Effects of Repetitive Shortwave Diathermy for Reducing Synovitis in Patients With Knee Osteoarthritis: An Ultrasonographic Study

Subjects

All subjects were referred to the study from the outpatient department by orthopedic surgeons. Subjects had a degenerative osteoarthritic knee of grade 3 or less based on radiographic diagnosis in the Kellgren and Lawrence grading of osteoarthritis,²⁰ had no limitation of range of motion except for minimum tightness in the knee joint, did not engage in any high joint-loading exercises^21,22 such as hiking or tennis playing, and had not undergone any specific treatments 3 months before entering the study. Participants were excluded from the study after on-site evaluation if they presented other musculoskeletal problems associated with the knee joint, such as fracture, tendon or ligament tears, or meniscus injury; musculoskeletal problems associated with the hip or ankle/foot joints; or central or peripheral neuropathy.

Initially, 63 patients fulfilled the above criteria from January 2002 to January 2003. However, 15 patients were unwilling to participate in this study, and 12 patients were unable to be present for SW therapy regularly; therefore, a total of 36 patients participated in this study. All participants signed consent forms before the experiment.

After being instructed in the study protocol, the participants were divided into 3 study groups, as determined by the participants’ own decision rather than random assignment. Fourteen participants received SW diathermy, 13 participants received SW diathermy and nonsteroidal anti-inflammatory drugs (NSAIDs), and 9 participants who did not receive any treatment but were willing to be present for follow-up served as a control group. Participants in the SW and control groups did not receive any medicine or treatment during the experiment period, but those in the SW+NSAIDs group, were allowed NSAIDs during the experiment period as needed. At the end of the experiment, 3 participants from each of the 2 treatment groups were unable to attend the last follow-up examination due to personal reasons not related to knee pain. Therefore, there were 30 participants for whom complete data were available, 14 of whom had bilateral osteoarthritis of the knee. Thus, a total of 44 knee joints of 30 subjects (6 men and 24 women) with knee osteoarthritis were investigated in this study. Baseline subject characteristics are listed in the Table ⇓.

Assessments

Ultrasonographic imaging examination

Ultrasonography was performed on the osteoarthritic knees at the initial and 3 follow-up sessions using the HDI 5000 ultrasound system.* A 12-5 MHz broadband linear-array transducer (L12-5 38 mm) was used, and all parameters were set at ranges suitable for assessment of the knee joint.²³

The measurement technique used to assess synovial sac thickness of knee joints was similar to the method described by van Holsbeeck et al,²⁴ except that no pressure was added on the examined area during the testing process. Van Holsbeeck et al applied some pressure on the examined area in order to expel the joint fluid apart from the synovial sac. However, in our experience, the pressure the examiner applies may not be the same for each measurement or for each patient, which may influence in the accuracy of the measurement. Therefore, we measured the synovial sac thickness without allowing any contact between the ultrasonic probe and the skin (ie, the measurement of synovial sac thickness included the thickness of both synovium and synovial fluid).

Participants were in a relaxed supine position during the whole measurement process. The tested leg was stabilized on a tripod with the knee joint at 30 degrees of flexion, which was confirmed with a universal goniometer (Fig. 1A). In such a position, overstretching of the synovial sac was avoided, and the patella did not overlie the suprapatellar recess and thus did not affect the measurement. Markers were made on bony prominences around the patella. The investigator (LYT) placed the water-soluble acoustic gel onto the examined area, and then gently placed the ultrasonic probe onto the gel without making contact with the patient’s skin. The position of the ultrasonic probe varied with assessment sites. When the suprapatellar recess was examined, the ultrasonic probe was placed to get a longitudinal view. It was placed centrally and superiorly to the patella, with the tail end near the proximal top of the patella. When either the medial or lateral parapatellar recess was examined, the ultrasonic probe was placed with the tail end lying in the middle trisection of the patella in order to have a longitudinal view. Throughout the examination process, the ultrasonic probe remained perpendicular to the skin surface, and the thickest part of the synovium on the image was taken (Fig. 1B). These images then were saved immediately on the hard disk of a desktop computer.

Figure 1.

(A) Longitudinal view of the suprapatellar recess. The probe is positioned above the patella with the patient in a supine position and the knee in 30 degrees of flexion. (B) Sonographic view of knee joint synovitis in a patient with osteoarthritis of the knee. We measured the largest anteroposterior diameter of the suprapatellar recess between the arrows. No compression was allowed in this picture. P=patella, f=fatty tissue, Q=quadriceps tendon, F=femoral cortex.

Intervention Protocol

The intervention protocol for 2 treatment groups consisted of 30 sessions of 20-minute induction-coil SW diathermy therapy.²⁶ The participants were positioned supine and comfortably on the treatment plinth with the affected knee extended. A towel was wrapped around the knee joint, and then the induction coil cable was applied circularly along the affected leg. All participants received treatments from the same SW diathermy machine.† The intensity of the current was set based on each participant’s sensation of warmth (a mild but pleasant sensation of heat).^27,28 Each treatment session lasted for 20 minutes. In cases where bilateral knees were involved, SW diathermy was applied to each knee separately and the total treatment time for both knees was extended to 40 minutes.

Experimental Procedure

Before the experiment, the ultrasonographic images of all 30 participants (44 osteoarthritic knees) were examined for synovial sac thickness, and a pain index assessment also was undertaken. After this initial examination, 21 participants (28 osteoarthritic knees) in the 2 treatment groups received SW diathermy 3 to 5 times a week for a total of 30 treatment sessions within 8 weeks. Follow-up evaluations were performed on the day after the 10th, 20th, and 30th treatment sessions. For the control group, the same series of evaluations were performed every 2 to 3 weeks after the initial evaluation.

Data Analysis

Although synovial sac thickness was measured at 3 different areas around the patella, the total synovial sac thickness was calculated as the sum of the thickness of the synovial sac located superiorly, medially, and laterally to the patella. For each outcome variable, including both total synovial sac thickness and pain index, each patient underwent measurements before the experiment and the day after the 10th, 20th, and 30th treatments, respectively. To control individual variance at the time of initial status assessment, the total synovial sac thickness for each measurement was normalized by that of the initial status and expressed as the percentage of the initial status (% initial). In addition, the difference in pain index for each measurement from the initial status was calculated and used for comparison of improvement in knee pain.

SPSS 10.0 statistical software‡ was used for all statistical analyses. One-way analysis-of-variance (ANOVA) tests were performed for participants’ age, sex, height, weight, duration of knee osteoarthritis, initial total sac thickness, and pain index to examine the differences among the 3 study groups. A 3 (group) × 4 (measurement) ANOVA for repeated measures was used to analyze the main effect of different study groups and measurements as well as their interactions in total synovial sac thickness. A P value of .05 or below was considered significant for all statistical analyses. If the interaction effect of study group and measurement was significant, an ANOVA for repeated measures was performed separately for each study group, and multiple comparisons using a Bonferroni adjustment were applied to examine the differences between measurements (P<.0083). If any statistical significance was noted in the main effect of study group, Bonferroni post hoc multiple comparisons were performed to compare the differences in the changes of those outcome variables among the 3 study groups (P<.017). The Friedman test was used to test for pretreatment-posttreatment differences in pain scale ratings for all 3 groups. The Kruskal-Wallis rank test was used to compare the pain scale difference between the initial (baseline) and follow-up assessments to determine whether any significance existed. In addition, the relationship between the total synovial sac thickness and pain index was examined by Spearman correlation coefficient.

Changes in Total Synovial Sac Thickness

Changes in total synovial sac thickness for the 3 study groups for the 4 measurements are presented in Figure 2. After 10 sessions of SW diathermy, the synovial sac thickness in both treatment groups decreased to approximately 81% to 84% of the initial thickness; after 30 treatment sessions, it was approximately 67% to 72% of the initial thickness. The synovial sac thickness in the control group, however, did not change during the follow-up period.

Figure 2.

Changes in total synovial sac thickness (% initial) at different sessions of treatment. The error bars represent standard deviations for each group. SW=shortwave diathermy, NSAIDs=nonsteroidal anti-inflammatory drugs.

The 3 × 4 ANOVA for repeated measures revealed a significant interaction effect of study group and treatment session (F_6,123=14.55, P<.0001). Therefore, reduction in total synovial sac thickness in different treatment sessions was observed among the 3 study groups, which supports the stated hypotheses. Furthermore, analysis of the data revealed that both treatment groups showed a significant decrease in total synovial sac thickness with more sessions of SW diathermy compared with the control group (F_1,28=36.12 and F_1,28=54.16, respectively; both P<.0001). Such a difference, however, did not appear between the treatment groups (F_1,26=.031, P>.05, power=.059), suggesting that SW diathermy decreases synovial sac thickness regardless of whether NSAIDs are used.

Further analysis of the data showed that the decrease in synovial sac thickness in both treatment groups was significantly greater after 10 sessions of SW diathermy (F_1,13=21.24 and F_1,13=24.97, respectively; both P<.0005). Such changes continued to take place after 20 and 30 treatment sessions (all P<.0001). Although total synovial sac thickness for both treatment groups continued to decrease till the end of 30 treatment sessions, the difference in total synovium sac thickness between 20 and 30 treatment sessions was not statistically significant (P>.0083).

Decrease in Pain Index

Decreases in pain index for 3 study groups for the 4 measurements are presented in Figure 3. After 10 sessions of SW diathermy, the pain index in the SW group and the SW+NSAIDs group decreased approximately 1.8±1.7 and 1.5±1.3, respectively. The pain index continued to decrease after 20 or 30 sessions of treatment. At the end of 30 sessions of SW diathermy, the decrease in pain index was approximately 5.0±2.0 and 4.1±2.4 from the initial status, respectively. The pain index in the control group remained the same for each follow-up session.

Figure 3.

Changes in pain index at different stages after shortwave treatment. The error bars represent standard deviations for each group. SW=shortwave diathermy, NSAIDs=nonsteroidal anti-inflammatory drugs.

Similarly, the Friedman test revealed a significant interaction effect of study group and treatment session on pain index (P<.0005). This finding indicates that reduction in pain index with different sessions of SW diathermy differed among the 3 study groups. Both treatment groups showed greater reduction in pain index than the control group (P<.005), but there was no significant difference in pain reduction between the treatment groups at each follow-up examination (P>.05). The pain index in both treatment groups was decreased significantly after 10, 20, and 30 treatment sessions (all P<.005). Pain improved significantly between each 2 examinations (P<.0083), except between the 10th and 20th treatment session follow-ups and the 20th and 30th treatment session follow-ups. However, participants in the control group did not experience any changes in their pain index (all P>.0083).

Correlation Between Total Synovial Sac Thickness and Pain Index

A total of 176 measurements were carried out on 44 osteoarthritic knees to examine the probable correlation between the total synovial sac thickness and the pain index. A small but significant correlation was found between these 2 variables (Spearman ρ=.174, P<.05). The Spearman correlation test also was performed to analyze the relationship between the total decreased pain index and the total decreased synovial sac thickness from the baseline measurements. We found that the greater the reduction in synovial sac thickness, the greater the decrease in pain index (Spearman ρ=.600, P<.0001).

Shortwave Diathermy Promoting Synovial Inflammatory Process

Compared with the control group, synovial sac thickness was reduced in both treatment groups after more than 10 sessions of SW diathermy. The results of this study demonstrated that SW diathermy therapy successfully decreases synovial sac thickness and suggest that SW diathermy can control synovial inflammation. The possible physiological mechanisms underlying such significant changes may be associated with improvement in circulation of vascular network in synovial membrane.²⁹ It has been found that synovial vascular density is lower in patients with chronic synovitis, adversely affecting the flow of blood and lymph fluid.^29–32 It also has been documented that SW diathermy may result in vascular dilation and increased blood flow.^3,11,14 Consequently, through the application of SW diathermy, the inflammatory response in the synovium of the knee joint is reduced and both synovial thickness and pain are decreased.

In this study, participants in the SW+NSAIDs group did not show better results than those in the SW group. Whether this finding was due to the participants’ nonadherence to medication use or to other causes was not possible to assess in this study. No attempt was made to follow up on their medication adherence, for either dosage or frequency that the medication was used. Further research is needed to determine the reasons for the lack of a difference between the treatment groups, but patients’ agreement to use medicine should be considered.

In the present study, total synovial sac thickness, including the thickness of both synovium and synovial fluid, was examined instead of synovium thickness only, which is different from the van Holsbeeck et al method.²⁴ Karim et al¹⁸ observed that increased synovial fluid could be detected using ultrasonography in patients with osteoarthritis and confirmed by arthroscopic findings with increased vigilance for synovitis. It is reasonable not to neglect the effect of thickness of synovial fluid for synovitis.

More Treatment Sessions Resulting in Better Improvement

This study demonstrated a decrease in synovitis with increased treatment sessions in patients with knee osteoarthritis. As treatments were repeated, the synovial sac became thinner and the amount of pain experienced by patients became lower, although the synovial thickness did not significantly change between 20 and 30 treatments. We do not think that the effect of treatment plateaued. However, we believe that the small sample size influenced the statistical results. The positive finding of SW diathermy is consistent with the results of previous studies that examined the therapeutic effects of SW diathermy.^11,13,15 This improvement in symptoms may be due to increased circulation around the knee joint and reduced retention of synovial fluid, which became more evident after a series of SW diathermy treatments. The results of this study represent direct evidence of the effect of SW diathermy on synovial sac thickness.

It is not surprising that single-session SW diathermy is not able to reduce synovitis in people with knee osteoarthritis immediately. In a preliminary study, we assessed the synovial sac thickness of 3 patients with knee osteoarthritis immediately, 30 minutes, and 60 minutes after single-session SW diathermy. The results showed that the synovial sac thickness around the patella remained unchanged in all 3 patients, indicating that single-session SW diathermy has no immediate effect on synovial sac thickness. In contrast, Jan and coworkers found increased cutaneous tissue temperature measured by infrared thermography³³ and increased blood flow using Doppler ultrasonic imaging¹⁴ after a single session of SW diathermy. Therefore, the mechanism of therapeutic effect between single- and multi-session SW diathermy may be different. We hypothesized that the immediate effect of SW diathermy in patients with knee osteoarthritis may arise from improved circulation after treatment, whereas the long-term effect may be associated with the reduction in synovial inflammation. Thus, a single SW diathermy session cannot detect an immediate change in synovial sac thickness. Further study is needed to determine whether a reduction in synovial thickness might be noted after a longer period has elapsed after a single session of treatment.

Following 30 sessions of SW diathermy, although the reduction of the synovial sac thickness was significant, some participants still experienced some knee pain. We hypothesized that SW diathermy can reduce the synovial sac thickness, but that it might be not relieve patients’ pain completely.

Ultrasonography Is a Useful Tool to Detect Changes in Synovial Sac Thickness

The results of this study show that ultrasonography is a useful tool to quantitatively detect changes in synovial thickness in the knee joint following SW diathermy therapy. The validity and reliability of data obtained with the ultrasonographic imaging technique in determining the synovial sac thickness in patients with inflammation of the knee joint has been established in previous studies. In the studies by Karim et al¹⁸ and van Holsbeeck et al,²⁴ ultrasonographic imaging techniques were shown to yield valid and reliable data as well as more accurate data than clinical examination only, although only examination of thickness of the synovium, rather than the whole synovial sac, was performed. Tsai et al,³⁴ using the same measurement protocol used in our study, also demonstrated high interrater and intrarater reliability of measurements of synovial sac thickness obtained with ultrasonography in patients with knee osteoarthritis. In the present study, a series of examinations was applied to the same patient in order to evaluate the effect of the SW diathermy and successfully demonstrated the changes in synovial sac thickness. It may be concluded that ultrasonographic imaging is a valuable tool in the evaluation of the severity and progression of synovitis.

In addition, the measured suprapatellar synovial sac thickness in patients with degenerative osteoarthritis in this study was 5.3±1.6 mm (X̅±SD), which is slightly less than that measured by Rubaltelli and colleagues¹⁷ in patients with rheumatoid arthritis (6.9±1.9 mm). This finding was consistent with the results of a histological study³⁵; that is, thickening of the synovial sac is less pronounced in degenerative osteoarthritis than in rheumatoid arthritis. These findings are consistent with clinical presentation of synovitis and further validate the use of ultrasonographic imaging in the measurement of synovial sac thickness.

Limitations of the Study

There are some limitations of this study including the lack of random assignment of participants to study groups, no monitoring of pain medication usage, and the fact that the control group had significantly less pain than those in the treatment groups. Perhaps the major concern of the first limitation is that the assignment of participants to study groups was dependent on the patients’ own decision rather than being randomly assigned. In addition, patients with severe knee pain would hardly choose to get into the control group, which received no SW diathermy. We agree that the motivation of patients to receive SW diathermy treatment might have been a confounding factor in this study. This factor warrants more attention in future research.

This study highlights the need to explore the relationship between ultrasonographic image and histological finding of synovitis. Future studies are needed to examine the cost-effectiveness of using ultrasonography for assessing the disease progress compared with other assessment techniques, such as magnetic resonance imaging, and the treatment effect of SW therapy combined with other interventions on control of synovitis.