Do intra-articular hyaluronic acid injections delay total knee replacement in patients with osteoarthritis

Nearly one in five people over 45 years old have sought treatment for knee osteoarthritis (OA) and 98% of initial knee replacements are due to OA (Arthritis Research UK, 2013). Due to the growing prevalence of knee OA and the medical costs associated with total knee replacement (TKR), there is increasing interest in the search for minimally invasive treatments that can delay OA progression. Intra-articular hyaluronic acid (IA-HA) injections is one such treatment that is frequently used. However, the available literature is varied about the effectiveness of IA-HA injections.

A retrospective and observational study conducted in France between 2006 and 2013 by Delbarre et al. sought to compare the delay from diagnosis of knee OA to TKR between patients who received IA-HA and those who did not (Delbarre, et al., 2017).

Methods and patient selection

Patients were selected from two medico-administrative databases which could be linked together using a common patient identifier. From these databases, information on a patient’s ambulatory care, including IA injections, and date of TKR could be determined and the association explored.

The databases do not contain lab test or X-ray results and therefore, an algorithm was established to determine which patients were likely to have knee OA. The inclusion criteria were as follows (Fig 1):

  • At least one X-ray or arthroscopy of the knee followed by an IA injection (corticosteroid (CS) or HA) within 12 months
  • Aged 50 or over at the time of the X-ray or arthroscopy
  • The injection was prescribed by a rheumatologist, an orthopaedic surgeon, physician specialising in physical medicine and rehabilitation or a GP

The date of the X-ray or arthroscopy was used as a proxy to date the diagnosis of knee OA.

Adjustment variables that were considered included socio-demographics, general health and knee OA conditions, such as initial prescriber’s speciality, mean annual number of visits to specialist physicians, and number of knees with OA.

Based on these criteria, the study population consisted of 14,782 patients, of which 5306 had CS injections only (non-HA group) and 9476 had at least one HA injection (HA group) during the follow-up period. 1662 patients had TKR before the end of the study (31st December 2013).

Fig 1. Flow-chart of patients’ selection. IA: intra-articular, HA: hyaluronic acid, CS: corticosteroids, TKR: total knee replacement. HA group: patients who received at least one IA HA injection during their follow-up period, N-HA group: patients who received only CS as IA injections. The four specialists are rheumatologist, orthopaedic surgeon, physical medicine and rehabilitation practitioner and general practitioner.

Results

The bivariate analysis performed on the patients who had TKR shows that the mean time from diagnosis to TKR was significantly higher in the HA group than the non-HA group (Fig 2).

Everything else being equal, the restricted mean survival time from diagnosis of knee OA to TKR was 842 days for the HA-group and 625 days for the non-HA group (95% CI for both groups).

At each time point, the mean time without TKR was significantly higher in the HA group – from +57 days at 1 year and +217 days at 7.5 years after diagnosis.

These findings support the effectiveness of intra-articular hyaluronic acid injections in delaying total knee replacement.

A retrospective longitudinal study involving knee osteoarthritis patients

Fig 2. Kaplan-Meier curves of time without TKR for HA and non-HA groups. HA group: patients who received at least one IA HA injection during their follow-up period, non-HA group: patients who received only CS as IA injections, TKR: total knee replacement, IA: intra-articular, HA: hyaluronic acid, CS: corticosteroids.

References

  1. Arthritis Research UK, 2013. Osteoarthritis in general practice. [Online] [Accessed 15 August 2019].
  2. Delbarre, A. et al., 2017. Do intra-articular hyaluronic acid injections delay total knee replacement in patients with osteoarthritis – A Cox model analysis. PLoS One, 11(12), p. e0187227.