Knee osteoarthritis (KOA) is one of the most common chronic progressive diseases, with a prevalence of 30%–40% among older adults over 65 years of age, posing a serious health risk (Helmick et al., 2008). KOA is associated with degenerative destruction of articular cartilage and bone, leading to knee pain (Glyn-Jones et al., 2015), impaired proprioception (Segal et al., 2010), and reduced functional performance (Hawker, 2019).

Pain is the predominant and most common clinical symptom of KOA, which was believed to be caused by the uneven load at the knee and the increased pressure on the tibiofemoral joint (Beaupré et al., 2000). Pain has been commonly assessed by the Western Ontario and McMasters Osteoarthritis Index (WOMAC) (Sisante et al., 2020). Proprioception is closely related to the developmental process of KOA (van der Esch et al., 2007), and the impairment of proprioception of the knee and ankle joints led to decreased lower extremity joint balance and coordination among older adults with KOA (Hassan et al., 2001; Knoop et al., 2011). Although KOA affects mostly the knee joint, ankle proprioception is closely related to balance control (Han et al., 2015). Compared with their healthy counterparts, older adults with KOA have worse functional performance, which affects their ability of daily activities and increases the risk of falls (Anderson et al., 2019). Timed up & Go (TUG) (Podsiadlo and Richardson, 1991; Dobson et al., 2012) and 20-m walk tests (Motyl et al., 2013) have been used effectively to quantify functional performance among older adults with KOA.

Older adults with KOA typically receive surgical, pharmacological, and non-pharmacological treatments (Brophy and Fillingham, 2022). Surgery is generally used in the late-stage and can improve symptoms of KOA (Xue et al., 2022), however, with a high cost (Serikova-Esengeldina et al., 2022); Non-steroidal antiinflammatory, opioids, or nutritional drugs are usually used in pharmacological treatments (Persson et al., 2018), and side effects on the liver and kidneys have been reported (Zhang et al., 2019b). Non-pharmacological treatments have been recommended as the primary treatments for KOA (Brophy and Fillingham, 2022). Among them, therapeutic exercise (TE), such as resistance training or joint mobilization, has the potential effects to relieve pain (Juhl et al., 2014), recover proprioception (Lai et al., 2018), and improve functional performance (Fransen et al., 2002). Resistance training improves muscle strength and functional performance (de Rooij et al., 2017). Joint mobilization balances the load at the knee and relieves pain (Chen H. et al., 2019), as well as improves proprioception by enhancing sensory input to the central nervous system about postural movements (Saunders et al., 2005). However, a previous meta-analysis study showed that the effect size of TE remained small to moderate (Fransen et al., 2015). Therefore, there is room to improve its effectiveness on KOA.

Traditional Chinese herb hot compress (TCHHC) has been widely used in the treatment of chronic diseases such as KOA (Yuan et al., 2015). It fully synergizes the warming effect, which has been shown to slightly improve pain, functional performance and quality of life in older adults with KOA (Aciksoz et al., 2017), and the pharmacological effect, which reduces inflammation for pain relief by promoting synovial fluid flow and reducing the release of inflammatory substances into the synovial fluid (Wang et al., 2010). Combined with the above-mentioned effects of resistance training and joint mobilization, TCHHC + TE has the potential to relieve pain, improve proprioception, and functional performance.

To the best of our knowledge, no studies have explored the effects of TCHHC + TE on pain relief, proprioception recovery, and functional performance improvement among older adults with KOA. It was hypothesized that compared with TE, TCHHC + TE was superior in relieving pain, recovering proprioception, and reducing the times of TUG and 20-m walk tests.

No side effects were observed during the intervention, e.g., pain exacerbation, allergic reaction, redness, or heat burn due to the hot compress on the superficial joints. The Shapiro-Wilk test confirmed that all dependent variables were normally distributed. Independent t-tests showed no significant differences in age (p = 0.794), weight (p = 0.962), height (p = 0.467), body mass index (p = 0.841), and leg length (p = 0.639) between the two groups (Table 2).

The pain score was presented in Table 3. Significant interactions were detected in 2 items (walk on the ground: p = 0.009, η2_p = 0.245 and go up or down stairs: p = 0.006, η2_p = 0.266) of the WOMAC scores, and the total score (p = 0.047, η2_p = 0.148). At week 9, the go up or down stair (p = 0.002, d = 1.327) and total (p = 0.035, d = 0.798) scores were lower in the TCHHC + TE group compared with those in the TE group. Compared to week 0, the sleep at night (p < 0.001, η2_p = 0.395), sit or lying (p = 0.001, η2_p = 0.364), and stand straight (p < 0.001, η2_p = 0.424) scores were lower at week 9 in both groups.

The proprioceptive thresholds were presented in Table 4. Significant interactions were detected in proprioception thresholds of knee extension (p = 0.018, η2_p = 0.203) and ankle plantarflexion (p = 0.010, η2_p = 0.235). Post hoc comparisons showed that compared with week 0, the proprioception thresholds of the knee extension and ankle plantarflexion of both groups (TCHHC + TE: p < 0.001, d = 1.870, p < 0.001, d = 2.434; TE: p = 0.007, d = 0.927; p < 0.001, d = 1.352) decrease at week 9. The proprioception threshold of the knee extension was lower in the TCHHC + TE group compared with those in the TE group at week 9 (p = 0.022, d = 0.967). Significant time effects were detected in proprioception thresholds of knee flexion (p < 0.001 η2_p = 0.540), ankle dorsiflexion (p < 0.001 η2_p = 0.446), ankle inversion (p = 0.048, η2_p = 0.148) and eversion (p = 0.008, η2_p = 0.246).

The times of TUG and 20-m walk tests were presented in Table 5. Significant interactions were detected in the TUG (p = 0.046, η2_p = 0.150) and 20-m walk (p = 0.008, η2_p = 0.249) tests. Post hoc comparisons showed that compared with week 0, the times of the TUG and 20-m walk tests of both groups (TCHHC + TE: p < 0.001, d = 4.420, p < 0.001, d = 3.400; TE: p < 0.001, d = 3.113; p < 0.001, d = 2.543) decrease at week 9. The times of the TUG and 20-m walk tests was lower in the TCHHC + TE group compared with those in the TE group at week 9 (p = 0.022, d = 0.957; p = 0.044, d = 0.820).

This study investigated the effects of an 8-week TCHHC + TE on pain, proprioception, and functional performance among older adults with KOA. The results supported our hypothesis. Compared with TE, TCHHC + TE was superior in relieving pain, recovering proprioception, and improving functional performance among older adults with KOA.

At week 9, pain scores decreased in both groups, and the decrease was more significant in the TCHHC + TE group than in the TE group. The effects of TE on pain relief have been well documented (Juhl et al., 2014; Zeng et al., 2021). Resistance training increases the muscle strength of the lower extremity, improved the stability of knee joints, and reduces the wear between articular cartilage to relieve pain (Chen H. et al., 2019). Joint mobilization reduces the excitability of nerves to relieve pain (Saunders et al., 2005). Our results showed that TCHHC + TE significantly relieved pain among older adults with KOA compared to TE. Similar results were reported, supporting that the addition of TCHHC to TE enhanced the relief of knee pain (Chen S. et al., 2019). There are several possible mechanisms for TCHHC + TE’s superior effects for pain relief, as follows. 1. Cao (Cao et al., 2016) et al. reported that the herbs have been demonstrated to be effective at promoting absorption of inflammatory substances and eliminating swelling by increasing blood circulation in clinical observation. In this study, some of the components of the herbs are effective in relieving pain, e.g., peony root reduces the expression level of substance P, which is the most important element in pain perception (Wang et al., 2013). Prepared sichuan black and sichuan ox knee reduces the level of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6, and IL-17A) in serum, increases the viscosity of intra-articular knee joint fluid, and eliminates swelling and promotes blood circulation (Weng et al., 2014; Li et al., 2017; Zhang et al., 2019a). Miyajima (Miyajima et al., 2013) et al. point out that hot compress increases the temperature of local tissues and reduces the excitability of afferent nerves, and helps control swelling and reduce pain. TCHHC allow herbs to penetrate the skin barrier through a warming effect, immediately reaching and acting on the affected areas. TCHHC can regulate the imbalance of the levels of anti-inflammatory factors (TGF-β1, IL-13), pro-inflammatory cytokines (IL-1β, IL-6), and pain mediators (PGE2, 5-HT), thus relieving pain (Orita et al., 2011). 2. TE facilitates TCHHC’s pain-relieving effects, by enhancing TE enhances blood and lymphatic fluid circulation, as well as promoting local vasodilation (Guo et al., 2021); thus, it increases the absorption of herbs into the skin and promotes anti-inflammatory and pain-relieving effects.

At week 9, the proprioception thresholds of the knee extension and ankle plantar flexion decreased significantly in both groups, and the decrease was more significant in the TCHHC + TE group than in the TE group. Previous studies showed that TE has a positive effect on the proprioception of knee and ankle joints (Shen et al., 2021), which is consistent with our study. As the most important proprioceptor within the knee joint, the muscle spindles are stimulated by muscle lengthening, speed, or acceleration (Orita et al., 2011).TE releases and stretches the tensed muscles around the knee joint, coordinates the contraction ability between muscles, and enhances muscle spindle sensitivity to improve proprioception among older adults with KOA (Fransen et al., 2002). There are several possible reasons why TCHHC + TE had a superior effect on proprioception, as follows. 1. In this study, some of the components of the herbs have been proven to recover proprioception, such as cassia twig, mulberry parasitic, and antler gum prevent the loss of proteoglycans and accelerate the proliferation of chondrocytes to protect joint cartilage cells and improve their metabolism (Xu and Liu, 2004; Fan et al., 2012), reduce the damage to mechanoreceptors, promote neuromuscular control, and improve proprioception (Chen et al., 2016). Nagashima et al. (Nagashima et al., 2006) and Rusminingsih et al. (Rusminingsih et al., 2020) pointed out that hot compresses applied to the knee joint to dilate blood vessels, increases the excitability of motor neurons and recruits many motor neurons to participate in activities (Levine, 2007), facilitates sensory input. 2. TCHHC enhanced TE’s effects on proprioception. By reducing the excitability of nerve endings and eliminating swelling, thereby allowing the adhesions and atrophy of the knee joint and its surrounding soft tissues to recover as much as possible (Zhang et al., 2015), which facilitates the sensitivity of muscle spindles (Wang et al., 2010; SH, 2012).

At week 9, the times of TUG and 20-m walk decreased significantly in both groups, and the decrease was more significant in the TCHHC + TE group than in the TE group. TCHHC + TE is more effective in improving functional performance, which is consistent with a previous study (Song, 2022). Another study has shown that herbs are used to enhance muscle strength and function performance (Sellami et al., 2018). The level of magnesium (Mg) (Zeng et al., 2015) and calcium (Ca⁺) (Li et al., 2016) in the serum are significantly decreased among older adults with KOA, which reduces their functional performance, increases the risk of falls, and accelerates the deterioration of KOA (Veronese et al., 2017; Heffernan et al., 2020). Some of the Traditional Chinese Herbs, e.g., paeony root, ephedra root, atractylodes macrocephala reduce pro-inflammatory cytokines such as TNF- α and interleukins, inhibit the production of matrix metalloproteinases (Yuan et al., 2015; Malemud, 2017), thereby improve the level of Mg and Ca⁺ in serum (Gunn et al., 2013; Zhang et al., 2019a). In addition, parsnip and anemarrhena asphodeloides reduce the expression of fibroblast-like synoviocytes and receptor activator for NF-kB ligand, inhibit the activation of osteoclasts, and reduce the destruction of cartilage, bone, and tendon (Yuan et al., 2015). Previous studies have pointed out that hot compress promotes blood circulation, activates the motor cortex, and facilitates the recovery of adhesions and atrophy of the muscles and soft tissues around the knee joint (Zhang et al., 2015). Therefore, TCHHC is more effective in improving functional performance by expands capillaries, promotes blood circulation, and releases muscle tension (Savaş et al., 2019). The combination of TE facilitates herb penetration in the affected area, reduces stress-related muscle tension, improves lower extremity coordination, and enhances muscle strength to improve functional performance (Guo et al., 2021).

This study has several limitations. First of all, the effects of the TCHHC are possibly attributed to Chinese herbs or hot compress, further studies are recommended to further investigate their specific effects. Second, there was no follow-up after the 8-week intervention, and it was impossible to determine how long the effects would last.

Compared with TE, the 8-week TCHHC + TE was superior in relieving pain, recovering proprioception, and improving functional performance among older adults with KOA. It is recommended that TCHHC should be adopted prior to TE to enhance the effects of KOA rehabilitation.