Injection route affects intra-articular hyaluronic acid distribution and clinical outcome in viscosupplementation treatment for knee osteoarthritis: a combined cadaver study and randomized clinical trial


The coverage of hyaluronic acid (HA) on the impaired cartilage should be the precondition to exert its beneficial effect on knee osteoarthritis (KOA) according to the pharmacological mechanism. However, the intra-articular distribution of HA might be correlated with the route of drug delivery. Forty-two cadaver knees with radiographic evidence of osteoarthritis were given anteromedial (AM) or medial midpatellar (MMP) injection of HA (molecular weight 600–1500 kD) followed by gait stimulation. Although 2.5 ml HA delivered through both routes failed to cover the entire cartilage, HA covered 96.12% cartilage of patellofemoral joint (PFJ) and 71.44% of medial femorotibial joint (FTJ) through MMP route, whereas mainly distributed into FTJ and posterior condyles through AM route. HA in the MMP group distributed more in PFJ than that in the AM group (P < 0.001), but no significant difference presented in medial FTJ (P = 0.084). The clinical efficacy was also associated with the route of drug delivery. One hundred patients with unilateral mild-to-moderate KOA were recruited and randomly assigned to receive five weekly HA injections with AM route (n = 50) or MMP route (n = 50). Patients in the MMP group obtained better improvement in WOMAC index total score, pain score, stiffness score, and Lequesne index total score over the entire follow-up period, as compared to patients in the AM group (all P < 0.01). More patients in the MMP group claimed pain relief (71.7%, P = 0.024) and felt satisfying (63.1%, P = 0.007) than in the AM group at the end of follow-up. Therefore, intra-articular HA injection through MMP route is recommended in treating mild-to-moderate KOA.



Knee osteoarthritis (KOA) is an age-related disorder of synovial joint characterized by progressive articular cartilage erosion, subchondral bone sclerosis, and reduced concentration of hyaluronic acid (HA) [1]. HA serves in multiple functions including the maintenance of the rheological properties of synovial fluid, shock absorption, joint lubrication, and energy dissipation. Notably, the reduced molecular weight and lower concentration of HA are believed to potentially contribute to the pain, stiffness, and loss of mobility [2]. Since HA is a natural component of synovial fluids, viscosupplementation with intra-articular injection of exogenous HA is recommended as a viable option for KOA, which has been shown to provide lubrication and elastic shock absorption to improve symptoms [2].

Although the efficacy and safety of viscosupplementation have been extensively reported [3, 4], studies concerned to improve its efficacy never cease. However, most of the current studies focus on the HA products [5,6,7], such as the proposed high molecular weight HA and chemically cross-linked HA. The influence of injection technique is easily neglected. The efficacy of viscosupplementation to treat KOA could be improved with the optimized injection technique [8].

Based on the mechanism of action, exogenous HA helps maintain the smoothness of the surface of articular cartilage and suppress degeneration through decreasing friction and permeating synovial and cartilage tissues to reach targeted cells (such as synoviocytes and chondrocytes), leading to an effect of anti-inflammatory, analgesic, and chondroprotection [9]. Hence, the coverage of HA on the impaired cartilage should be the precondition to exert its mechanical and biological effects. Previous study had proven that intra-articular HA distribution into non-target regions compromised the effective symptomatic benefits in patients [10]. It is commonly believed that HA easily expands and evenly distributes into the entire cartilage surface, regardless of the injection route. However, this assumption has never been verified. There might be such a possibility that injection route affects HA coverage to the damaged cartilage; thus, the therapeutic efficacy of intra-articular HA injection will possibly be compromised.

Medial midpatellar (MMP) and anteromedial (AM) routes are two common injection sites for viscosupplementation. We hypothesize that diverse injection routes lead to different intra-articular HA distribution, and a better clinical efficacy would be achieved by HA covering the damaged regions of articular cartilage; thus, the injection route may directly influence the clinical outcome. Therefore, in this study, we firstly examined the intra-articular distribution of HA delivered through different injection routes in cadaver knees. A clinical trial was then carried out to determine whether different injection routes (MMP and AM) would affect clinical outcome of viscosupplementation in the treatment of mild-to-moderate KOA. The study was carried out after the approval of Medical Ethics Committee of Nanfang Hospital (NFEC-2014-074).

Materials and methods

The cadaver study


Cadaver knees were collected from the anatomy department of our university after permission, and the current work was carried out in accordance with national and international ethical criteria. The knees were screened by X-ray examination. Finally, 42 knees with radiographic evidence of OA [11] were included and numbered 1 to 42.

Injection technique

The MMP route was administered with the cadaver knee extension. The needle was advanced transversely between the articular surfaces of the patellofemoral joint at the point where the mid-horizontal line of the patella met the medial border of the patella (Fig. 1a). The AM route was performed with the target knee at 90° flexion. The needle was advanced towards the intercondylar notch at the site of one fingerbreadth proximal to the tibial joint surface and medial to the patella tendon (Fig. 1b). Forty-two cadaver knees were randomly assigned into two groups based on the injection routes: MMP group (n = 21) and AM group (n = 21).

Fig. 1

The injection routes for a injection with medial midpatellar (MMP) portal and b injection with anteromedial (AM) portal

An avian-derived non-crosslink and no-gel solution of sodium hyaluronate with an intermediate molecular weight (600–1500 kD) (Baibei; Jingfeng Pharmaceuticals, China) were used in the current study. Methylene blue has been proven a reliable measure to assess intra-articular distribution [12, 13]. The exogenous 25 mg/2.5 ml HA was mixed with 75 μl of 5% methylene blue. The compatibility of the mixture was confirmed (Fig. 2). The mixture was then injected into each cadaver knee using a 1.25-in. (3.5-cm), 23-gauge needle. The accuracy of needle placement was determined by exposing the knee joint and viewing the distribution of the mixture; that is, with incorrect needle placement in the soft tissue, the mixture cannot be observed inside the joint.

Fig. 2

a The original mixture. b Standing for 6 h. c Standing for 12 h

Gait stimulation

After HA injection, gait stimulation of the knee specimens was performed through the electro-force biomechanical device (Bose Corporation ElectroForce Systems Group, USA) [14, 15]. Knee OA mainly influenced the swing phase of the frontal kinematic waveforms, and the swing phase primarily affected the intra-articular spreading of HA with the flexion and extension of knee [16]. Therefore, simple swing phase was stimulated in our cadaver study. Five to 45 degrees of motion with a frequency of 45 cycles per min were adopted in the gait stimulation, referred from the gait of normal and symptomatic knee OA [17, 18]. A force of 110 N proximally relative to the axial load of the knee was applied using the above electro-force biomechanical device. The input force was based on the measurement of contract force for level walking [19, 20]. For each knee specimen, the stimulation process lasted for 2 h, which fulfilled the basic activity length of patients.

Compartments of the knee joint

Referring to the study reported by Peterfy et al. [21], the compartments of the knee joint were divided into 14 regions (Fig. 3). The patella was divided into two regions: the lateral patella (LP) and the medial patella (MP). Both the medial femur (MF) and lateral femur (LF) could be further divided into three regions: anterior (a), central (c), and posterior (p). Similarly, the medial and lateral tibial plateaus (MT and LT) were also divided into three regions: anterior (a), central (c), and posterior (p). The anterior region of the femoral cartilage corresponded to the patellofemoral joint, the central region between the anterior and posterior meniscal ligaments corresponded to the area of the femoral cartilage loaded during standing and normal walking, and the posterior region corresponded to the area loaded during deep flexion of the knee. Therefore, the patellofemoral joint (PFJ) regions included MP, LP, MFa, and LFa. The medial femorotibial joint (MFTJ) regions included MFc, MTa, MTc, and MTp, and the lateral femorotibial joint (LFTJ) regions included LFc, LTa, LTc, and LTp. The medial posterior condyle and lateral posterior condyle were categorized as two independent regions.

Fig. 3

Regional subdivision of the articular surfaces. The patella was divided into two regions: the lateral patella (LP) and the medial patella (MP). Both the medial femur (MF) and lateral femur (LF) could be further divided into three regions: anterior (a), central (c), and posterior (p). Similarly, the medial and lateral tibial plateaus (MT and LT) were also divided into three regions: anterior (a), central (c), and posterior (p)

Analysis of HA distribution

Peterfy et al. [21] have successfully employed semiquantitative scoring methods to grade the cartilage damage based on the extent of regional involvement. Accordingly, the distribution in the abovementioned 14 regions was calculated with quantification in our study. The coverage area was then assessed by image analysis. The AutoCAD software (Autodesk, USA) was used to estimate the percentage of colored area in the individual regions, followed by scoring. The grades were defined as follows: 0 = “none,” 1 = “< 25% of the region,” 2 = “25% to 50% of the region,” 3 = “50% to 75% of the region,” and 4 = “> 75% of the region.” The score of PFJ was the sum scores from its each component, i.e., MP, LP, MFa, and LFa. Similar method was applied for MFTJ and LFTJ. The maximum scores for PFJ, MFTJ, LFTJ, medial posterior condyle, lateral posterior condyle, and the entire knee were therefore 16, 16, 16, 4, 4, and 56, respectively. The fractional coverage area in PFJ is equal to the estimated score divided by the maximum score of PFJ.

The clinical study


A prospective, single-blind, randomized, and controlled study was performed. The procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975. The current clinical trial has registered in ClinicalTrial

Gov (NCT03600571). A written informed consent was obtained for each participant before enrollment.

From September 2014 to January 2017, mild-to-moderate KOA patients, who fulfilled the American Rheumatism Association criteria [22] and met the following standards [23]: (1) radiographic manifestation of Kellgren-Lawrence (K-L) grade 2 or 3 and (2) a score of 20% on at least 1 pain dimension of Western Ontario and McMaster Universities Osteoarthritis (WOMAC) VA3.0 index, were recruited from our clinic.

Following were the exclusive criteria: pregnancy, acute fracture, rheumatoid arthritis, gout, traumatic arthritis, inflammatory arthritis, severe knee joint effusion, oral NSAIDs in the past 2 weeks, HA allergy history, intra-articular injection of HA or corticosteroid or surgery to the target knee in the past 6 months, K-L grade 4 KOA, and infectious/inflammatory skin diseases in the area of the affected knee. Patients with active liver, kidney, cardiovascular, and cerebrovascular diseases were also excluded.

The sample size was calculated assuming a SD of 23 mm based on a previous study of Berenbaum et al. [24], resulting in 50 patients per group to achieve a power of 80% at a significance level of 0.05 with a 10% predicted dropout rate [25]. Randomization was done on an individual basis with consecutively numbered sealed envelopes (from 1 to 100) that were opened after patient recruitment. Patients with odd number were assigned to the AM group while with those with even number were assigned to the MMP group. Finally, a total of 100 patients were enrolled in the study: AM group (n = 50) and MMP group (n = 50). Each patient received five weekly intra-articular injections of HA (2.5 ml/25 mg, 600–1500 kD; Baibei, Jingfeng Pharmaceuticals). At the baseline visit, the following data were collected: patient demographics, symptom duration, K-L grade, WOMAC index, and Lequesne index.

Injection technique

Intra-articular injection of HA was performed under strict aseptic technique in the clinic by the same sophisticated senior surgeon. The injection approach was similar to that of the cadaver study (Fig. 1).

According to the study of Jackson et al. [26], the accuracy of injection was judged based on the following indications: (1) incorrect placement of a soft tissue injection causes more discomfort to the patient during and after the procedure, and (2) once the instantaneous discomfort of needle placement has subsided, the injection of HA should not be painful. After injection, patients were advised to take a relative rest period for 48 h (walk slowly, avoid frequently going upstairs and downstairs, etc.) [27]. Physical exercises, such as running, jogging, and tennis, were not recommended at this period.

Outcome measurements

Clinical outcomes were evaluated by an investigator who was blind to the injection route. WOMAC and Lequesne indexes, the most widely used assessment of OA-specific health status, were applied as the assessment tool. The WOMAC index [28] includes 5 questions about pain, 2 about stiffness, and 17 about difficulty in daily activities. Each subscale has a 5-point scale grading (0 = none, 1 = mild, 2 = moderate, 3 = severe, 4 = extreme). The Lequesne index [29] questionnaire includes 10 questions about pain, stiffness, and function. The total score ranges from 0 (no pain or stiffness, no disability) to 24 (maximum pain, stiffness, and disability). The WOMAC index and Lequesne index questionnaires were completed at the initial screening procedure as baselines and at 1 week, 2 weeks, 3 weeks, 4 weeks, 5 weeks, 14 weeks, and 24 weeks post the first injection. Patients’ global assessment was conducted in the end of follow-up, including pain relief, joint function, and satisfaction.

Adverse event was defined as any emergent signs, symptoms, or undesirable events that occur while a patient is using HA (2.5 ml/25 mg, 600–1500 kD; Baibei, Jingfeng Pharmaceuticals) during the study. The adverse event information obtained from the patient included the problem description, onset and stop dates, history, severity, outcome of event, and action taken. Referring to the report of Berenbaum et al. [24], the adverse events include allergic reaction, pain in the injection site, bleeding in the injection site, arthralgia, and others (swelling, redness, or effusion) at each visit following injection.

Statistical analysis

Statistical analysis was performed using the IBM SPSS Statistics 20.0 (version 20.0; Armonk, New York, USA). The distribution of values in each group has been checked. In the cadaver study, Student’s t test was used for the comparison of the intra-articular distribution of HA between the MMP group and the AM group. When a heteroscedasticity occurred, the Mann-Whitney test was used for comparison. In the clinical study, the Friedman test was used for the comparison of repeated non-parametric measurements, the Wilcoxon signed-rank test for the paired comparison of pre- and post-injection scores, and Pillai’s trace for the comparison of WOMAC index values and Lequesne index values between the MMP and AM groups. Differences at a P level of < 0.05 were identified as statistically significant.


Distribution of intra-articular injection of HA

HA delivered through both injection routes failed to cover the entire articular cartilage. The intra-articular distribution of HA is shown in Table 1. With the MMP injection route, intra-articular HA was able to cover 96.12% articular cartilage of PFJ and 71.44% of MFTJ as well as 52.69% of the LFTJ. However, HA covers a small part of the posterior condyle region, with only 17.75% coverage to the medial and 26.25% to the lateral.

Table 1 Distribution of hyaluronate in cadaver knees under MMP or AM injection

In contrast, HA injection with AM route showed a predominant distribution inside the FTJ and posterior condyles, with an 84.48% coverage to the MFTJ and 83.62% to the LFTJ and a 73.75% coverage to the medial posterior condyle and 72.70% to the lateral posterior condyle. However, the HA distribution inside the PFJ was very limited, which showed only 28.87% coverage to the PFJ cartilage.

As compared to the AM group, intra-articular HA injection in the MMP group leads to more extensive distribution into PFJ (P < 0.001), while less into LFTJ (P = 0.001; 95% CI, 2.18–7.72) and bilateral posterior condyles (both P < 0.001). No significant difference was observed in the MFTJ (P = 0.084) and the entire knee joint (P = 0.571) between the two groups. Figure 4 shows the feature of HA distribution with MMP and AM injection routes.

Fig. 4

HA distribution was visualized with methylene blue dye in cadaver knees with MMP route (a, c) that shows HA concentrates in patellofemoral joint and medial femorotibial joint, and AM route (b, d) that shows HA mainly distributes into femorotibial joint and posterior condyle

Clinical outcomes of intra-articular HA injection

The recruitment was completed in January 2017, and the follow-up was finished in July 2017. Forty-six patients in the MMP group (92%) and forty-four patients in the AM group (88%) completed all five intra-articular HA injections and 24-week follow-up assessment (Fig. 5).

Fig. 5

The flow of participants for the MMP group (HA injection with medial midpatellar portal) and AM group (HA injection with anteromedial portal)

No significant difference was present for the demographic features of the participants between the two groups. Table 2 shows the results. Patients’ baseline clinical measurements had no significant difference in the two groups including the Lequesne index (P = 0.077) and WOMAC index (pain score [P = 0.052, power = 0.6760], stiffness [P = 0.180], function [P = 0.814], total score [P = 0.059, power = 0.9910]) (Table 2).

Table 2 Demographic characteristics and clinical baselines of the patients who completed the 24-week follow-up

Lequesne index scores in both groups were significantly reduced at all assessment time points after the initial injection (all P < 0.05) (Table 3, Fig. 6). Significant reduction was also noted in the pain score, stiffness score, function score, and total score (all P < 0.05) of the WOMAC index in both groups at all assessment time points as compared to the individual baseline (Table 3, Fig. 7).

Table 3 Clinical outcomes of HA treatment in patients measured by the Lequesne index and WOMAC index (mean ± SD)
Fig. 6

In both groups, the total score of the Lequesne index at every assessment time point was significantly improved as compared to the baseline (P < 0.05). Patients in the MMP group exerted a better efficacy over the 24-week follow-up than in the AM group (P = 0.01)

Fig. 7

ad In both groups, the WOMAC pain score, stiffness score, function score, and total score were significantly improved at every assessment time point as compared to the baselines (P < 0.05). Patients in the MMP group exerted a better efficacy in total score (P = 0.004), pain score (P < 0.001), and stiffness score (P = 0.001) over the 24-week follow-up than in the AM group

Moreover, the improvement of the Lequesne index total score (P = 0.010; 95% CI, − 1.762 to 2.3941) (Fig. 6), WOMAC index subscale pain score (P < 0.001; 95% CI, 0.355 to 2.414) (Fig. 7a), subscale stiffness score (P = 0.001; 95% CI, − 0.469 to 1.205) (Fig. 7b), and total score (P = 0.004; 95% CI, − 3.424 to 20.483) (Fig. 7d) over the 24-week follow-up was significantly better in the MMP group compared to the AM group (Table 3). However, there was no statistically significant difference in the subscale function score of the WOMAC index between the two groups (P = 0.119) (Table 3, Fig. 7c). At assessment time points of 1 week, 2 weeks, 3 weeks, 4 weeks, 5 weeks, and 14 weeks, the Lequesne index total score showed more improvement in the MMP group than in the AM group (P < 0.05). The WOMAC total score and subscales of pain score at each assessment time point presented more improvement in the MMP groups than the AM group (P < 0.05). The rest subscales of WOMAC showed no significant difference between the two groups at each assessment time point.

The global assessment at the end of follow-up was also collected (Table 4). More patients in the MMP group claimed pain relief (71.7%, P = 0.024) and felt satisfying (63.1%, P = 0.007) than in the AM group (61.4% and 40.9%, respectively). Seventy-four percent of patients claimed improvement in joint function in the MMP group while 66% in the AM group. But, it showed no significant difference between the two groups (P = 0.383). During the study, two patients in the AM group suffered from significant pain in the injection site after injection.

Table 4 Global assessment at the end of follow-up


Alternative treatments, i.e., non-surgical treatment or minimally invasive surgery for KOA prior to total knee arthroplasty, have been extensively studied, but no definitive consensus has been reached so far. The high prevalence of cardiovascular and metabolic comorbidities in patients with KOA often restricts the administration of pharmacological medications. The benefits of intra-articular HA injection in relieving pain associated with knee OA is still controversial. Some studies suggested that it may be a placebo effect. In a randomized, controlled, and double-blinded trial, van der Weegen et al. [30] demonstrated that HA was not superior to saline in the management of knee OA with an improved knee pain and functional outcome. However, many more research (13 out of 17 meta-analyses) supported the beneficial effect of HA injection [31]. The reported discrepancy may result from the variations in HA preparation, including its concentration, molecular weight, and molecular organization (linear or cross-linked) and also the protocol of injection [27, 31, 32]. The current study aimed to evaluate the HA injection route on its intra-articular distribution and the corresponding clinical outcome in treating knee OA; thus, a single preparation of HA, the non-crosslink, and no-gel solution with an intermediate molecular weight (600–1500 kD) (sodium hyaluronate, Baibei; Jingfeng Pharmaceuticals, China) was used to exclude the influence of product preparation. However, bear in mind that HA preparation with different concentrations, molecular weights, and molecular organizations theoretically may lead to different rheological and viscous properties that will influence the spreading capacity and the intra-articular distribution of HA and eventually result in different clinical outcomes. A more complex study design to investigate the choice of HA preparation in combination with the route of injection is warranted in the future.

Our protocol with five weekly HA injections was found safe and effective. Merely two patients in the AM group suffered from significant injection site pain after injection, and the other participants did not report local adverse effect. More than 60% patients in both groups experienced pain relief at the end of follow-up. Importantly, it even reached 71.8% in the MMP group. Seventy-four percent patients claimed improvement in joint function in the MMP group while 66% in the AM group. The average efficacy including pain relief and functional improvement occurred shortly at 1 week after injection. Such positive effects persisted up to 24 weeks of follow-up.

Previous studies have suggested that injection route of HA may affect its treatment efficacy. Conrozier et al. [8] suggested that the injection route might influence clinical outcomes, such as patient satisfaction, safety, pain, and function. Toda and Tsukimura [33] reported a greater improvement in patients’ pain relief resulted from HA injection by the modified Waddell portal (an AM portal with manipulative ankle traction at 30° of knee flexion) as compared to the lateral patella portal. A difference also found in our study, which showed that the intra-articular HA injection with the MMP route, led to a greater improvement in the total score, the subscale pain score, the subscale stiffness score of the WOMAC index, and the total score of the Lequesne index than the AM route. By the end of the follow-up, 16 out of 46 patients in the MMP group felt very satisfied with the treatment whereas only 3 patients were very satisfied in the AM group. The global satisfactory rate was 63.1% in the MMP group and 40.9% in the AM group with a statistically significant difference.

Based on the result of the cadaver study, we believed that the different intra-articular distributions of HA between the two injection routes probably contributed to the diverse clinical outcomes. Although HA of the package dose 2.5 ml injected through both MMP and AM injection routes failed to cover the entire articular cartilage, different distributions of HA were present between the two injection routes. Similar to what Jackson and Simon [34] have reported, we found that HA was mainly localized to the articular surfaces in FTJ by the AM injection route. Luo et al. [35] also reported HA distribution to FTJ after injection through the joint line and to PFJ by medial patellar injection in a rabbit model.

More importantly, HA injection with MMP route provides a better coverage to the susceptible cartilage in the osteoarthritic knee, which was considered closely associated with better clinical efficacy. HA exerts topical beneficial effects in osteoarthritic knee with the permeability to reach the targeted cells [9, 36], i.e., modifying structural damage, eliminating oxygen-derived free radicals, anti-inflammation, and suppressing cartilage matrix degradation. Hence, the coverage of HA on the impaired cartilage should be the precondition to exert its effect. Although the region of MFTJ was usually of greatest clinical concern in knee osteoarthritis, Peterfy et al. [21] proposed an inconsistent conclusion with a more sensitive and specific instrument (magnetic resonance imaging) to detect the involved regions. They indicated that cartilage abnormalities were most frequent in the PFJ (94%) in KOA patients rather than MFTJ (89%) and LFTJ (71%), as well as the chance to bone cysts and bone attrition. On the other hand, the posterior region of femur condyle corresponded to the area loaded during deep flexion of the knee; videlicet, the region might less be involved in KOA [21].

In the present study, we found that MMP injection route was superior in delivering HA to the above damaged cartilage regions of osteoarthritic knee, which had a nearly full coverage to PFJ and large coverage to MFTJ as well as half coverage to LFTJ. In contrast, distribution of HA in PFJ was seriously inadequate with AM injection route. Distribution of HA in PFJ was found more extensive in the MMP group, while in MFTJ, it showed no significant difference between the two groups. The mechanical properties of exogenous HA are noted as lubricating the cartilage surface and decreasing friction, which was accomplished in the condition of coating the cartilage. It was noted that Bonnevie et al. [37] recently have proposed two pathways by which chondrocytes respond negatively to friction: an immediate necrotic response and a longer-term pathway involving mitochondrial dysfunction and apoptosis. It was concluded that the features of HA in terms of friction coefficient and lubricating ability to cartilage were predictive to clinical outcomes in the treatment of knee OA [38]. Singh et al. [39] also suggested that exogenous HA coating the cartilage surface can drastically alter the lubricating response when tuning its affinity to the cartilage. Apart from the mechanical effects, the biological benefits of HA also contribute to a better efficacy. Tobetto et al. [40] proposed that HA provided an ideal environment to inhibit the neutrophil-mediated cartilage degradation, as it was a major constituent of cartilage, possessed excellent biocompatibility, and was well known for its anti-inflammatory properties. Therefore, viscosupplementation with MMP injection route could improve the efficacy to treat KOA.

Interestingly, our patients claimed the maximal efficacy at 14 weeks after injection, which is consistent with the report by Arrich et al. [41]. Similar to other non-modified HA products [42], the intra-articular half-life of the sodium hyaluronate used in the present study is 12–24 h. The biological mechanism of exogenous HA has been proposed with the simple fact that its clinical effects often last for months despite its rapid clearance from the joint [43]. The biological effect may be accomplished in two steps, reaching the surface of cartilage and binding to the specific receptors through permeating [9]. Therefore, the activities of HA are associated with both the intra-articular distribution and its clearance rate. The biological effects of HA on cartilage would be affected if either the cartilage is not covered or HA is cleared rapidly and unable to bind to the receptor. As presented in our study, the intra-articular distribution of HA is significantly associated with the injection route. Komatsu et al. [44] suggested that higher molecular weight HA was associated with longer residual time in the knee joint. Altman et al. [43] summarized in a review that the basic preclinical science–discovered higher molecular weight HAs were associated with longer clearance rate that may provide superior chondroprotection and promote proteoglycan/glycosaminoglycan, anti-inflammatory, mechanical, and analgesic effects. Whether these in vitro observations can be translated into clinical evidence has not been clearly established yet.

Although HA products in general have a short residual time in the knee joint, our results showed that the efficacy of HA injection lasted for 24 weeks, which was the end point of our follow-up study. HA may work through a progressive and cumulative process. Akmal et al. [45] suggested that HA had a significant stimulatory effect on the metabolic activity of chondrocytes. Chareancholvanich et al. [46] also showed that HA treatment had a beneficial effect on knee cartilage preservation and cartilage volume increase. The benefit of resurfacing the damaged articular cartilage was also reported, which attributed to enhanced mechanical properties, reduced susceptibility to enzymatic degradation, and reduced adhesion of macrophages [47]. With the progress of continuous repairing, the articular cartilage has the potential to recover its thickness and biomechanical properties to some extent, even not to its primary status.

HA has other reported functions including inhibition of immune complex adherence to polymorphonuclear cells [48], inhibition of topical inflammatory reaction [34], and scavenging of reactive oxygen-derived free radicals [49], which may all contribute to pain relief. Ghosh and Guidolin [50] suggested that HA formed a pericellular coat around cells, interacted with proinflammatory mediators, and bound to cell receptors such as cluster determinant CD44 and receptor for hyaluronate-mediated motility RHAMM, through which it modulated cell proliferation, migration, and gene expression. Some investigators also suggested that HA altered the pain sensation by coating pain fiber receptor endings and thus reduced evoked sensory nerve transmission [51, 52]. These may account for the long-term beneficial effect. Although the average WOMAC score on pain, stiffness, and function and the Lequesne score seemed to rise again after 14 weeks, the treatment effect of HA was maintained up to the end of follow-up (24 weeks).

The impact of HA product difference on treating knee OA has been previously discussed [5, 43]. Intra-articular HA exerted both biomechanical and potential disease-modifying effects in treating knee OA, which might be influenced by individual HA preparations with regard to rheological, lubricating, viscous, and permeability properties, leading to different clinical outcomes. In the current study, we demonstrated that the injection route influenced the intra-articular distribution of HA and the coverage of cartilage. HA coating of the impaired cartilage should be the precondition to provide lubricating benefits and biological effects through binding to its specific receptors (HA-CD44). The product difference of HA might also affect its intra-articular distribution and the corresponding clinical outcomes. Therefore, both HA preparation and the route of injection are of considerable importance in clinical application of intra-articular HA injection to treat knee OA.

There are more than eighty HA preparations on the worldwide market with different degrees of cross-linking versus no cross-linking, the use of gels versus no-gel solutions, and various molecular weights. Our current study chose an intermediate molecular weight HA that was commonly supplied in China market. Our study revealed that the MMP portal injection route led to a significant superior coverage on PFJ than the AM portal route (averagely 96.12% versus 28.87%). This manifest difference indicated a vital importance of the route of injection in the potential clinical benefits of HA.

Another limitation of the study was that two most common injection routes (MMP route and AM route) were included, but other injection routes, such as lateral mid-patella route and superomedial route, need further investigation. It is proposed that chemical exchange saturation transfer (Visco-CEST) has a potential to assess the spatial and temporal variations of HA in vivo [53]; however, it is still limited to the laboratory experiments. Methylene blue has been proven a suitable indicator to demonstrate the accuracy of different injection sites [12]. Therefore, the mixture of methylene blue and HA was employed to discuss the intra-articular distribution of HA in cadaver knee. Besides, in the current study, we chose the commonly used dose of HA to investigate the intra-articular distribution and clinical outcome. However, whether other larger doses, i.e., 3.0 ml, sometimes used in clinic will increase the pharmaceutical coverage to the cartilage or alter the distribution, and its impact on clinical efficacy, needs further discussion. Our cadaver study was unable to accurately evaluate the joint space loss of the involved knee; thus, the relationship between joint space narrowing and intra-articular distribution of HA is unknown.

In conclusion, intra-articular HA injection is an effective and safe treatment for patients with mild-to-moderate KOA to relieve pain and improve joint function. Injection through MMP route is observed to provide a better coverage of HA to the vulnerable regions of the articular cartilage in osteoarthritic knee. That is, intra-articular HA with MMP injection route showed a nearly full coverage to PFJ and large coverage to MFTJ, as well as half coverage of LFTJ. Correspondingly, injection through the MMP route is superior to AM route in achieving pain relief and global patient satisfaction. The inadequate coverage in PFJ might be responsible for the compromised effect in the AM group. Therefore, we recommend intra-articular HA injection through the MMP route for patients with mild-to-moderate KOA in the future clinical practice.


  1. 1.

    Dieppe PA, Lohmander LS. Pathogenesis and management of pain in osteoarthritis. Lancet. 2005;365(9463):965–73.

    CAS  Article  PubMed  Google Scholar 

  2. 2.

    Moreland LW. Intra-articular hyaluronan (hyaluronic acid) and hylans for the treatment of osteoarthritis: mechanisms of action. Arthritis Res Ther. 2003;5(2):54–67.

    CAS  Article  Google Scholar 

  3. 3.

    Altman RD, Rosen JE, Bloch DA, Hatoum HT, Korner P. A double-blind, randomized, saline-controlled study of the efficacy and safety of EUFLEXXA for treatment of painful osteoarthritis of the knee, with an open-label safety extension (the FLEXX trial). Semin Arthritis Rheum. 2009;39(1):1–9.

    CAS  Article  PubMed  Google Scholar 

  4. 4.

    Navarro-Sarabia F, Coronel P, Collantes E, Navarro FJ, de la Serna AR, Naranjo A, et al. A 40-month multicentre, randomised placebo-controlled study to assess the efficacy and carry-over effect of repeated intra-articular injections of hyaluronic acid in knee osteoarthritis: the AMELIA project. Ann Rheum Dis. 2011;70(11):1957–62.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  5. 5.

    Altman RD, Bedi A, Karlsson J, Sancheti P, Schemitsch E. Product differences in intra-articular hyaluronic acids for osteoarthritis of the knee. Am J Sports Med. 2016;44(8):2158–65.

    Article  PubMed  Google Scholar 

  6. 6.

    Guo Y, Yang P, Liu L. Origin and efficacy of hyaluronan injections in knee osteoarthritis: randomized, double-blind trial. Med Sci Monit. 2018;24:4728–37.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7.

    Trojian TH, Concoff AL, Joy SM, Hatzenbuehler JR, Saulsberry WJ, Coleman CI. AMSSM scientific statement concerning viscosupplementation injections for knee osteoarthritis: importance for individual patient outcomes. Br J Sports Med. 2016;50(2):84–92.

    Article  PubMed  Google Scholar 

  8. 8.

    Conrozier T, Mathieu P, Schott AM, Laurent I, Hajri T, Crozes P, et al. Factors predicting long-term efficacy of Hylan GF-20 viscosupplementation in knee osteoarthritis. Joint Bone Spine. 2003;70(2):128–33.

    Article  Google Scholar 

  9. 9.

    Altman RD, Dasa V, Takeuchi J. Review of the mechanism of action for Supartz FX in knee osteoarthritis. Cartilage. 2018;9(1):11–20.

    CAS  Article  PubMed  Google Scholar 

  10. 10.

    Pradal J, Maudens P, Gabay C, Seemayer CA, Jordan O, Allemann E. Effect of particle size on the biodistribution of nano- and microparticles following intra-articular injection in mice. Int J Pharm. 2016;498(1–2):119–29.

    CAS  Article  PubMed  Google Scholar 

  11. 11.

    Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957;16(4):494–502.

    CAS  Article  Google Scholar 

  12. 12.

    Esenyel C, Demirhan M, Esenyel M, Sonmez M, Kahraman S, Senel B, et al. Comparison of four different intra-articular injection sites in the knee: a cadaver study. Knee Surg Sports Traumatol Arthrosc. 2007;15(5):573–7.

    Article  PubMed  Google Scholar 

  13. 13.

    Wind WM Jr, Smolinski RJ. Reliability of common knee injection sites with low-volume injections. J Arthroplast. 2004;19(7):858–61.

    Article  Google Scholar 

  14. 14.

    Podraza JT, White SC. Effect of knee flexion angle on ground reaction forces, knee moments and muscle co-contraction during an impact-like deceleration landing: implications for the non-contact mechanism of ACL injury. Knee. 2010;17(4):291–5.

    Article  PubMed  Google Scholar 

  15. 15.

    Makinejad MD, Abu Osman NA, Abu Bakar Wan Abas W, Bayat M. Preliminary analysis of knee stress in full extension landing. Clinics. 2013;68(9):1180–8.

    Article  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Abid M, Mezghani N, Mitiche A. Knee joint biomechanical gait data classification for knee pathology assessment: a literature review. Appl Bionics Biomech. 2019;2019:1–14.

    Article  Google Scholar 

  17. 17.

    Elbaz A, Mor A, Segal G, Debi R, Shazar N, Herman A. Novel classification of knee osteoarthritis severity based on spatiotemporal gait analysis. Osteoarthr Cartil. 2014;22(3):457–63.

    CAS  Article  Google Scholar 

  18. 18.

    McCarthy I, Hodgins D, Mor A, Elbaz A, Segal G. Analysis of knee flexion characteristics and how they alter with the onset of knee osteoarthritis: a case control study. BMC Musculoskelet Disord. 2013;14(1):169.

    Article  PubMed  PubMed Central  Google Scholar 

  19. 19.

    Bergmann G, Bender A, Graichen F, Dymke J, Rohlmann A, Trepczynski A, et al. Standardized loads acting in knee implants. PLoS One. 2014;9(1):e86035.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  20. 20.

    Sutton LG, Werner FW, Haider H, Hamblin T, Clabeaux JJ. In vitro response of the natural cadaver knee to the loading profiles specified in a standard for knee implant wear testing. J Biomech. 2010;43(11):2203–7.

    Article  PubMed  Google Scholar 

  21. 21.

    Peterfy CG, Guermazi A, Zaim S, Tirman PF, Miaux Y, White D, et al. Whole-organ Magnetic Resonance Imaging Score (WORMS) of the knee in osteoarthritis. Osteoarthr Cartil. 2004;12(3):177–90.

    CAS  Article  Google Scholar 

  22. 22.

    Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Arthritis Rheum. 1986;29(8):1039–49.

    CAS  Article  Google Scholar 

  23. 23.

    Wluka AE, Stuckey S, Snaddon J, Cicuttini FM. The determinants of change in tibial cartilage volume in osteoarthritic knees. Arthritis Rheum. 2002;46(8):2065–72.

    Article  PubMed  Google Scholar 

  24. 24.

    Berenbaum F, Grifka J, Cazzaniga S, D’Amato M, Giacovelli G, Chevalier X, et al. A randomised, double-blind, controlled trial comparing two intra-articular hyaluronic acid preparations differing by their molecular weight in symptomatic knee osteoarthritis. Ann Rheum Dis. 2012;71(9):1454–60.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  25. 25.

    Wittes J. Sample size calculations for randomized controlled trials. Epidemiol Rev. 2002;24(1):39–53.

    Article  Google Scholar 

  26. 26.

    Jackson DW, Evans NA, Thomas BM. Accuracy of needle placement into the intra-articular space of the knee. J Bone Joint Surg Am. 2002;84-A(9):1522–7.

    Article  Google Scholar 

  27. 27.

    Henrotin Y, Raman R, Richette P, Bard H, Jerosch J, Conrozier T, et al. Consensus statement on viscosupplementation with hyaluronic acid for the management of osteoarthritis. Semin Arthritis Rheum. 2015;45(2):140–9.

    CAS  Article  PubMed  Google Scholar 

  28. 28.

    Bellamy N, Campbell J, Stevens J, Pilch L, Stewart C, Mahmood Z. Validation study of a computerized version of the Western Ontario and McMaster Universities VA3.0 Osteoarthritis Index. J Rheumatol. 1997;24(12):2413–5.

    CAS  PubMed  Google Scholar 

  29. 29.

    Lequesne MG, Mery C, Samson M, Gerard P. Indexes of severity for osteoarthritis of the hip and knee. Validation--value in comparison with other assessment tests. Scand J Rheumatol Suppl. 1987;65:85–9.

    CAS  Article  Google Scholar 

  30. 30.

    van der Weegen W, Wullems JA, Bos E, Noten H, van Drumpt RA. No difference between intra-articular injection of hyaluronic acid and placebo for mild to moderate knee osteoarthritis: a randomized, controlled, double-blind trial. J Arthroplast. 2015;30(5):754–7.

    Article  Google Scholar 

  31. 31.

    Maheu E, Bannuru RR, Herrero-Beaumont G, Allali F, Bard H, Migliore A. Why we should definitely include intra-articular hyaluronic acid as a therapeutic option in the management of knee osteoarthritis: results of an extensive critical literature review. Semin Arthritis Rheum. 2019;48(4):563–72.

    Article  PubMed  Google Scholar 

  32. 32.

    Legre-Boyer V. Viscosupplementation: techniques, indications, results. Orthop Traumatol Surg Res. 2015;101(1 Suppl):S101–8.

    CAS  Article  PubMed  Google Scholar 

  33. 33.

    Toda Y, Tsukimura N. A comparison of intra-articular hyaluronan injection accuracy rates between three approaches based on radiographic severity of knee osteoarthritis. Osteoarthr Cartil. 2008;16(9):980–5.

    CAS  Article  Google Scholar 

  34. 34.

    Jackson DW, Simon TM. Intra-articular distribution and residence time of Hylan A and B: a study in the goat knee. Osteoarthr Cartil. 2006;14(12):1248–57.

    CAS  Article  Google Scholar 

  35. 35.

    Luo Y, Lin B, Xiao J, Shi Z. Iohexol tracing of hyaluronic acid distribution in the knee joint cavity of rabbits. Nan fang yi ke da xue xue bao J South Med Univ. 2015;35(6):820–5.

    CAS  Google Scholar 

  36. 36.

    Brandt KD, Smith GN Jr, Simon LS. Intraarticular injection of hyaluronan as treatment for knee osteoarthritis: what is the evidence? Arthritis Rheum. 2000;43(6):1192–203.<1192::AID-ANR2>3.0.CO;2-L.

    CAS  Article  PubMed  Google Scholar 

  37. 37.

    Bonnevie ED, Delco ML, Bartell LR, Jasty N, Cohen I, Fortier LA, et al. Microscale frictional strains determine chondrocyte fate in loaded cartilage. J Biomech. 2018;74:72–8.

    Article  PubMed  PubMed Central  Google Scholar 

  38. 38.

    Bonnevie ED, Galesso D, Secchieri C, Bonassar LJ. Frictional characterization of injectable hyaluronic acids is more predictive of clinical outcomes than traditional rheological or viscoelastic characterization. PLoS One. 2019;14(5):e0216702.

    Article  PubMed  PubMed Central  Google Scholar 

  39. 39.

    Singh A, Corvelli M, Unterman SA, Wepasnick KA, McDonnell P, Elisseeff JH. Enhanced lubrication on tissue and biomaterial surfaces through peptide-mediated binding of hyaluronic acid. Nat Mater. 2014;13(10):988–95.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  40. 40.

    Tobetto K, Nakai K, Akatsuka M, Yasui T, Ando T, Hirano S. Inhibitory effects of hyaluronan on neutrophil-mediated cartilage degradation. Connect Tissue Res. 1993;29(3):181–90.

    CAS  Article  PubMed  Google Scholar 

  41. 41.

    Arrich J, Piribauer F, Mad P, Schmid D, Klaushofer K, Müllner M. Intra-articular hyaluronic acid for the treatment of osteoarthritis of the knee: systematic review and meta-analysis. CMAJ. 2005;172(8):1039–43.

    Article  Google Scholar 

  42. 42.

    Brown TJ, Laurent UB, Fraser JR. Turnover of hyaluronan in synovial joints: elimination of labelled hyaluronan from the knee joint of the rabbit. Exp Physiol. 1991;76(1):125–34.

    CAS  Article  PubMed  Google Scholar 

  43. 43.

    Altman RD, Manjoo A, Fierlinger A, Niazi F, Nicholls M. The mechanism of action for hyaluronic acid treatment in the osteoarthritic knee: a systematic review. BMC Musculoskelet Disord. 2015;16:321.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  44. 44.

    Komatsu S, Iwata H, Nabeshima T. Studies on the kinetics, metabolism and re-utilisation after intra-articular administration of hyaluronan to rabbits. Arzneimittel-Forschung. 1999;49(5):427–33.

    CAS  Article  PubMed  Google Scholar 

  45. 45.

    Akmal M, Singh A, Anand A, Kesani A, Aslam N, Goodship A, et al. The effects of hyaluronic acid on articular chondrocytes. J Bone Joint Surg Br Vol. 2005;87(8):1143–9.

    CAS  Article  Google Scholar 

  46. 46.

    Chareancholvanich K, Pornrattanamaneewong C, Narkbunnam R. Increased cartilage volume after injection of hyaluronic acid in osteoarthritis knee patients who underwent high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc. 2014;22(6):1415–23.

    Article  PubMed  Google Scholar 

  47. 47.

    Grenier S, Donnelly PE, Gittens J, Torzilli PA. Resurfacing damaged articular cartilage to restore compressive properties. J Biomech. 2015;48(1):122–9.

    Article  PubMed  Google Scholar 

  48. 48.

    Hideki Sato MD, Takahashi T, Ide H, Toshiyuki Fukushima MD, Minoru Tabata MD, Fusao Sekine MD, et al. Antioxidant activity of synovial fluid, hyaluronic acid, and two subcomponents of hyaluronic acid. synovial fluid scavenging effect is enhanced in rheumatoid arthritis patients. Arthritis Rheum. 1988;31(1):63.

    Article  Google Scholar 

  49. 49.

    Larsen NE, Lombard KM, Parent EG, Balazs EA. Effect of hylan on cartilage and chondrocyte cultures. J Orthop Res. 1992;10(1):23–32.

    CAS  Article  Google Scholar 

  50. 50.

    Ghosh P, Guidolin D. Potential mechanism of action of intra-articular hyaluronan therapy in osteoarthritis: are the effects molecular weight dependent? Semin Arthritis Rheum. 2002;32(1):10–37.

    CAS  Article  Google Scholar 

  51. 51.

    Gomis A, Pawlak M, Balazs EA, Schmidt RF, Belmonte C. Effects of different molecular weight elastoviscous hyaluronan solutions on articular nociceptive afferents. Arthritis Rheum. 2004;50(1):314–26.

    CAS  Article  PubMed  Google Scholar 

  52. 52.

    Pozo MA, Balazs EA, Belmonte C. Reduction of sensory responses to passive movements of inflamed knee joints by hylan, a hyaluronan derivative. Exp Brain Res. 1997;116(1):3–9.

    CAS  Article  Google Scholar 

  53. 53.

    Haris M, Singh A, Reddy S, Bagga P, Kneeland JB, Tjoumakaris FP, et al. Characterization of viscosupplementation formulations using chemical exchange saturation transfer (ViscoCEST). J Transl Med. 2016;14:92.

    Article  Google Scholar 

Download references


We thank the staff of anatomy department for their important support and help, the National Natural Science Foundation of China (NNSFC) for the fund source support, and Tahsin Tarik Torsha and Dr. Cindy for the generous help in modifying the grammar and spelling.

Author information



Corresponding author

Correspondence to Jun Xiao.

Ethics declarations

Conflict of interest

Jun Xiao receives a grant (81101389) from the National Natural Science Foundation of China (NNSFC) to support the current research, but there is no financial conflict to disclose. The rest of the authors also certify that he or she has no commercial associations that might pose a conflict of interest in connection with the submitted article.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Xiao, J., Hu, Y., Huang, L. et al. Injection route affects intra-articular hyaluronic acid distribution and clinical outcome in viscosupplementation treatment for knee osteoarthritis: a combined cadaver study and randomized clinical trial. Drug Deliv. and Transl. Res. 11, 279–291 (2021).

Download citation


  • Hyaluronate
  • Intra-articular distribution
  • Knee
  • Osteoarthritis
  • Medial midpatellar portal