Hylan G-F 20 injections have been established as a safe and effective approach for the treatment of knee osteoarthritis (OA) [1][2][3]. Side effects of erythema, joint swelling, decreased range of motion, and pain at the site of injection are variable in incidence, with studies reporting between 1–21%, particularly in those receiving additional treatments [1][3] [4][5][6] [7][8][9][10]. Although rare, more serious side effects have been documented and include severe acute inflammatory reaction (SAIR) also known as pseudo-sepsis, pseudo-gout, and injection induced calcium pyrophosphate dihydrate arthritis [11] [12][13][14][15][16][17][18]. Pseduo-sepsis is defined more specifically as an inflammatory reaction causing effusion in the knee that’s not attributable to sepsis, gout/pseudo-gout or a traumatic event. Treatment for the side effects has typically involved non-steroidal anti-inflammatory medications and intra-articular corticosteroid injections, with patients experiencing complete or near complete symptomatic recovery [1][11][18]. It is unclear why some patients remain symptom free after hylan G-F 20 injections while others respond poorly, but it is important to identify the ideal treatment modality to counter these side effects when an adverse event occurs.

The patient is a 41-year-old female with a 10-year history of bilateral knee pain previously treated with oral pain relievers including acetaminophen/hydrocodone, tramadol, gabapentin, and ibuprofen, physical therapy, and corticosteroid injections. Initial X-ray imaging of the knees revealed only bilateral mild narrowing of the medial weight bearing compartments. Magnetic resonance imaging scan revealed no ligamentous damage but some evidence of lateral patellar tracking disorder. The patient underwent a successful first round of Synvisc-One^® injections to her bilateral knees with no adverse reactions and adequate pain relief. Approximately, three months after the injections the patient experienced a traumatic injury to the left knee with follow-up MRI scan showing mild chondromalacia of the patella, with intact menisci, cruciate, and collateral ligaments. There was no large cartilaginous defect, lesion, or wear. After discussing treatment options the patient elected for a second course of Synvisc-One^® injections, which was administered bilaterally approximately five months after the initial Synvisc-One^® treatment.

Two days after the injection, the patient presented to the emergency department complaining of left knee swelling, erythema, warmth, and pain, with no right sided symptoms. The patient also reported subjective fevers and chills. Physical examination was positive for significant effusion, overlying erythema anteriorly, with intact muscle strength and sensation. The patient was unable to ambulate, with range of motion limited to 15–25 degrees due to pain. Supine X-rays showed a mild suprapatellar joint effusion and re-demonstrated mild medial joint space compartment narrowing, similar to the prior exam. The osseous structures were intact without fracture, dislocation, or bony erosions. A bedside arthrocentesis was performed in the emergency department and sent for fluid analysis with complete results listed in Table 1. Patient was observed in the emergency department overnight and subsequently discharged the following day.

On clinic follow-up five days later, the patient reported continued pain and decreased range of motion in the left knee, but had improvement in swelling. On physical exam, there was a moderate effusion with diffuse tenderness, primarily over the medial and lateral joint lines. There was painful range of motion of the knee with only 15-60 degrees of flexion. The knee was slightly warm, but not erythematous. A second arthrocentesis was performed with results listed in Table 1. The patient was to return within one week for re-evaluation.

The patient returned to clinic with unchanged left knee pain. Physical examination was positive for slight effusion, joint line tenderness medially and laterally, and pain with motion of her left knee. Exam was negative for significant instability to varus, valgus stress, or anterior, posterior testing. The patient was grossly neurovascularly intact. At this time, the treatment options were discussed including the possibility of a diagnostic arthroscopy. The patient opted to pursue surgery, as nonoperative treatment had been unsuccessful to that point. Surgery was scheduled for three weeks later.

A left knee diagnostic arthroscopy was performed. Intra-operative findings included outerbridge grade 0 changes of the majority of the knee. There was a small area on the medial aspect of the medial femoral condyle with grade three changes to the cartilage measuring approximately a centimeter in diameter. The same was seen on the lateral femoral condyle. There was a small area of fraying on the trochlea as well in the mid-portion, but no loose fragment to debride. She had a hypertrophic fat pad. The anterior cruciate ligament (ACL) and posterior cruciate ligament (PCL) were intact as well as the ligamentum submucosa. The medial meniscus was intact. The lateral meniscus had some fraying about the edges with a small radial tear as well. The findings on the scope were consistent with normal degeneration for a 41-year-old female, nothing pathologic was observed within the knee that would cause this reaction. After the procedure the patient was made weight bearing as tolerated, given crutches for balance, and instructed to use a continuous passive motion (CPM) machine. Intra-operative photographs of the changes seen can be found in Figure 1.

On follow-up one week postoperative, the patient reported improved subjective left knee symptoms with 45–50 degrees range of motion on the CPM. The patient also reported calf pain, but left lower extremity duplex scan ruled out DVT. Further treatment plans included physical therapy with scheduled follow-up in five weeks.

The patient returned to clinic as scheduled complaining that the original pain had returned to the pre-surgery level. She described the pain as 5/10, constant, located anteriorly, and aggravated by walking and stairs. She also complained of the occasional buckling sensation in the left knee. Only three physical therapy sessions had been completed due to scheduling issues. On exam, she had full range of motion of both knees from 0–120 degrees, with no clinical instability. No calf tenderness to palpation, no effusion, edema, erythema, or ecchymosis bilaterally. Portal sites were well healed. Our treatment plan involved increased physical therapy and pain control with over the counter medication. The patient was to continue to follow-up in our clinic for evaluation of pain, range of motion, and strength.

On four-month postoperative follow-up, the patient continued to experience knee pain despite a lack of further effusion or evidence of reaction. Physical examination revealed intact intra-articular structures with full active and passive range of motion and no point tenderness. Due to the unclear nature of the persistent pain, the patient has been referred to a pain clinic for further evaluation and treatment.

Pseudo-sepsis, also referred to as severe acute inflammatory reaction, is defined in literature as an instance of predominantly non-self-limiting severe inflammation causing pain within 72 hours, generally after a second exposure to an injection, with other causes such as sepsis, gout and pseudo-gout ruled out, and productive of a mononuclear rich effusion [15][16][17],[19][20][21]. Studies have also suggested that the presence of macrophages in the aspirate fluid may be a sign of immunologic sensitization, as was seen in the subject of our report [18]. This case study presents another example of possible immune related reaction to hylan G-F 20 as it parallels outcomes observed in multiple case reports in patients receiving the same treatment, although the underlying cause remains unclear [16][17][20].

Studies have suggested that injection technique may be to blame for a large number of the adverse reactions, up to 72%, but further data on the subject is limited and likely is not the cause of the more serious inflammatory effusions [19]. A second potential hypothesis involves a host-mediated response to foreign proteins. These antibodies would be targeted against the hyaluronan which is taken from chicken combs to produce the injection, or possibly the chemicals used to crosslink the hylan [8][22][23][24]. Studies have shown increased titers in animals previously treated with Synvisc^® suggesting that the involved cross-linked proteins may be antigenic if contaminated, however a small study with human subjects did not determine the reactions to be antibody mediated [25][26]. In our case, the occurrence of a significant reaction in only one of the two injected joints is further evidence against a systemic cause. Without a clear cause, treatment for our patient required additional measures than traditionally used in the treatment of a pseudoseptic reaction.

Goldberg and Coutts outline the basic approach for pseudo-sepsis from hylan GF-20 injection, starting with rest and cold compress in addition to analgesics or NSAIDs as needed [15]. More aggressive treatment includes intra-articular steroids as well as arthrocentesis, performed to both diagnose and provide relief [19]. Unfortunately, treatment modalities are seemingly limited to these in literature. There was no evidence on how to proceed in cases that were not adequately resolved with the above mentioned interventions outside of surgical intervention to address a chronic granuloma, which did not apply in this specific case [15]. As the joint in our patient was determined to be non-septic there was no need to administer antibiotics. An intra-articular steroid injection was also deferred in this case based on shared decision making with the patient, who wished to proceed directly to surgical intervention at this point.

Diagnostic arthroscopy with before mentioned additional procedures appeared to resolve her pain and persistent effusion. The patient’s physical examination at the end of six weeks improved overall, including a reduction in swelling and erythema with an associated increased range of motion, despite the continuation of pain symptoms. Based on these findings the surgical intervention to counteract the adverse reaction was justified on follow-up physical examination. Although the patient reports continued pain at a pre-surgery baseline the acute phase of the reaction appears to have passed and should allow for further physical therapy to improve strength and reduce pain.

Potential confounding aspects of this case to consider are the subjective nature of knee pain and the adherence to treatment. As the patient improved clinically on exam, one could argue that the interventions were successful despite the continuation of knee pain. One other intervention that could have been considered is an intra-articular corticosteroid injection postoperatively, which could have provided pain relief. The joint in question was no longer erythematous, swollen, or with decreased ROM and strength on testing although a limp and reported lack of functional capacity remained. The lack of adherence to a prescribed physical therapy regimen since the onset of the adverse reaction may also have contributed to the need for surgical intervention. Despite these drawbacks, this case still serves as an informative example of a treatment progression when dealing with a seemingly treatment refractory hylan GF-20 induced case of pseudo-sepsis.

Pseudoseptic reactions to hylan G-F 20 (Synvisc-One^®) injections are a common occurrence, particularly after a second or third round of treatment. Although distressing to the patient, the reactions are predominately controlled with non-steroidal anti-inflammatory medications and steroid injections. When a reaction persists despite conservative treatments, it may be necessary to advance to diagnostic surgical intervention. This decision is one that must be made in direct collaboration with the patient based on treatment history and available options.

1. Andrzejewski T, Golda W, Gruszka J, et al. Assessment of Synvisc treatment in osteoarthritis. Ortop Traumatol Rehabil 2003 Jun 30;5(3):379–90.   [Pubmed]    
2. Pavelka K, Uebelhart D. Efficacy evaluation of highly purified intra-articular hyaluronic acid (Sinovial^®) vs hylan G-F20 (Synvisc^®) in the treatment of symptomatic knee osteoarthritis. A double-blind, controlled, randomized, parallel-group non-inferiority study. Osteoarthritis Cartilage 2011 Nov;19(11):1294–300.   [CrossRef]   [Pubmed]    
3. Wobig M, Beks P, Dickhut A, Maier R, Vetter G. Open-label multicenter trial of the safety and efficacy of viscosupplementation with Hylan G-F 20 (Synvisc) in primary osteoarthritis of the knee. J Clin Rheumatol 1999;5(6 Suppl):S24–S31.    
4. Lee S, Park D, Chmell SJ. Viscosupplementation with hylan G-F 20 (Synvisc): Pain and mobility observations from 74 consecutive patients. J Knee Surg 2004 Apr;17(2):73–7.   [Pubmed]    
5. Clarke S, Lock V, Duddy J, Sharif M, Newman JH, Kirwan JR. Intra-articular hylan G-F 20 (Synvisc) in the management of patellofemoral osteoarthritis of the knee (POAK). Knee 2005 Jan;12(1):57–62.   [CrossRef]   [Pubmed]    
6. Leopold SS, Warme WJ, Pettis PD, Shott S. Increased frequency of acute local reaction to intra-articular hylan GF-20 (synvisc) in patients receiving more than one course of treatment. J Bone Joint Surg Am 2002 Sep;84–A(9):1619–23.   [CrossRef]   [Pubmed]    
7. Huskisson EC, Donnelly S. Hyaluronic acid in the treatment of osteoarthritis of the knee. Rheumatology (Oxford) 1999 Jul;38(7):602–7.   [Pubmed]    
8. Lussier A, Cividino AA, McFarlane CA, Olszynski WP, Potashner WJ, De Médicis R. Viscosupplementation with hylan for the treatment of osteoarthritis: Findings from clinical practice in Canada. J Rheumatol 1996 Sep;23(9):1579–85.   [Pubmed]    
9. Scale D, Wobig M, Wolpert W. Viscosupplementation of osteoarthritic knees with hylan: A treatment schedule study. Curr Ther Res 1994;55(3):220–32.    
10. Wobig M, Dickhut A, Maier R, Vetter G. Viscosupplementation with hylan G-F 20: A 26-week controlled trial of efficacy and safety in the osteoarthritic knee. Clin Ther 1998 May–Jun;20(3):410–23.   [CrossRef]   [Pubmed]    
11. Kroesen S, Schmid W, Theiler R. Induction of an acute attack of calcium pyrophosphate dihydrate arthritis by intra-articular injection of hylan G-F 20 (Synvisc). Clin Rheumatol 2000;19(2):147–9.   [CrossRef]   [Pubmed]    
12. Hammesfahr JF, Knopf AB, Stitik T. Safety of intra-articular hyaluronates for pain associated with osteoarthritis of the knee. Am J Orthop (Belle Mead NJ) 2003 Jun;32(6):277–83.   [Pubmed]    
13. Conduah AH, Baker CL, Baker CL. Managing joint pain in osteoarthritis: Safety and efficacy of hylan G-F 20. J Pain Res 2009 Jul 6;2:87–98.   [CrossRef]   [Pubmed]    
14. Tahiri L, Benbouazza K, Amine B, Hajjaj-Hassouni N. Acute pseudoseptic arthritis after viscosupplementation of the knee: A case report. Clin Rheumatol 2007 Nov;26(11):1977–9.   [CrossRef]   [Pubmed]    
15. Goldberg VM, Coutts RD. Pseudoseptic reactions to hylan viscosupplementation: Diagnosis and treatment. Clin Orthop Relat Res 2004 Feb;(419):130–7.   [CrossRef]   [Pubmed]    
16. Martens PB. Bilateral symmetric inflammatory reaction to hylan G-F 20 injection. Arthritis Rheum 2001 Apr;44(4):978–9.   [CrossRef]   [Pubmed]    
17. Bernardeau C, Bucki B, Lioté F. Acute arthritis after intra-articular hyaluronate injection: Onset of effusions without crystal. Ann Rheum Dis 2001 May;60(5):518–20.   [Pubmed]    
18. Pullman-Mooar S, Mooar P, Sieck M, Clayburne G, Schumacher HR. Are there distinctive inflammatory flares after hylan g-f 20 intraarticular injections? J Rheumatol 2002 Dec;29(12):2611–4.   [Pubmed]    
19. Puttick MP, Wade JP, Chalmers A, Connell DG, Rangno KK. Acute local reactions after intraarticular hylan for osteoarthritis of the knee. J Rheumatol 1995 Jul;22(7):1311–4.   [Pubmed]    
20. Rees JD, Wojtulewski JA. Systemic reaction to viscosupplementation for knee osteoarthritis. Rheumatology (Oxford) 2001 Dec;40(12):1425–6.   [CrossRef]   [Pubmed]    
21. Allen E, Krohn K. Adverse reaction to Hylan GF-20. J Rheumatol 2000 Jun;27(6):1572.   [Pubmed]    
22. Bucher W, Otto T, Hamburger MI. Differentiation of hyaluronate products by qualitative differences in their immunogenicity in rabbits: possible mechanism for product-specific severe adverse reactions? Comment on the article by Martens. Arthritis Rheum 2002 Sep;46(9):2543–4; author reply 2544.   [CrossRef]   [Pubmed]    
23. Schiavinato A, Finesso M, Cortivo R, Abatangelo G. Comparison of the effects of intra-articular injections of Hyaluronan and its chemically cross-linked derivative (Hylan G-F20) in normal rabbit knee joints. Clin Exp Rheumatol 2002 Jul–Aug;20(4):445–54.   [Pubmed]    
24. Balazs E, Leshchiner A, Leshchiner A, Larsen N, Band P. Hylan preparation and method of recovery thereof from animal tissues. US patent: 1992.5099013.    
25. Ohara N, Matsuoka C, Nagata T, et al. Topical reactions against intracutaneous injection of three hyaluronan products, Artz^®, Synvisc^® and Durolane^® in guinea pigs. Japanese Pharmacol Ther 2004;32(10):663–77.    
26. Marino AA, Waddell DD, Kolomytkin OV, Pruett S, Sadasivan KK, Albright JA. Assessment of immunologic mechanisms for flare reactions to Synvisc. Clin Orthop Relat Res 2006 Jan;442:187–94.   [CrossRef]   [Pubmed]