Keywords
hyaluronic acid - osteoarthritis - knee - viscosupplementation - Hymovis® - viscoelastic
Introduction
Osteoarthritis (OA) is the most common form of arthritis in industrialized countries,
with prevalence in the general population of Caucasian ethnicity largely stable for
decades between 3.4 and 6.1%.[1] The prevalence increases progressively with age and becomes significant after 60
years (range, 10–15%), with women affected more frequently.[2] The most commonly affected joints are the hip and knee. Patients with OA experience
functional disability and progressive deterioration in the quality of life (QoL).
The impact on function and QoL depend on the number of joints involved and the severity
of joint damage. Mobility can be seriously impaired in patients with severe bilateral
knee and/or hip OA. Such patients are at an increased risk of accidental falls that
can result in fractures, emergency surgery, permanent disability, and premature death.[3] Recent studies show that functional disability from hip or knee OA is associated
with a higher incidence of metabolic disease (e.g., diabetes, obesity) and cardiovascular
disease (e.g., heart attack, stroke, atherothrombotic vascular disease) due to physical
inactivity and weight gain.[4]
[5]
The causes of OA are only partially known, even though the major risk factors and
pathogenic mechanisms of cartilage damage have been identified in recent decades.[6]
[7] Articular cartilage wear is a common denominator, into which the effects of various
risk factors are summed. For example, knee OA is approximately 10 times more common
in weight lifters and 5 times more common in runners,[8] as a result of repeated microtrauma to cartilage. These values further increase
in individuals with a history of the ruptured meniscus or cruciate ligament. In such
patients, altered biomechanics can cause dynamic loads to impinge on the cartilage
along nonphysiological pressure lines.[9] Recently, obesity has emerged as a potent promoter of hyaline cartilage degeneration.
When overweight begins in childhood, the risk of early (<50 years) knee and hip OA
is much higher compared with that in the normal weight population.[10]
[11]
[12]
OA treatment may comprise conservative interventions (physical therapy), medication
(local or systemic) or surgery.[13] The guidelines of the Osteoarthritis Research Society International (OARSI) suggest
that interventions should be progressive and proportionate to disease severity.[14] Pharmacological interventions may include nonsteroidal anti-inflammatory drugs (NSAIDs),
pure analgesics (acetaminophen) and disease-modifying therapies.[15]
[16]
The costs associated with pharmaceutical therapy and arthroplasty for OA are substantial
and increasing. Piscitelli et al have calculated that between 2001 and 2005, the knee
arthroplasty interventions in Italy increased by 16.6% for men and 12.4% for women,
with an exponential increase in direct and indirect economic costs.[17] Consequently, also on the basis of economic considerations, there is a tendency
to use measures that prevent or delay recourse to arthroplasty.
Intra-articular infiltration (viscosupplementation) with derivatives of hyaluronic
acid (HA), a polysaccharide polymer naturally present in articular cartilage, is a
therapeutic measure that restores the rheological properties of synovial fluid and
stimulates the chondrocytes that produce endogenous HA.[18]
[19] Whereas intra-articular HA injection is included in several major guidelines on
managing OA,[14]
[20] conflicting evidence for its efficacy has resulted in a lack of agreement among
various guidelines regarding its use.[21] The discrepancy between the beneficial effects of this procedure observed in clinical
practice and guideline recommendations may result from study characteristics, such
as inclusion criteria and the form of HA used.[22]
[23]
[24] Currently, several HA formulations are approved for clinical use in Europe and the
United States. These formulations differ in the origin of the HA and production methods
used, in their chemical–physical properties, joint space half-life, rheological properties,
as well as their administration schedules and cost.[25]
[26]
HYADD® 4 (Hymovis®; Fidia Farmaceutici, Abano Terme, Italy) is a derivative of HA
obtained by controlled chemical synthesis, with a molecular weight between 500 and
730 kDa. In Hymovis®, 2% of the carboxy radicals on the glucuronic acid present in
the polysaccharide chain are conjugated with an aliphatic amine (hexadecylamine).
This chemical structure makes the three-dimensional lattice of Hymovis® particularly
hygroscopic and forms a hydrogel with excellent viscoelastic, lubricating, and rheological
properties.
Hymovis® has been shown to significantly delay cartilage degeneration in a rabbit
model of anterior cruciate ligament rupture.[27] Moreover, controlled clinical studies with Hymovis® have demonstrated its superiority
to Hyalgan (Fidia Farmaceutici) both in terms of reduction in synovial hyperplasia
and increased synthesis of high-molecular-weight HA by chondrocytes, with fewer Hymovis®
infiltrations.[28]
[29] The present study investigated the efficacy and safety of viscosupplementation with
Hymovis® in patients with knee OA. The hypothesis of the study was that viscosupplementation
with Hymovis® in patients with knee OA improves subjective outcome and knee function
with no significant adverse events.
Methods
Study Design
This was a retrospective observational study designed to evaluate the clinical efficacy
and tolerability of two intra-articular Hymovis® infiltrations (24 mg/3 mL) administered
1 week apart in everyday clinical practice. Data were obtained from the ANTIAGE national
register, a nonprofit database of clinical data on the effects of ultrasound-guided
intra-articular viscosupplementation. All patients were evaluated at 1 month and subsequently
at 3, 6, 9, and 12 months ([Table 1]).
Table 1
Monitoring of clinical parameters during the study
|
End point
|
Baseline
(n = 937)
|
Month 1
(n = 821)
|
Month 3
(n = 743)
|
Month 6
(n = 698)
|
Month 9
(n = 137)
|
Month 12
(n = 106)
|
|
Pain at rest (VAS score)
|
55.9 ± 11.4
|
31.8 ± 9.7
|
24.9 ± 10.2
|
24.1 ± 9.1
|
25.3 ± 7.3
|
25.9 ± 6.9
|
|
Pain on movement (VAS score)
|
72.1 ± 9.8
|
48.3 ± 6.7
|
38.2 ± 7.7
|
37.9 ± 7.3
|
38.4 ± 6.5
|
38.9 ± 11.5
|
|
WOMAC score
|
56.8 ± 8.7
|
45.7 ± 7.9
|
27.6 ± 10.3
|
24.8 ± 8.8
|
N/A
|
N/A.
|
|
Acetaminophen/NSAIDs use
|
48.8%
|
27.4%
|
19.4%
|
19.6%
|
N/A
|
N/A
|
|
Patient satisfaction
|
|
Worsening (%)
|
|
2.4
|
3.2
|
4.8
|
|
|
|
Unchanged (%)
|
|
9.3
|
8.9
|
9.6
|
|
|
|
Improved (%)
|
|
88.3
|
87.9
|
85.6
|
|
|
Abbreviations: N/A, not available; NSAID, nonsteroidal anti-inflammatory drug; VAS,
visual analog scale; WOMAC, Western Ontario McMasters University Osteoarthritis Index.
Patients
Eligible patients met the following evaluation criteria: age ranging from 41 to 80
years; symptomatic knee arthritis diagnosed according to the American College of Rheumatology
(ACR) criteria;[30] Kellgren–Lawrence disease severity grades II to IV; disease duration of ≥2 years;
and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) total score
>30 points. Patients were excluded if they had other chronic rheumatic or autoimmune
diseases, cancers, joint line collapse, a history of allergy to HA or its derivatives.
Also, patients receiving anticoagulants and those who had received viscosupplementation
or corticosteroids (CS) infiltration in the previous 12 months were excluded.
Outcomes Measures
The study duration was 6 months for all preestablished end points, with a 12-month
extension for the parameters pain at rest and pain in motion. Data on these two parameters
were available at month 9 (n = 137) and month 12 (n = 106).
The reference standard for evaluating patients with hip or knee OA is the WOMAC questionnaire,
which detects clinically relevant data in three specific areas: joint pain at rest
and in motion, articular function, and joint stiffness. The total WOMAC score can
range from 0 (no impairment) to 96 points (maximum impairment). Conventionally, treatments
are considered efficacious if they provide a clinically significant reduction in WOMAC
score of ≥20 points.[31] The WOMAC questionnaire measures pain using a Likert scale, where 0 corresponds
to no pain and 4 to maximum pain; it also provides an accurate assessment of functionality
and joint stiffness by measuring the ability to perform specific movements. Scores
below 25 indicate mild impairment, while scores above 50 indicate moderate-to-severe
impairment. Scores above 75 indicate severe joint impairment.
Pain at rest and pain in motion were also evaluated on a visual analog scale (VAS)
in millimeters, with scores from 0 to 100 mm.
Because 48.8% of included patients were using acetaminophen and/or NSAIDs at baseline,
a reduction in the number of patients requiring these medications was considered a
reliable endpoint of clinical efficacy.
All patients were asked to answer a question on a subjective evaluation of the results,
in terms of pain reduction and joint function. The three options were: (1) improvement;
(2) worsening; and (3) no change.
All local and systemic adverse events (AEs) were recorded. AE associated with the
first infiltration were recorded during the first week, whereas those after the second
infiltration were recorded at months 1, 3, and 6.
Statistical Analysis
The changes from baseline for each end point under study were analyzed with the chi-square
test. Two measurements were performed for each parameter, and the mean ± standard
deviation (SD) reported. Excel spreadsheets were used for data management (Microsoft,
Redmond, Washington, United States). Statistical analysis was performed with Sigma
Stat software (SPSS 9; SPSS, Chicago, Illinois, United States). Significance was set
for values of p < 0.05 (5%).
Results
Data from 937 eligible patients, 534 women (57%), were extracted from the database
([Table 2]). Of these, 210 had bilateral disease. In total, 1,147 knee joints were treated,
for a total of 2,294 infiltrations. Mean disease duration was 6.8 ± 6.6 years. Disease
severity according to the Kellgren–Lawrence classification ranged between grade II
(38%) and grade IV (7%). Most patients (55%) had grade III disease, and 39.2% had
received the previous infiltration with HA or CS.
Table 2
Clinical and demographic characteristics of the study population
|
Item
|
Value
|
|
Age (y) (mean ± SD)
|
66 ± 11.4
|
|
Women (n [%])
|
534 (57%)
|
|
Men (n [%])
|
403 (43%)
|
|
BMI (kg/m2) (mean ± SD)
|
27.09 ± 3.6
|
|
Duration OA (y) (mean ± SD)
|
6.8 ± 6.6
|
|
Kellgren–Lawrence grade
|
|
Grade II (n [%])
|
356 (38%)
|
|
Grade III (n [%])
|
515 (55%)
|
|
Grade IV (n [%])
|
66 (7%)
|
|
Right knee (n [%])
|
484 (51.7%)
|
|
Left knee (n [%])
|
243 (25.9%)
|
|
Bilateral (n [%])
|
210 (22.4%)
|
|
Previous use of HA or CS (mean ± SD)
|
39.2 ± 5.6
|
|
Pain at rest (VAS score) (mean ± SD)
|
55.9 ± 11.3
|
|
Pain on movement (VAS score) (mean ± SD)
|
72.1 ± 9.8
|
|
Acetaminophen/NSAIDs use ≥2 times/wk (%) (mean ± SD)
|
48.8 ± 6.1
|
|
Comorbidity (%) (mean ± SD)
|
19.5 ± 3.9
|
|
WOMAC
|
|
Pain (Likert 0–4) (mean ± SD)
|
11.5 ± -2.7
|
|
Physical function (mean ± SD)
|
39.8 ± 8.3
|
|
Stiffness (mean ± SD)
|
5.5 ± 3.7
|
|
Total score (mean ± SD)
|
56.8 ± 9.9
|
Abbreviations: CS, corticosteroids; HA, hyaluronic acid; NSAID, nonsteroidal anti-inflammatory
drug; SD, standard deviation; VAS, visual analog scale; WOMAC, Western Ontario McMasters
University Osteoarthritis Index.
Impairment at baseline measured with the WOMAC questionnaire was 56.8 ± 9.9 points
out of a possible 96. Accordingly, 48.8% ± 6.1% made regular use (≥2 times/week) of
NSAIDs and/or acetaminophen to relieve pain when performing daily activities.
At the first follow-up (1 month), 821/937 patients (87.6%) were evaluable; at the
second follow-up, 743/937 patients (79.3%) were evaluated; in the sixth month, the
698/937 patient data were available (74.5%). By month 12, it was possible to examine
the data relating to only two parameters (pain at rest and in motion) in a subpopulation
of 106 patients ([Table 1]). The dropout was due to the failure of patients to present for scheduled visits
or incomplete compilation of medical records, with missing data relevant to the clinical
evaluation (e.g., missing annotation about acetaminophen/NSAIDs use). Patients with
bilateral knee OA were excluded if results were discordant between the two affected
joints because the protocol was set to report only one answer per patient. After the
12-month extension, 106 patients had evaluable data for the pain outcomes.
The effect on pain was rapid, with a reduction of 24.1 points (43.1%) after the first
month ([Fig. 1]). The analgesic action lasted until the third month, a decrease of 31 points (55.4%),
remaining largely stable between the second and third visits. At the end of the first
phase of the study (6 months), pain at rest was reduced by 56.8%. The same trend was
observed for pain on moving, with a slightly smaller reduction from baseline in the
sixth month of 47.4%. These changes were statistically significant (p < 0.05) for both parameters already after the first month.
Fig. 1 Trend in pain scores and WOMAC scores after two Hymovis® infiltrations. WOMAC, Western
Ontario and McMaster Universities Osteoarthritis Index.
For the pain parameters, available data allowed assessments to a total of 12 months
for a small cohort of patients (n = 106). Neither of the pain parameters had clinically or statistically significant
changes between months 6 and 12. Compared with baseline, the reduction in pain at
rest was 54.7% at month 9 (p < 0.05) and 53.6% at month 12 (p < 0.05). Compared with the maximum reduction in resting pain recorded at month 3
(56.8%), the loss of efficacy through month 12 was 3.2% (p = nonsignificant). A similar trend was observed for pain in motion, with reductions
equal to 47.4 and 46.0% at months 6 and 12, respectively. Also, in this case, the
difference between months 6 and 12 is not statistically significant (p > 0.91).
The mean WOMAC (SD) score was reduced by 17.1 points (30.1%) 1 month after the second
infiltration (p < 0.05), and was reduced by 24.8 points (56.3%) after 6 months (p < 0.05). The mean (SD) score after 6 months was 24.8 ( ± 8.8) points, corresponding
to mild impairment. It was not possible to assess the WOMAC score during the extension
to 12 months. At the first visit, only 225/821 evaluable patients (27.4%) continued
to take acetaminophen or NSAIDs at least twice a week ([Fig. 2]).
Fig. 2 The use of NSAIDs or acetaminophen during the study. NSAID, nonsteroidal anti-inflammatory
drug.
At 6 months, only 137 patients continued to take NSAIDs or acetaminophen (19.6%) with
a frequency of more than two times per week. A subanalysis conducted on 561 patients
who had stopped taking painkillers with this frequency, revealed that 178 patients
had not taken any such medication from the second month until study end; these were
classified as “drug-free.” The reduction from baseline in use was highest in the third
month (19.4%). These data suggest that the effect of viscosupplementation on the use
of anti-inflammatory and analgesic drugs is both rapid and durable.
Data on the pain response and the number of patients taking NSAIDs/acetaminophen correlate
well with the degree of subjective satisfaction with therapy: 3 months after infiltration,
87.9% of the patients reported an improvement in symptoms ([Fig. 3]). This improvement remained largely stable over the first 6 months of observation.
Fig. 3 The degree of patient satisfaction with therapy.
AEs were reported in 105 patients (11.2%). Arthralgias were the most frequent AEs,
and they had a relatively rapid onset of 12 to 36 hours, rarely lasting more than
2 days. No AEs were observed after the first 4 weeks from infiltration, suggesting
that surveillance should be focused in this time span. None of the AEs prevented the
second infiltration, which was planned at a distance of 1 week from the first. There
are no recorded cases of allergic reactions or systemic effects of the drug.
Discussion
The effects of viscosupplementation with HA are supported by numerous experimental
and clinical studies and several meta-analyses, although there is no unanimous agreement
on the effect size.[25]
[32] In a recent consensus organized in Italy, 52 clinicians with expertise in intra-articular
infiltration agreed that the viscosupplementation with HA is effective and appropriate
in patients with Kellgren–Lawrence grade II to IV knee OA.[33] In vitro studies and experiments in animal models have demonstrated the efficacy
and tolerability of Hymovis®.[27] Our retrospective study demonstrated that two administrations of Hymovis® at a distance
of 1 week not only reduce pain at rest and pain in the movement for approximately
1 year but also provide considerable sparing of NSAIDs and/or acetaminophen use. This
sparing effect was documented for at least 6 months. However, the favorable effects
on the pain that we documented for up to 12 months suggest that the medication-sparing
effect may last longer. The lack of corroborating data from the WOMAC score at 12
months does not invalidate the significant results observed on resting and movement-related
pain.
Because patients with symptomatic OA are elderly and receiving multiple drug therapies,
a reduction in NSAID consumption may results in a lower risk of AEs, both direct events
(e.g., gastrointestinal bleeding, ulcers from NSAIDs) or indirect events caused by
pharmacological interference (e.g., with antihypertensive drugs, cortisone). Based
on these results, it may be appropriate to consider treating also patients with grade
I knee OA who are at an increased risk of pharmacological interference or NSAID-induced
gastropathy, in accordance with the opinions expressed by a significant proportion
of the participants in the aforementioned clinical consensus.[33] Regarding overall tolerability, we recorded a slightly lower AE rate than that reported
by other authors, although the average age of our sample is higher.[16]
As with all retrospective studies, ours also has limitations related to design, but
these are offset by a large number of evaluable patients and the long duration of
the observation period.
Conclusion
In conclusion, our results indicate that Hymovis® is effective and safe in patients
with mild-to-severe knee OA. Prospective studies with adequate population size and
design are needed to better assess clinical effect size, optimal infiltration schedule,
and criteria for patient selection. We await the results of the ongoing prospective
randomized, double-blind efficacy study (Safety and Effectiveness Study of a Non-Crosslinked
HA Alkylamide HYADD® 4 Hydrogel for Osteoarthritis of the Knee; ClinicalTrials.gov
NCT02187549).
