle |
| 19 | R-hip | None (could not be photographed) |
R-front thigh: +0.48–0.54 °C R-knee: +0.48 °C R-anterior foot: +0.59 °C |
L-front thigh: +0.67 °C L-popliteus: +0.92 °C L-back leg: +0.36–0.52 °C L-back foot: +1.57 °C |
None |
| 20 | L-knee | + 1.42 °C |
L-anterior thigh: +0.41–0.71 °C L-back thigh: +0.71–0.83 °C L-popliteus: +0.77 °C L-back leg: +1.06 °C L-anterior foot: +0.45 °C L-back foot: +0.55 °C |
R-front thigh: +0.41–0.71 °C R-front leg: +0.39 °C |
L-shoulder Bilateral elbow R-THA Bilateral TKA Bilateral ankle |
| 21 | L-knee | + 0.91 °C |
L-front thigh: +0.44 °C L-back thigh: +0.35–0.89 °C L-back leg: +1.03 °C L-back foot: +0.57 °C |
R-front thigh: +0.43–0.88 °C R-knee: +1.19 °C R-front leg: +0.69 °C R-ankle: +0.41 °C R-anterior foot: +0.74 °C |
L-shoulder Bilateral elbow Bilateral ankle |
| 22 | L-knee | + 0.47 °C |
L-front thigh: +0.43–0.74 °C L-front leg: +0.46 °C L-back foot: +0.45 °C |
R-back leg: +0.30 °C R-anterior foot: +1.29 °C |
Bilateral elbow Bilateral THA L- ankle arthrodesis R-ankle |
| 23 | L-ankle | + 0.39 °C | - |
R-gluteus: +0.38–0.53 °C R-back thigh: +0.37–0.66 °C |
L-elbow |
| 24 | L-knee | -0.07 °C | L-back thigh: +0.39–0.63 °C |
R-back thigh: +0.45 °C R-front leg: +0.66 °C R-back leg: +0.44 °C R-ankle: +1.65 °C R-achilleas: +1.77 °C R-anterior foot: +1.58 °C R-back foot: +2.21 °C |
Bilateral elbow R-knee Bilateral ankle |
R, right; L, left; THA, total hip arthroplasty; TKA, total knee arthroplasty
Among the 6 participants with after bleeding, 4 had intramuscular bleeding, 1 had intra-articular bleeding, and 1 had subcutaneous bleeding. Two cases involved iliopsoas haemorrhages and could not be photographed because the participants were wearing underwear. After those 2 participants were excluded, we found that 3 of the 4 remaining participants had a Δ skin temp higher than 0.30 °C in the affected area (median: +0.62 °C, maximum: +1.00 °C, minimum: +0.52 °C).
After orthopaedic surgery, all 4 participants had a Δ skin temp increase of more than 0.30 °C in the affected area (median: +0.80 °C, maximum: +2.80 °C, minimum: +0.37 °C).
Among the 14 participants with haemophilic arthropathy, 1 had haemophilic arthropathy in the elbow, 1 in the hip, 8 in the knee, and 4 in the ankle. One participant with hip haemophilic arthropathy could not be photographed because he was wearing underwear. After that participant was excluded, we found that 10 of the remaining 13 participants had a Δ skin temp increase of more than 0.3 °C in the affected area (median: +0.91 °C, maximum: +1.76 °C, minimum: +0.39 °C). Nine out of the 14 participants had haemophilic arthropathy in areas other than the affected area.
Representative data on the Δ skin temp in the affected area are shown in Fig. 2. Patient 1 had a haematoma in the left triceps surae muscle, with a temperature increase of 0.52 to 0.63 °C compared with that of the contralateral side. Patient 2 had a haematoma in the right ankle joint, with a temperature increase of 0.78 °C compared with that of the contralateral side. Patient 9 underwent arthroscopic synovectomy of the right elbow joint, with a temperature increase of 0.8 to 2.8 °C compared with that of the contralateral side. Patient 10 underwent total knee arthroplasty on the left knee joint, with a temperature increase of 1.67 °C compared with that on the contralateral side. Patient 14 had haemophilic arthropathy in the right ankle joint, with a temperature increase of 1.30 °C compared with that of the contralateral side. Patient 20 had haemophilic arthropathy in the left knee joint, with a temperature increase of 1.42 °C compared with that of the contralateral side.
Fig. 2.
Representative data on the temperature increase in the affected area. In Patients 1 and 2 (after bleeding), Patients 9 and 10 (postoperative orthopaedic surgery), and Patients 14 and 20 (haemophilic arthropathy), the area enclosed by the blue square represents the affected area. In all three types of cases, representative data revealed a Δ skin temp of 0.3 °C or greater in the affected area
Temperature increases in the surrounding area
We observed two typical patterns of temperature elevation among the study participants: (i) a pattern of temperature elevation ipsilateral to the affected area, more proximal or distal to the affected area (vertical pattern), and (ii) a pattern of temperature elevation contralateral to the affected area, more proximal or distal to the affected area (diagonal pattern).
In 6 participants following a bleeding event, a Δ skin temp of more than 0.30 °C in the vertical direction relative to the affected area was observed in 4 out of 6 participants (median: +0.68 °C, maximum: +1.43 °C, minimum: +0.32 °C). A Δ skin temp of more than 0.30 °C in the diagonal pattern relative to the affected area was observed in 5 out of the 6 participants (median: +0.48 °C, maximum: +1.92 °C, minimum: +0.30 °C).
After orthopaedic surgery, a Δ skin temp of more than 0.30 °C in the vertical and diagonal patterns relative to the affected area was observed in all the participants (vertical pattern: median: +0.82 °C, maximum: +2.29 °C, minimum: +0.31 °C; diagonal pattern: median: +0.81 °C, maximum: +2.17 °C, minimum: +0.42 °C).
In 14 participants with haemophilic arthropathy, a Δ skin temp of more than 0.30 °C in the vertical direction relative to the affected area was observed in 13 out of 14 participants (median: +0.59 °C, maximum: +2.43 °C, minimum: +0.30 °C). A Δ skin temp of more than 0.3 °C in the diagonal pattern relative to the affected area was observed in all the participants (median: +0.57 °C, maximum: +2.21 °C, minimum: +0.30 °C).
Representative data on Δ skin temp at the vertical and diagonal positions of the affected area are shown in Fig. 3. Participants 2, 10, and 20 exhibited a vertical pattern of temperature increase in the affected area. Participants 2, 10, and 20 are examples where a vertical pattern of temperature increase was observed in the affected area. In contrast, Participants 3, 8, and 13 are examples where a diagonal pattern of temperature increases was observed in the affected area.
Fig. 3.
Representative data on the temperature increase at the vertical and diagonal positions of the affected area. In Patients 2 and 3 (after bleeding), Patients 10 and 8 (postoperative orthopaedic surgery), and Patients 20 and 13 (haemophilic arthropathy), the area enclosed by the blue square is the affected area. The arrows indicate a Δ skin temp of 0.3 °C or more at the vertical and diagonal positions relative to the affected area. The cases in the upper and lower panels show the vertical pattern and diagonal pattern, respectively. A Δ skin temp of 0.3 °C or more was observed in the vertical and diagonal positions relative to the affected area enclosed by the blue square
Discussion
We previously reported that in people with haemophilia with new-onset pain or exacerbation of existing pain, an increase in the mean temperature of the painful area of more than 0.3 °C compared with that of the contralateral side is likely to indicate an acute-phase condition such as bleeding or inflammation [10]. In this study, we confirmed for the first time that in PwH with joint and/or muscle pain, Δ skin temp also occurs in asymptomatic areas surrounding the pain area. These areas of elevated temperature may have caused early damage and/or increased soft tissue or joint load; therefore, it is necessary to monitor for future injury and functional decline.
In previous studies that attempted to use infrared thermography (IRT) to detect injury sites based on increases in body surface temperature, it was reported that areas affected by conditions such as whiplash or fractures showed significant temperature elevations (>1 °C) [13, 14]. Higher values of markers of muscle damage were observed not only with high-intensity exercise but also with an increase in surface temperature detected by IRT, and a correlation was observed between these two parameters [15]. On the basis of these previous reports, we hypothesized that it would be possible to use body surface temperature to assess the presence of secondary inflammation and/or muscle strain in areas distant from the affected region in PwH with pain. The mechanism of Δ skin temp at remote areas relative to the affected area was inferred as follows. When there is bleeding or tissue damage, inflammatory mediators dilate blood vessels and increase local blood flow [16]. In addition, during muscle exercise, increased oxygen demand and sympathetic excitation cause vasodilation and increased local blood flow [14]. When blood flow increases, more warm blood is transported to the surface of the skin, where heat can be dissipated, leading to an increase in skin temperature.
Two typical patterns of temperature elevation observed in the study participants were the vertical pattern and the diagonal pattern. The raised temperature areas located away from the affected area were not symptomatic, and there was no other sign of haemorrhage. Given the aforementioned mechanisms of temperature increase, these areas are expected to be subjected to inflammation or muscular loading, which increases the oxygen demand. In the orthopaedic field, increased knee strain secondary to hip osteoarthritis is known as coxitis [17]. Osteoarthritis is a disease in which the cartilage in the joints wears away and the bones of the joints come into direct contact, causing pain and stiffness. The following mechanisms are considered responsible for increased knee strain in individuals with hip osteoarthritis: (i) Changes in gait occur, i.e., hip joint pain and limited range of motion result in gait changes and increased strain on the knee joint. Specifically, the knee is used excessively when walking, leading to progressive deformity and pain in the knee joint. (ii) Postural changes occur, i.e., hip joint disorders result in strain on other joints and muscles because of the need to balance the body. (iii) Muscle weakening occurs, i.e., hip joint pain results in restricted activity and weakening of the surrounding muscles, which impairs the stability of the knee joint. Weakened knee muscles increase the strain on the knee joint and increase the risk of knee osteoarthritis. (iv) Unilateral strain occurs i.e., if there is a deformity in one hip joint, more weight is placed on the opposite leg, which also places excessive strain on the knee joint. Like patients with osteoarthritis, PwH, whether their perceived pain is acute or chronic, may have an increased risk of secondary injury around the painful area owing to the increased load caused by pain-induced changes in gait and/or posture. For osteoarthritis of the hip, interventions to prevent injury and functional decline include physiotherapy for muscle strengthening and flexibility [18], proper footwear to reduce joint strain [19], and the use of assistive devices such as canes or walkers [20]. Similarly, in PwH, early implementation of such therapeutic interventions may help to avoid future injury. There are reports of the avoidance of injury in professional athletes through the use of infrared body surface temperature assessment, which is expected to be an effective adjunct to treatment decision-making in the field of disease [11]. Even when infrared assessment is not possible, careful observation is necessary to consider the possibility of increased loading in areas that are vertical or diagonal to the affected area.
Limitations
There are several limitations to this research. First, annual bleeding rates in PwH have declined in recent years due to advances in therapeutic agents [21], resulting in fewer opportunities to investigate the early detection of compensatory overload, even among PwH who present with symptoms unrelated to acute bleeding episodes. Second, as our method is designed to assess temperature differences between the left and right sides, a significant temperature increase on the affected side may prevent the detection of a slight temperature increase on the unaffected side. Consequently, the evaluation of the side opposite to a region with a pronounced temperature rise may be insufficient. Third, since the evaluation was conducted on patients who actually visited the hospital, there are diverse backgrounds, which may have influenced the results. However, despite the diverse backgrounds of these patients, temperature changes were observed at distant sites different from the sites where pain was felt in almost all cases, suggesting that our method may be useful for individual evaluation of such cases. Finaly, it is unknown whether the increased temperature of the surrounding area relative to the affected area caused subsequent pain and haemorrhage, as this study was only cross-sectional. This study will serve as a starting point for a longitudinal investigation to determine whether the surrounding area later becomes symptomatic.
Conclusions
The temperature increases, which may reflect the physical load, occurred in areas other than the painful area. When PwH present with pain, the surrounding area, especially in the diagonal and/or vertical position, may be under strain, even if it is asymptomatic. Therefore, considering options such as ultrasound examinations, rehabilitation therapy, or changes in treatment agents or administration frequency may help ensure the patient’s future physical function. At this stage, there are few indicators for effective intervention, but our IRT method may serve as a useful tool in this regard.
Acknowledgements
Medicinal Biology of Thrombosis and Haemostasis was established in collaboration between Nara Medical University and Chugai Pharmaceutical Co., Ltd. First, we thank the PwH who participated in the measurements. We thank American Journal Experts (AJE) for their help with English language editing.
Abbreviations
- IRT
Infrared thermography
- ROI
Region of interest
- Δ skin temp
Difference in skin temperature
Author contributions
Conceptualization: YM, AS, MS, and KN. Data curation: YM, AS, CH, NM, RK, KO, SF, NS, TS, KT, and YI. Formal analysis: YM, AS, CH, NM, and RK. Investigation: YM, AS, CH, KO, SF, NS, and YI. Methodology: YM, AS, CH, RK, and NM. Project administration: AS, MS, TS, KT, AK, and KN. Resources: YM, MS, CH, KO, SF, NS, YI, TS, KT, AK, and KN. Supervision: TS, AK and KN. Visualization: YM, AS, and MS.Writing -original draft: YM, AS, MS, CH, and NM. Writing -review and editing: All the authors.
Funding
Chugai Pharmaceutical Co., Ltd.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethical approval
All participants were given a clear and comprehensive explanation of the study on the basis of the Declaration of Helsinki. Subsequently, participants signed informed consent forms. Ethical approval for the studies was obtained from both the Nara Medical University (ID: 2097) and Chugai Ethics Committees (ID: E19086).
Consent for publication
No identifiable subject information is provided in the manuscript, figures, or tables.
Competing interests
YM, AS, MS, NM, RK, TS, KT, and KN are members of the Medicinal Biology of Thrombosis and Haemostasis established by Nara Medical University and Chugai Pharmaceutical Co., Ltd. NM, RK, and TS are employees of Chugai Pharmaceutical Co., Ltd. NM and RK have stock ownership of Chugai Pharmaceutical Co., Ltd. MS is a representative of the Medicinal Biology of Thrombosis and Haemostasis collaborative research laboratory; research support from Chugai Pharmaceutical Co., Ltd.; Takeda Pharmaceutical Co., Ltd.; and CSL Behring; received honoraria or consultation fees from Chugai Pharmaceutical Co., Ltd.; is a speaker for Chugai Pharmaceutical Co., Ltd.; and CSL Behring, Sanofi, Bayer, Novo Nordisk Pharma, Takeda Pharmaceutical Co., Ltd., Pfizer, and Fujimoto Seiyaku Corp.; and has patents for inventions relating to products of Chugai Pharmaceutical Co., Ltd.CH has no conflict of interest to decare. KO has received personal fees from Chugai Pharmaceutical Co., Ltd., and CSL Behring. SF taught a course endowed by CSL Behring and received personal fees from Chugai Pharmaceutical Co., Ltd., CSL Behring, Takeda Pharmaceutical Co., and Sanofi S.A.NS taught a course endowed by CSL Behring.KT received grants or research support from the Japan Blood Products Organization, The Mother and Child Health Foundation and Novo Nordisk Pharma. YI received specified clinical trial funding from MediaMart Co., Ltd.; research funding from Chugai Pharmaceutical Co., Ltd., Johnson & Johnson K.K., KM Biologics, Olympus Terumo Biomaterials Corp., and Zimmer Biomet; and honoraria from Asahi Kasei Pharma Corporation, Bayer Yakuhin Ltd., Chugai Pharmaceutical Co., Ltd., CSL Behring, Healious K.K., Hisamitsu Pharmaceutical Co., Johnson & Johnson K.K., Ono Pharmaceutical Co., Ltd., Novo Nordisk Pharma Ltd., Sanofi K.K., and Takeda Pharmaceutical Co., Ltd.; and is a medical advisor for Chugai Pharmaceutical Co., Ltd. and KM Biologics.AK received personal fees from Hisamitsu Pharmaceutical Co. and Tsumura & Co. KN received grants, personal fees, and nonfinancial support from Chugai Pharmaceutical Co., Ltd.; personal fees from Hoffmann-La Roche Ltd.; grants and personal fees from Sysmex Co., SEKISUI MEDICAL., Takeda Pharmaceutical Co., Sanofi S.A., Co., Ltd., CSL Behring Co., KM Biologics Co., Ltd., Novo Nordisk A/S, Bayer AG, and Fujimoto Seiyaku; is an inventor of patents relating to emicizumab; and is a member of the Medicinal Biology of Thrombosis and Haemostasis established by Nara Medical University and Chugai Pharmaceutical Co., Ltd.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Shared first authorship: Yuya Mawarikado and Asuka Sakata.
References
- 1.Bolton-Maggs PH, Pasi KJ. Haemophilias A and B. Lancet. 2003;361(9371):1801–9. [DOI] [PubMed] [Google Scholar]
- 2.Srivastava A, et al. WFH guidelines for the management of hemophilia, 3rd edition. Haemophilia. 2020;26(Suppl 6):1–158. [DOI] [PubMed] [Google Scholar]
- 3.Melchiorre D, Manetti M, Matucci-Cerinic M. Pathophysiology of hemophilic arthropathy. J Clin Med. 2017;6(7):63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lafeber FP, Miossec P, Valentino LA. Physiopathology of haemophilic arthropathy. Haemophilia. 2008;14(Suppl 4):3–9. [DOI] [PubMed] [Google Scholar]
- 5.Zhang T, et al. Clinical outcomes of arthroscopic synovectomy for adolescent or young adult patients with advanced haemophilic arthropathy. Exp Ther Med. 2018;16(5):3883–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Song SJ, et al. Mid-term outcomes and complications of total knee arthroplasty in haemophilic arthropathy: a review of consecutive 131 knees between 2006 and 2015 in a single Institute. Haemophilia. 2018;24(2):299–306. [DOI] [PubMed] [Google Scholar]
- 7.Ahn J, Yoo MC, Seo J, Park M, Jeong BO. Comparison of total ankle arthroplasty and ankle arthrodesis in end-stage hemophilic arthropathy. Foot Ankle Int. 2020;41(8):937–44. [DOI] [PubMed] [Google Scholar]
- 8.Sillero-Quintana M, et al. Infrared thermography as a support tool for screening and early diagnosis in emergencies. J Med Imaging Health Inf. 2015;5(6):1223–8. [Google Scholar]
- 9.Selfe J, Whitaker J, Hardaker N. A narrative literature review identifying the minimum clinically important difference for skin temperature asymmetry at the knee. Thermol Int. 2008;18:51–4. [Google Scholar]
- 10.Kawasaki R, et al. The use of infrared thermography for non-invasive detection of bleeding and musculoskeletal abnormalities in patients with hemophilia: an observational study. Thromb J. 2023;21(1):70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gómez-Carmona P, Fernández-Cuevas I, Sillero-Quintana M, Arnaiz-Lastras J, Navandar A. Infrared thermography protocol on reducing the incidence of soccer injuries. J Sport Rehabil. 2020;29(8):1222–7. [DOI] [PubMed] [Google Scholar]
- 12.Fernández-Cuevas I, et al. Classification of factors influencing the use of infrared thermography in humans: a review. Infrared Phys Technol. 2015;71:28–55. [Google Scholar]
- 13.Castonguay T, Dover G. Infrared thermography-a novel tool for monitoring fracture healing: a critically appraised topic with evidence-based recommendations for clinical practice. J Sport Rehabil. 2023;32(7):834–9. [DOI] [PubMed] [Google Scholar]
- 14.Lee YS, et al. The effectiveness of infrared thermography in patients with whiplash injury. J Korean Neurosurg Soc. 2015;57(4):283–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Alburquerque Santana PV, et al. Relationship between infrared thermography and muscle damage markers in physically active men after plyometric exercise. J Therm Biol. 2022;104:103187. [DOI] [PubMed] [Google Scholar]
- 16.Ashina K, et al. Histamine induces vascular hyperpermeability by increasing blood flow and endothelial barrier disruption in vivo. PLoS ONE. 2015;10(7):e0132367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hauge MF. The knee in patients with hip joint ankylosis. Clinical survey and bio-mechanical aspects. Acta Orthop Scand. 1973;44(4):485–95. [DOI] [PubMed] [Google Scholar]
- 18.Poquet N, Williams M, Bennell K. Exercise for osteoarthritis of the hip. Phys Ther. 2016;96(11):1689–94. [DOI] [PubMed] [Google Scholar]
- 19.Paterson KL, et al. Shoes for self-managing chronic hip pain: the SCHIPP randomized clinical trial protocol. BMC Musculoskelet Disord. 2023;24(1):141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Fernandes L, et al. EULAR recommendations for the non-pharmacological core management of hip and knee osteoarthritis. Ann Rheum Dis. 2013;72(7):1125–35. [DOI] [PubMed] [Google Scholar]
- 21.Hassan S, et al. Health and treatment outcomes of patients with hemophilia in the Netherlands, 1972–2019. J Thromb Haemost. 2021;19(10):2394–406. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
No datasets were generated or analysed during the current study.