Article Text


Clinical, radiographic, and thermographic assessment of osteoarthritis in the knee joints
  1. H Warashina,
  2. Y Hasegawa,
  3. H Tsuchiya,
  4. S Kitamura,
  5. K-I Yamauchi,
  6. Y Torii,
  7. M Kawasaki,
  8. S Sakano
  1. Department of Orthopaedic Surgery, Nagoya University School of Medicine, Nagoya, Aichi, Japan
  1. Correspondence to:
    Dr H Warashina, Department of Orthopaedic Surgery, Nagoya University School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya, 466–8550, Japan;

Statistics from

It has been reported that thermography enables the evaluation of non-steroidal anti-inflammatory drugs for the treatment of rheumatic synovitis.1,2 However, few reports mentioned the relation between skin temperature and symptoms, physical signs, and radiographic features in patients with osteoarthritis (OA). In this study we evaluated the correlation between radiographic and clinical findings and skin temperature measured by thermography for the assessment of OA in the knee joints in the Comprehensive Health Examination Programme at Y-town, Japan in 1999.3


The assessments were performed on 974 knees of 169 men and 318 women who had a mean (SD) age of 57.6 (4.3). The following physical examinations of the knees were carried out: local tenderness, joint swelling, local heat, McMurray test, Lachmann test, instability of collateral ligaments, patellofemoral joint (PFJ) grinding test, range of joint motion, and thigh girth. Walking and stair climbing ability were graded using the Hospital for Special Surgery (HSS) knee rating scores and Lysholm knee scoring scale.4 Present or past pain continuing for more than one month was recorded.

The distance between the medial and lateral tibiofemoral joints (MTFJ and LTFJ) was measured on the anteroposterior radiographs. Radiographic scoring of osteophytes (0, none; 1, doubtful; 2, minimal; 3, moderate) at four sites (medial and lateral condyle of both tibia and femur) was recorded.5

The thermographic index (TI) and heat distribution index (HDI) were obtained by thermographic apparatus (TH5108ME, NEC Medical Systems, Japan). The TI has been defined as the average surface temperature of local skin area6 and HDI as ±1SD from the mean surface temperature.7 The whole anterior knee of 10 cm2 was assessed for HDI. Three different areas of 3×3 cm were studied for TI: Tm, TI of MTFJ; Tl, TI of LTFJ; and Tp, TI of 10 cm above the patella. Tp was considered as the control value. To minimise the effect of the variation of individual skin temperature, normalised Tm and Tl (ΔTm and ΔTl ) were calculated by subtracting Tp from Tm and Tl (ΔTm=Tm−Tp, ΔTl=Tl−Tp). High ΔTm means high skin temperature on MTFJ and high ΔTl means high skin temperature on LTFJ. Residents with severe knee deformity, rheumatoid arthritis, vascular disease, positive sign of PFJ grinding test were excluded, because we focused on the OA change of the MTFJ and LTFJ in this study.

Statistical analyses were carried out by unpaired Student’s t test, Pearson’s correlation coefficient test, and Kruskal-Wallis test. A p value of <0.05 was considered significant.


The thermographic and radiographic findings showed that narrowing of the MTFJ space was significantly associated with a rise in skin surface temperature (p<0.01) (table 1). ΔTm correlated significantly with the size of osteophytes at all sites measured, and ΔTl correlated at two of four sites (fig 1). These data indicate that residents with larger osteophytes in the MTFJ have higher skin surface temperature. We could not find any correlation between HDI and the radiographic findings.

Table 1

Correlation between radiographic, clinical assessments, and normalised TI (ΔTm, ΔTl)

Figure 1

Correlation between the site of the osteophytes and the normalised TI (ΔTm, ΔTl). Upper figure: ΔTm and osteophytes in MTFJ and LTFJ; lower figure: ΔTl and osteophytes in MTFJ and LTFJ. Larger osteophytes at each site of the MTFJ have a higher skin surface temperature.

The correlation between skin temperature and clinical symptoms showed that the residents with present or past pain have significantly higher ΔTm or ΔTl (p<0.01). We also found that many clinical findings correlated with the increase of skin temperature indicated by normalised TI (ΔTm and ΔTl ), as shown in table 1. HDI on the knee only correlated with joint swelling and tenderness (p<0.01).


In this study we used HDI and normalised TI to indicate the skin surface temperature. The normalised TI correlated more significantly with the clinical and radiographic parameters than the HDI. Therefore, it was suggested that the normalised TI is more useful for evaluation of the disease activity than the HDI. Further study is needed to confirm the accuracy and sensitivity of normalised TI for the screening of OA.

In the clinical situation raised skin temperature is sometimes seen as local heat in patients with OA in the knee joints. However, as far as we know no published report has assessed the relation between skin temperature and clinical or radiographic findings in OA knees. In this study, thermography provided objective data of skin temperature that was compared with other physical parameters, showing a correlation between skin temperature and clinical, and radiological evidence of OA in the knee joints. Synovial inflammation and subchondral change have been reported to affect the OA cascade.8–11 Dieppe et al used scintigraphy to show that subchondral bone metabolism is related to the progression and symptoms of OA in the knees.12 It is suggested that the skin temperature evaluated by thermography may be due to the increased synovitis or subchondral bone activity, or both, in the OA knee joint.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.