Effect of various factors on articular cartilage and their implications on arthroscopic procedures: A review of literature

1. Introduction

Arthroscopy is one of the most common orthopaedic procedures performed worldwide. It is reported that in 2006, nearly a million arthroscopies of only the knee joint were performed in the United States.¹ As arthroscopic interventions are minimally invasive, patients undergoing these procedures experience lesser postoperative pain, have quicker recovery, faster rehabilitation and better functional outcomes. Shoulder joint arthroscopy has probably attained equal acceptance among orthopaedic surgeons and it is estimated that more than half a million arthroscopic procedures for shoulder pathology including impingement are performed annually in the US alone.² Hip, ankle, elbow, and wrist joints are the other common sites for arthroscopic procedures.

Though the articular cartilage of the synovial joints have a histological structure tailored to resist the damage due to physiological stresses of day to day activities, it is now known to be vulnerable for damage during arthroscopic procedures. As the arthroscopic procedures are performed quite frequently in active young adults in second and third decades of their life, it will be of great importance to know ways and means to prevent or reduce the chances of articular cartilage damage during these procedures. Though there is a lag up to 5 years from articular cartilage damage to development of osteoarthritis the association is predictable with certain injury patterns.³ Post arthroscopic chondrolysis has been reported most often in the gleno-humeral joint but has also been documented in the knee joint 4, 5.

Many factors that can cause articular damage during arthroscopic procedures were studied like local anesthetics, irrigation fluid, the duration of exposure to the fluids and temperature of the fluids, radiofrequency device, and the level of training of the surgeon. The purpose of this article is to systematically review the current evidence available about each of the above factors which can lead to articular cartilage damage.

2. Materials and methods

The review was based on the article search performed in MEDLINE and EMBASE databases. The search strategy used combinations of the following keywords Arthroscopy, Cartilage, local anesthetic, irrigation fluid, duration of exposure, temperature, pain pump, learning curve, iatrogenic injury and chondrolysis. All human studies, animal studies and clinical investigations performed to assess the adverse effects of different factors on articular cartilage were included. Letters to editor, systematic reviews abstracts in language other than English were excluded. Studies quoted in systematic reviews were included.

The articles were grouped into clinical cases and laboratory studies (including animal studies). Articles published and available online from 1960 until April 2019 were included on both the databases. Studies mainly looking into effects of local anesthetic, corticosteroids, pain pump, irrigation fluid and its temperature, effect of radiofrequency energy, damage secondary to suture anchors and Knots facing articular surfaces and iatrogenic injury due to sharp instruments and surgeon factors were reviewed. The controlled trials register was searched for any relevant randomized controlled trials and were included to the articles.

3. Results

A total of 1639 articles were identified using the two databases. Manual searching of references revealed 11 articles further (Fig. 1). After removal of duplicates and screening the articles a total of 76 articles were included for the review.

Fig. 1.

Flow chart demonstrating the study selection process used for the systematic review. tag caption.

3.1. Effect of different irrigation fluids

Various laboratory studies have been performed assessing the effect of commonly used irrigation fluids on articular cartilage. Arciero (Arciero, Little et al., 1986)⁶ measured proteoglycan synthesis with normal saline, Ringer's solution, sterile water or an un-irrigated control group and found no difference between the three solutions. Bert et al. (Bert, Posalaky et al., 1990)⁷ found that specimens exposed to 1.25% Glycine had the most smooth appearance while those exposed to saline and Ringer's lactate showed fibrillation and ridges on scanning electron microscopy.

Reagan et al. (Reagan, McInerny et al., 1983)⁸ compared ³⁵SO₄ uptake by bovine articular cartilage specimens and found that the uptake was higher with Ringer's lactate compared to normal saline. They felt that the acidic pH (5.3) of the saline solution used in their study could be responsible for this difference in uptake.

Bulstra et al. (Bulstra, Kuijer et al., 1994)⁹ and Gulihar et al. (Gulihar et al., 2012)¹⁰ concluded that all irrigation solutions caused some degree of inhibition of proteoglycan metabolism but this was least with Ringers solution. Jurvelin et al. (Jurvelin, Jurvelin et al., 1994)¹¹ measured instant, total and creep deformation in bovine articular cartilage after applying an indenting force, Gradinger et al. (Gradinger, Trager et al., 1995)¹² measured proteoglycan loss from bovine cartilage and they both concluded that nonionic solutions like Mannitol probably are more safer.

3.2. Effect of duration of exposure

The duration of exposure to an irrigation fluid may have an effect on the metabolism of articular cartilage. Duration of arthroscopy depends on the nature of the procedure and is usually shorter for diagnostic purposes than for therapeutic procedures. Jurvelin et al. (Jurvelin, Jurvelin et al., 1994)¹¹ observed that creep deformation of articular cartilage increased as the immersion time in Ringer's solution was increased from 2 h to 20 h. However, Yang et al. (Yang, Cheng et al., 1993)¹³ studied effects of normal saline, Ringer's lactate, distilled water, and 3% sorbitol on articular cartilage of knee joints using an animal model with a scanning electron microscope (SEM) and observed similar surface ultrastructure at 1 and 2 h after exposure to these four different fluids.

3.3. Effect of temperature

Some authors have also studied the effect of temperature of the irrigation fluid. Cheng et al. (Cheng, Jou et al., 2004)¹⁴ irrigated rat articular cartilage with normal saline and observed that irrigation at 37 °C caused less damage on SEM to articular cartilage than at 4 °C. Brand et al. (Brand, De Koning et al., 1991)¹⁵ incubated bovine articular cartilage explants at two different temperatures and found that the rate of proteoglycan synthesis and the release of newly synthesized proteoglycans were decreased in cultures incubated at 32 °C compared to 37 °C.

3.4. Effect of local anaesthetics and pain pump

Local anaesthetics are often used post operatively as intra-articular injections or infused via pain pumps for post-operative pain relief. The potentially harmful effects of Bupivacaine on articular cartilage were first reported by Dogan et al. (Dogan, Erdem et al., 2004)¹⁶ who injected saline or 0.5% Bupivacaine into the knee joints of rabbits and examined cartilage specimens histo-pathologically after 24 h, 48 h or 10 days. They found that the specimens from animals that received Bupivacaine showed significantly more inflammatory changes in the form of inflammatory cell infiltration and synovial membrane cell hyperplasia and hypertrophy compared to saline control.

Piper and Kim (Piper and Kim 2008)¹⁷ harvested cartilage explants as well as chondrocyte cultures from the femoral heads or tibial plateau of five patients and exposed them to 0.9% saline control, 0.5% Bupivacaine or 0.5% Ropivacaine for 30 min. The explants and chondrocytes were analysed using live/dead cell viability analysis after 24 h. They found that Bupivacaine had significantly lower cell viability at 78% in the explants and 37% in the cell cultures compared to 94% and 64% for Ropivacaine. They found no significant difference between Ropivacaine and saline control in the explants. However, we believe that saline was a poor choice of control because it has been shown to inhibit proteoglycan metabolism in in-vitro studies (Bulstra, Kuijer et al., 1994, Gulihar, Bryson et al., 2012).^9,10 The use of a cartilage culture medium instead probably would have shown more toxicity with Ropivacaine.

Farkas et al. (Farkas, Kvell et al., 2010)¹⁸ found that even Ropivacaine can be harmful to articular cartilage when used at a higher concentration. They found significantly increased number of dead/ necrotic cells with all anaesthetics and this toxicity worsened when Betamethasone was added or when the duration of exposure was increased from 2 h to 6 h or 24 h. Studies performed by Park et al. (Park, Sutradhar et al., 2011)¹⁹ and Bolt et al. (Bolt, Ishihara et al., 2008)²⁰ on equine articular cartilage comparing toxicities of different local anaesthetics suggested that Mepivacaine was probably least toxic among them.

Some of the laboratory studies have not found significant toxicity with local anaesthetics on their own and have tested the effect of epinephrine or preservatives used in local anesthetic solutions. Dragoo et al. (Dragoo, Korotkova et al., 2008)²¹ found that at all time periods, solutions containing Epinephrine were more toxic than control. Syed et al. (Syed, Green et al., 2011)²² suggested that the low pH of local anaesthetics was unlikely to be a contributor to toxicity. The mechanism of chondrocyte toxicity at a molecular level has been studied by Grishko et al. (Grishko, Xu et al., 2010)²³, Mitochondrial dysfunction in the form of mitochondrial DNA damage, decrease in ATP and mitochondrial protein levels was seen with all local anaesthetics.

4. Clinical cases

Chondrolysis was established with clinical symptoms of decreasing range of movements of the joint, with increasing pain, associated crepitus and radiological findings consistent with the diagnosis. The characteristic radiographic findings include joint-space narrowing, periarticular bone erosion, subchondral cysts, and lack of osteophyte formation. On MRI diffuse loss of the articular cartilage on both the glenoid and humeral head surfaces, with cortical irregularity and patchy areas that show a change in signal intensity in subchondral marrow that is consistent with subchondral sclerosis and marrow edema are typically noted. In patients who undergo subsequent arthroscopic or open procedures diffuse loss of articular cartilage with associated synovitis, thickening of capsule and inflammatory changes were reported.

A total of 167 cases of chondrolysis in 163 patients associated with intra-articular local anesthetic infusion or infiltration are reported in literature. Of these, 149 (89.3%) cases involved the Gleno-humeral joint. Other joints involved included the knee (16 cases) and one case each involving the ankle and elbow joints. 163 (97.7%) cases involved continuous intra-articular infusion via a pain pump. In some ways chondrolysis following exposure of articular cartilage to local anesthetic appears to be an idiosyncratic response. However there may be a dose response-element. Anderson (Anderson, Buchko et al., 2010)²⁴ found that chondrolysis developed in 16 of 32 (50%) patients who received high-flow (5 ml/h) intra-articular Bupivacaine injection but only in 2 of 12 (17%) who received low flow (2 ml/h) infusion. Rapley et al. (Rapley, Beavis et al., 2009)²⁵ found that chondrolysis did not develop in 13 cases where 100 ml of 0.5% Bupivacaine was infused at 2.08 ml/h but developed in 3 out of 16 cases where 270 ml of 0.5% Bupivacaine was infused at 4.16 ml/h.

Wiater et al. (Wiater, Neradilek et al., 2011)²⁶ followed up a single surgeon case series of 365 patients. They found that chondrolysis only occurred in 49 patients out of 109 who all had local anesthetic pain pumps. No chondrolysis was observed in the remaining 256 patients who did not have a pain pump but had other described risk factors such as suture anchors, thermal and radiofrequency devices. They concluded that pain pumps were the single most important risk factor for chondrolysis in their series of 365 patients.

4.1. Effect of intra-articular corticosteroids

Steroid injection post arthroscopy though not a standard practice is practiced by some surgeons intraoperatively for better postoperative pain control and functional outcome. While this is mainly due to their anti-inflammatory action, there also is evidence of Corticosteroids having a protective effect on chondrocytes and articular cartilage. Pelletier et al. (Pelletier, Martel-Pelletier et al., 1987)²⁷ found that chondral specimens from patients with knee osteoarthritis, who had received intra-articular steroids, had lower level of proteoglycan degrading Matrix metalloproteinases (MMPs) compared to those who did not receive steroids. In a subsequent study, they induced osteoarthritis in dogs by transecting the ACL and manifested by the development of femoral cartilage erosions (Pelletier and Martel-Pelletier 1989).²⁸ Dogs that were given an intra-articular injection of Triamcinolone did not develop any lesions and did not show any evidence of cell death or degeneration. In a similar model, they further demonstrated that Methylprednisolone reduced osteophyte formation and histological changes of OA and also suppressed the synthesis of the metalloproteinase stromelysin (Pelletier, Mineau et al., 1994).²⁹ Triamcinolone also reduced fibrillation, osteophyte formation, histological abnormalities and chondrocyte loss in knees of guinea pigs injected with sodium iodoacetate (Williams and Brandt 1985).³⁰

Corticosteroids also seem to have a protective effect on non-arthritic articular cartilage. Wang et al. (Wang, Elewaut et al., 2004)³¹ investigated the effect of Hydrocortisone on chondrocytes obtained from femoral condyles of five donors. They found that levels of aggrecan, type II collagen and fibronectin were increased in the steroid treated cells and ILα and β were inhibited. Koyonos et al. (2009)³² performed a randomized double blinded control trial on 58 patients (59 knees) where either lidocaine mixed with saline or lidocaine mixed with methyl prednisolone were administered into the knee after closing the arthroscopic portals and found that postoperative corticosteroid injection resulted in improved pain and function at an early time point; however, it provided no lasting difference compared with only local anesthetic injection.

4.2. Effect of Surgeon's experience and level of training

Arthroscopic procedures unlike open procedures are technically demanding and require specialized training to acquire the hand eye coordination and to learn a safe standard technique. As the procedure involves using needles, scalpel blades, trochars, motorised shavers, arthroscopic instruments, cutters, drills, screws, etc., there is a risk of iatrogenic articular cartilage injury (iACI). Cannon et al.³³ in their study involving six year one post graduate residents, six year five residents and six community orthopaedic surgeons evaluated mean time to perform a standard diagnostic knee arthroscopy on ArthroSim virtual reality arthroscopy simulator in each of the three groups to identify that both the postgraduate year-5 residents and the community-based orthopaedic surgeons performed the procedure in significantly less time (p = 0.006) than the postgraduate year-1 residents. There was a trend toward significance (p = 0.055) in time to complete the procedure when the postgraduate year-5 residents were compared with the community-based orthopaedic surgeons.

The risk of iACI is greater with hip and ankle arthroscopy due to more tighter articular compartment and the unique anatomy making the access to pathological areas like central compartment in hip and talar dome difficult thus the articular cartilage becoming more vulnerable to mechanical trauma. Mehta et al.³⁴ from HSS followed 8041 hip arthroscopies performed by 251 surgeons to find that at 981 (12.3%) had to undergo additional hip surgery within 5 years. They concluded that cases performed by surgeons with career.

Volumes >/ = 519 had significantly lower risk of subsequent hip surgery than those performed by lower-volume surgeons. Kautzner et al.³⁵ evaluated 150 hip arthroscopies performed by single surgeon and concluded that at least 100 procedures are needed to gain the basic technical skills. Vega et al.³⁶ reviewed 74 ankle arthroscopies to find that iatrogenic cartilage injuries happened in 23 instances of which 5 injuries were deep and clinically significant. DiGiovanni³⁷ in their cadaveric experiment with 2.7 mm rigid 30° scope in comparison with flexible bronchoscopes concluded that flexible bronchoscope provided access to posterior compartment of the ankle as well from the ankle without causing any articular cartilage damage offering a potential solution to the problem.

4.3. Effect of radiofrequency energy

Radiofrequency energy was introduced in 1990s to replace other mechanical devices to treat cartilage lesions to produce a compact scar which is more stable with the aim of being safe and least harmful to chondrocytes. The evidence however is discordant.

Barber et al. (Barber et al., 2006)³⁸ compared RFE and mechanical debridement in grade III outerbridge lesions and found comparable IKDS and Lysholm scores. Owens et al.³⁹ performed a similar study on patellar chondral lesions, however this study found that bipolar radio frequency group had a clear superior Patellofemoral Joint Evaluation Score than the mechanical shaver group P = 0.0006. Spahn et al.⁴⁰ evaluated outcomes of bipolar RFE (bRFE) group vs mechanical debridement group at 1 year and 4 year and showed significantly better outcomes in the bRFE group.

Voloshin et al.⁴¹ in his series reported that patients who underwent RFE chondroplasty the chances of progression of lesion is less and is more like to result in partial or complete filling of the cartilage defect. Osti et al.⁴² in their RCT compared RFE with micro fracture method to note that there was a statistically significant difference between the groups (P < 0.001) in WOMAC scores.

4.4. Effect of suture anchors and knots

Chondrolysis was reported by McCarty et al.⁴³ following use of poly-l-lactic acid (PLLA) implants. In this series of 44 patients apart from chondrolysis, anchor failure, osteolysis and glenohumeral synovitis was reported. Mc Nickle et al.⁴⁴ reported 2 out of 20 cases of glenohumeral Chondrolysis was probably due to proud or suboptimal placement of suture anchor. Bailie et al.⁴⁵ reported that in their series of 23 cases of Chondrolysis, 14 patients had bio absorbable device used for shoulder stabilization. They reported loosening of suture anchors resulted in prominent anchors and knots and ultimately leading to severe chondral injury to the humeral head. It is unclear if this is related to any specific type of suture anchor, however there are few reports indicating higher incidence with knotless suture anchors.^46,47

In a study on Rabbit model Magnum Wire and Ticron sutures stimulate a more intense foreign body inflammatory response and can at least lead to synergistic effect in inducing chondrocyte damage. ⁴⁸ Kelly et al.⁴⁹ reported aseptic synovitis and chondral injury after usage of outside-in Mulberry knot repair of the medial meniscus. The presence of synovitis was diagnosed clinically and by multiple arthrocentesis and confirmed, along with a chondral lesion, by repeat arthroscopy. Meniscal repair with arrow technique is reported to lead to complications like such as inflammatory foreign-body reaction, cartilage lesions, and arrow displacement. Seibold et al.⁵⁰ reported a high incidence of failure rate (81.5%) with this device in a series of 113 patients. They reported chondral damage due to prominent arrow head during revision procedures.

5. Discussion

Arthroscopic surgical procedures are increasingly becoming more frequent especially in younger patients. The articular cartilage of the synovial joints is vulnerable to damage from various insults. Though the resulting chondral damage may not be immediate there can be significant lag in its presentation. It is important for the treating surgeon to be aware of these agents which can be potentially be addressed by changes in technique etc. This is very relevant if the patient is young and has intact articular cartilage.

There is currently no clinical evidence to suggest that any of the irrigation fluids are safer or produce better outcomes than other fluids. In vitro studies seem to suggest that non-ionic fluids are safer than ionic fluids. The effect of duration of exposure, pH and osmolarity of the fluid is not clear but using fluid at 37 °C seems to be safer than that at room temperature.

Chondrolysis is a devastating complication of arthroscopic surgery especially in young patients. Almost all patients require further surgery and a large proportion of them will eventually end up having arthroplasty. Pain pumps are associated with high incidence of chondrolysis and should be avoided. Almost all of the commonly used local anesthetic agents have been found to be chondrotoxic in laboratory studies even with a single exposure. However, there are only four documented cases of Chondrolysis following a single intra-articular injection of local anesthetic. The long-term effects of a single exposure are therefore unknown.

Problems with suture anchors can be avoided by ensuring optimal placement and avoiding the devices with known higher failure rates. Placing knots away from the articular cartilage may also help prevent articular cartilage damage from any ensuing tissue reaction.

Most of the documented cases of Chondrolysis had multiple risk factors such as irrigation fluid, radio-frequency devices, local anesthetics etc. The individual effects of any one of these factors are therefore difficult to define. Further clinical studies especially in the form of randomized clinical trials will help to delineate the exact role of different factors in the development of chondrolysis.

6. Conclusion

To summarize Ringer's solution was the least toxic arthroscopic irrigation fluid and should replace normal saline in clinical practice. Injecting local anesthetics into joints needs careful consideration of risks and benefits and should not be routine practice post arthroscopy and pain pumps must be avoided. More clinical studies are required to assess whether any real damage occurs to a joint injected with local anesthetic or irrigated with normal saline. If a clinical problem can be identified then the protective effect of Glucosamine or Corticosteroid could be assessed. Non-invasive and highly accurate imaging modalities like MRI is recommended over a routine arthroscopy especially in the young individuals to establish the diagnosis. Orthopaedic surgeons should consider small and inconsequential benefits of arthroscopic debridement against the risks associated in degenerative osteoarthritis where probably physical therapy probably could provide equal or greater benefit.

Conflicts of interest

We, the authors have no conflict of interest to declare.

Appendix A. Supplementary data

The following is the Supplementary data to this article: