Surgical Approach

Incision

The vascular supply to the skin around the knee is rather tenious due to the lack of underlying muscle or intramuscular septum to allow for arterial perforators, and thus the skin relies on an anastamotic ring within the subcutaneous fascia. The majority of these subq arterioles originate medially: the saphenous artery and the lateral geniculate artery provide medial blood supply. 

If there are multiple skin incisions, using the lateral incision will minimize the risk of vascular compromise and wound dehiscence (because the lateral aspect of the wound will have decreased oxygenation and thus its key to not disrupt any further perfusion).

If you need to make a new incision, near an prior incision, a skin bridge of 6 cm is ideal. Avoid a skin incision that intersects prior incisions at acute angles the skin at this junction will be hypoxic.

If the skin quality is poor, if there are many prior incisions, it may be wise to consult a plastic surgeon for muscle flap coverage, or tissue expanders.

Full thickness skin flaps must be made to avoid devasularizing the skin (which is often quite thin).  The arthrotomy should be offset from the skin incision to prevent direct pathway to the external environment.

Arthrotomy

tka surgical approach medial parapatellar midvastus subvastus

Medial Parapatellar.  The goal standard for TKA arthrotomy, initially credited to von Langenbeck. Start midline over knee 2 finger-bredths above patella to 1 cm below tibial tubercle.  The VMO is detached from its insertion into the quad tendon, leaving a cuff of tendon (~5 mm) attached to the VMO proximally and a cuff of tendon attached to the medial edge of the patella distally to facilitate repair at closure.   The patella is either dislocated or dislocated + everted.  There is no definitive evidence to suggest that one technique leads to delayed recovery. 

The term for "minimally invasive" doesn't just relate to the size of the incision (although this plays a role) but most importantly, it refers to the amount of soft tissue injury.   The goal of the midvastus, subvastus discussed below is to avoid violating the quad tendon [2] A minimally invasive median parapatellar approach has also been developed which avoids violation of the quad tendon by stopping the proximal extent of the incision at the superior pole of the patella [3].  This limited arthrotomy creates a “mobile window” whereby the entire joint is not visible at once, but rather the incision must be moved during the steps of surgery to expose the necessary bone.  

Subvastus (and “mini-subvastus”).

This approach is promoted as an “extensor-mechanism preserving” arthrotomy by staying below the VMO, avoiding an incision into the quad tendon [4].   

Some studies suggest that it improves postop quad recovery and reduces pain, while other studies argue that it not only offers no short-term/long-term benefit, but also increases complications due to the reduced visualization of the knee (this approach is not recommended in knees with significant deformity or in obese patients.

Midvastus (and “mini-midvastus”). This approach is a type of compromise between the parapatellar and subvastus, as it violates less of the quad tendon, yet provides improved visualization to the subvastus.  The proximal arthrotomy splits thru the VMO muscle fibers. The standard midvastus muscle split goes to the intermuscular septum while the mini approach only violates 2-3 cm of VMO [5].

Extended Approaches (Revision).

revision tka surgical approach quad snip v-y turndown coonse adams tibial tubercle osteotomy

Quad Snip. This technique extends the standard medial parapatellar approach by creating a 45° oblique incision proximally across the quad tendon toward the vastus lateralis.   This technique is sometimes necessary in revision knees that are stiff.  At 30 months, there is no appreciable quad strength deficit compared to a standard medial parapatellar approach [6].  

V-Y Turndown (Coonse & Adams). This technique extends the Quad Snip by creating an inverted-V via an oblique incision aimed distal and lateral from the apex of the Quad Snip, which parallels the vastus lateralis.  This completely mobilizes the patella, however, it also significantly increase the risk of patellar AVN, and is associated with a 10 - 20° extensor lag. 

Tibial Tubercle Osteotomy. Expose in a stiff knee can place significant tension on the patellar tendon insertion, and risk the dreaded complication of a patellar tendon avulsion.  A controlled mobilization of the patellar tendon offers great visualization and prevents the risk of iatrogenic tendon rupture in the revision setting. The tubercle osteotomy is typically 5 cm in length, tapers distally, and hinges on the lateral anterior cortex.  It is fixed into position with cerclage wires. 


Gap Balancing & Soft Tissue Tension

One goal of TKA is to achieve balanced tension within the knee throughout range of motion.  This balanced tension is important for implant stability and longevity. A balanced knee has rectangular Flexion and Extension gaps. A balanced knee also has equal sized Flexion and Extension gaps.  

The rectangular Flexion and Extension gaps demonstrate that the medial and lateral compartment share equal tension.  The angle of bone cuts and the surrounding soft tissue tension can both affect the balance of the flexion gap (90°) and extension gap (0°).  Because the gaps are affected by the bone cut and the soft tissue tension, surgeons can adjust either to balance the gaps, and each approach represents a difference school of thought.  Some surgeons argue that bone cuts should be used to create rectangular gaps (measured resection), others argue that soft tissue releases should used to create rectangular gaps (gap balancers).  

Gap Balancing Technique. In this approach, the soft tissue tension is assumed to be correct (its not altered) and its used to determine bony cuts. The tibial cut is made first and then acts as guide to plan the cuts for the distal femur and posterior femur.   Essentially you put a 9 mm block on top of your tibial cut and draw a line across the top of it with the knee in extension.  That line that’s where your distal femoral cut will be, and that is your extension gap. Then with the knee in flexion, draw a line across the spacer block and that is where your posterior femoral cut will be, and that is your flexion gap.

Measured Resection Technique. In this approach, the soft tissue tension is not trusted and will be edited once the bony cuts are made.  The bony cuts are made based on anatomic landmarks (or navigation), then a spacer block is used to examine the flexion and extension gaps.  If the gaps are not symmetric (trapizoidal) after the bony cuts are made, and assuming the bony cuts were done properly, the imbalance is due to soft tissue changes secondary to a chronic knee deformity.  Over time, a varus or valgus deformity  will affect the soft tissue tension on the medial and lateral side. If the soft tissue tension becomes unbalanced, it needs to be corrected during surgery.  The need for ligament balancing can be evaluated pre-operatively by performing varus-valgus stress x-rays to determine whether the knee deformity is “fixed” or “flexible”.  A fixed deformity (does not correct on stress x-ray) will often require ligament balancing. 

A stepwise approach for sequential ligament releases, dictated by the type of deformity, is essential to good outcomes.

Posterior ligaments lead to tightness in extension (just bend down and touch your toes, you feel the pull posteriorly in your knee with this hyper-extension stretch). Anterior ligaments lead to tightness in flexion.

Varus deformity.  If tight in flexion release the anterior portion of the deep MCL, if tight in extension release the posterior portion of the deep MCL and capsule. If still tight release some of the superficial MCL.  If still tight posterior, release the semimembranosus.  If still tight anterior release the Pes. 

Valgus deformity. If tight in flexion, release the IT band. If still tight then release the posterior capsule and the biceps femoris tendon.  If tight in extension release the posterior lateral capsule, then the LCL, then the popliteus.


Patellar Resurfacing

The patella is a mobile fulcrum increasing the extensor mechanism's mechanical advantage at all positions of knee motion. The extensor mechanism is 50% weaker in a knee without the patella. The patella is subject to considerable forces, 5x body weight rising from a chair, 2x body weight going up stairs, and 20x body weight with jumping.  Patellofemoral pressure increases with knee flexion and peaks around 120 degrees. The native patella articular surface is not symmetric, the medial facet is small and steeply angled, while the lateral facet is large with a shallow angle.  The typical patellar implant is an all-polyethylene dome to prevent it from binding within the trochlear groove. The implant is placed along the medial border of the patella to improve tracking.   

The first knee replacements only addressed the tibio-femoral articulation.  These patients reported anterior knee pain in over 50% of cases [6].  The first patellar resurfacing options appeared in the early 1970s, yet 6% experienced lateral patellar dislocation. Changes in TKA design and improved alignment techniques have significantly reduced this risk to < 1%.  However, patella resurfacing still presents with its own unique risk of complication including fracture, mal-alignment and subluxation. Furthermore, while modern designs have reduced the incidence of anterior knee pain to about 10%, the number remains significant and thus the underlying etiology of anterior knee pain remains unclear.  [7] [8].

1. Always resurface the Patella.  

The patella is a mobile fulcrum increasing the extensor mechanism's mechanical advantage at all positions of knee motion. The extensor mechanism is 50% weaker in a knee without the patella. The patella is subject to considerable forces, 5x body weight rising from a chair, 2x body weight going up stairs, and 20x body weight with jumping.  Patellofemoral pressure increases with knee flexion and peaks around 120 degrees. The native patella articular surface is not symmetric, the medial facet is small and steeply angled, while the lateral facet is large with a shallow angle.  The typical patellar implant is an all-polyethylene dome to prevent it from binding within the trochlear groove. The implant is placed along the medial border of the patella to improve tracking.   

The first knee replacements only addressed the tibio-femoral articulation.  These patients reported anterior knee pain in over 50% of cases [1].  The first patellar resurfacing options appeared in the early 1970s, yet 6% experienced lateral patellar dislocation. Changes in TKA design and improved alignment techniques have significantly reduced this risk to < 1%.  However, patella resurfacing still presents with its own unique risk of complication including fracture, mal-alignment and subluxation. Furthermore, while modern designs have reduced the incidence of anterior knee pain to about 10%, the number remains significant and thus the underlying etiology of anterior knee pain remains unclear. [2] [3].

ALWAYS RESURFACE THE PATELLA

 

The majority of TKAs in the United States resurface the patella, however, the utility of this procedure remains controversial [4].   

Resurfacing the patella decreases the risk of revision surgery for anterior knee pain. The Australian Registry Data reports a 1.4% higher rate of revision surgery in TKA without resurfacing, and other reports cite higher rates [5] [6]], up to 8% higher [7] (the most common revision procedure being secondary resurfacing).  However, a difference in the risk of revision surgery does not necessarily mean that resurfacing reduces the risk of anterior knee pain.  The rate of revision only means that anterior knee pain in a TKA without patellar resurfacing is more likely to undergo surgery than anterior knee pain in a TKA with a patellar resurfacing (which makes sense from a practical standpoint as its easier for a surgeon to operate when there is something definite to do, ie resurface the patella, whereas, if the patella is already resurfaced, addressing anterior knee pain is often a more complicated procedure).  Anterior knee pain in TKA with or without resurfacing is 10% on average and overall there is no difference between the two [8].   

From a cost analysis standpoint, there is some evidence to support patellar resurfacing.  While there may be controversy as to why the revision rates are higher for a non-resurfaced patella, the fact remains that revision rates are higher , and the cost associated with this difference more than compensates for any “savings” obtained by not implanting a patellar button. In summary, there is cost savings for resurfacing every patella [10]

It is best to primarily resurface the patella because there are unpredictable and generally poor results associated with secondary resurfacing for painful TKA when the patella was not resurfaced at the index procedure.  Over 50% of patients report continued anterior knee pain and are dissatisfied with the procedure [11, 12].  Part of the problem is correctly diagnosing the patella was the cause of anterior knee pain.  In patients with anterior knee pain and a bone scan identifying a “hot patella”there were good results for secondary resurfacing. However, secondary resurfacing in patients with generalized knee pain and a “hot patella” showed poor results [13].  Secondary resurfacing in patients with knee pain and a “cold patella” similarly showed poor results [11].

From a biomechanical standpoint there is some evidence to support patellar resurfacing.  The patella has the thickest cartilage in the body due to the high level of forces in the patellofemoral joint.  Yet these forces increase 3x in a TKA exposing the patella to super-physiologic loads, possibly leading to acute symptoms or accelerated chondrolysis and later pain.

technical considerations. The patella is between 22-26 mm thick.  After resurfacing, a minimum bony thickness of 12 mm is necessary to avoid fracture. The goal is to replicate the native patellar thickness, ie 22-26 mm.  It is critical to obtain a flat patellar cut (commonly too much bone is resected from the medial side, increasing risk of fracture).  Best results for patellar resurfacing include: maximizing size of patellar button without overhang (this decreases crepitus) [14], error on side of increased patellar thickness (concern for "overstuffing" is not demonstrated in literature), place component slightly superiorly to avoid patella baja, slightly medialize to improve tracking [15], inlay fixation is better than onlay.  

NEVER RESURFACE THE PATELLA

There is not strong evidence to suggest that the un-resurfaced patella is a pain generator (despite the higher rates of revision surgery for un-resurfaced TKA).

On RCT examined the non-resurfaced TKA and found a low incidence of chondrolysis (7/46) that was not correlated with anterior knee pain[16].  Another study examined bilateral TKA comparing one side resurfaced versus one side non-resurfaced, and failed to demonstrate significant differences [17].  

A metaanalysis examined 13 RCT studies and found that rates of anterior knee pain were not statistically different between resurfaced and non-resurfaced TKA (10% overall) [8] and there was no difference in functional outcome scores. Among the 13 trials there were differences in outcomes.  5/13 found the resurfaced patella group had lower anterior knee pain, 1/13 found lower anterior knee pain with non-resurfaced patella, while 7/13 found no difference.  Interestingly, when the 5 studies using TKA implants with better patellofemoral kinematics (“patellar friendly”designs) were isolated, non-resurfacing lead to better functional outcomes in 4/5.  In studies of TKA designs without patellofemoral kinematic considerations “unfriendly designs”, the resurfaced patellar groups demonstrated better outcomes in 6/6 studies. Such findings suggest that TKA design has some effect on the outcome of resurfacing. 

Anterior knee pain is probably not related to resurfacing but rather patellar tracking (based on implant positioning) and ligamentous balancing. Furthermore, advances in TKA design have developed more "patella-friendly" implants to improve patellofemoral kinematics and thus decrease anterior knee pain.  Design changes include: deepening the trochlear groove and increasing conformity of patella to the trochlear groove, extending the anterior flange, adjusting the radius of curvature of femoral component to improve rollback and ligament isometry during flexion.  These changes alone have had more of an impact on anterior knee pain than any studies suggesting resurfacing. The patellar offset and lateral patellar tilt are both decreased in a resurfaced patella, which appear to alter patellofemoral pressure and kinematics, as compared to a native knee, and may be related to anterior knee pain [18].

Resurfacing the patella is not a benign procedure.  The complication rate for patellar resurfacing is between 3-7%. 

Sometimes resurface

There are also those in the middle, who selectively resurface the patella in cases with obvious patella-femoral arthritis, patella-femoral knee pain preop, or inflammatory arthropathy, but preserve the native patella when there is no obvious pathology. 

REFERENCES

1.         Ranawat, C.S., The patellofemoral joint in total condylar knee arthroplasty. Pros and cons based on five- to ten-year follow-up observations. Clin Orthop Relat Res, 1986(205): p. 93-9.

2.         Brander, V.A., et al., Predicting total knee replacement pain: a prospective, observational study. Clin Orthop Relat Res, 2003(416): p. 27-36.

3.         Elson, D.W. and I.J. Brenkel, Predicting pain after total knee arthroplasty. J Arthroplasty, 2006. 21(7): p. 1047-53.

4.         Meneghini, R.M., Should the patella be resurfaced in primary total knee arthroplasty? An evidence-based analysis. J Arthroplasty, 2008. 23(7 Suppl): p. 11-4.

5.         Kim, B.S., et al., Selective patellar nonresurfacing in total knee arthroplasty. 10 year results. Clin Orthop Relat Res, 1999(367): p. 81-8.

6.         Bourne, R.B., et al., Resurfacing versus not resurfacing the patella during total knee replacement. Clin Orthop Relat Res, 1995(321): p. 156-61.

7.         Parvizi, J., et al., Failure to resurface the patella during total knee arthroplasty may result in more knee pain and secondary surgery. Clin Orthop Relat Res, 2005. 438: p. 191-6.

8.         Pavlou, G., et al., Patellar resurfacing in total knee arthroplasty: does design matter? A meta-analysis of 7075 cases. J Bone Joint Surg Am, 2011. 93(14): p. 1301-9.

9.         Waters, T.S. and G. Bentley, Patellar resurfacing in total knee arthroplasty. A prospective, randomized study. J Bone Joint Surg Am, 2003. 85-A(2): p. 212-7.

10.       Meijer, K.A. and V. Dasa, Is resurfacing the patella cheaper? An economic analysis of evidence based medicine on patellar resurfacing. Knee, 2015. 22(2): p. 136-41.

11.       Toro-Ibarguen, A.N., et al., Secondary Patellar Resurfacing as a Rescue Procedure for Persistent Anterior Knee Pain After Primary Total Knee Arthroplasty: Do Our Patients Really Improve? J Arthroplasty, 2016. 31(7): p. 1539-43.

12.       Garcia, R.M., M.J. Kraay, and V.M. Goldberg, Isolated resurfacing of the previously unresurfaced patella total knee arthroplasty. J Arthroplasty, 2010. 25(5): p. 754-8.

13.       Ahmad, R., et al., Significance of a "hot patella" in total knee replacement without primary patellar resurfacing. Knee, 2009. 16(5): p. 337-40.

14.       Dennis, D.A., et al., The John Insall Award: control-matched evaluation of painful patellar Crepitus after total knee arthroplasty. Clin Orthop Relat Res, 2011. 469(1): p. 10-7.

15.       Anglin, C., et al., Biomechanical consequences of patellar component medialization in total knee arthroplasty. J Arthroplasty, 2010. 25(5): p. 793-802.

16.       Campbell, D.G., et al., Patellar resurfacing in total knee replacement: a ten-year randomised prospective trial. J Bone Joint Surg Br, 2006. 88(6): p. 734-9.

17.       Keblish, P.A., A.K. Varma, and A.S. Greenwald, Patellar resurfacing or retention in total knee arthroplasty. A prospective study of patients with bilateral replacements. J Bone Joint Surg Br, 1994. 76(6): p. 930-7.

18.       Fuchs, S., et al., Retropatellar contact characteristics before and after total knee arthroplasty. Knee, 2005. 12(1): p. 9-12.


Bilateral TKA


Its not uncommon for patients to present with end stage arthritis in both joints. TKA can occur simultaneous (both knees replaced at the same time), sequentially (both knees replaced in same surgery, but one after the other), or staged (one knee replaced, then rehabbed, then second knee replaced with second surgery).

There are reports that bilateral TKA (single surgery) increases mortality risk and slows rehab.

Recent studies suggest that many of the risks reported in the earlier studies may be modified with utilization of newer technology.  Computer Navigation for example allows surgeons to obtain femoral alignment without an intra-medullary guide, and thus reduce the risk of fat emboli and decrease bleeding.

Patient satisfaction with bilateral TKA is generally good, albeit variable.  One study reported that only 65% would undergo the same procedure, while another reported 95%. [15, 16]


Principles of Revision TKA


Over 50,000 revision TKA are performed in the USA every year and this number is growing as patients are living longer. Its projected to rise up to 6x by 2030. The most common reasons for revision include aseptic loosening (28%), septic loosening (14%), pain without known etiology (9%).

Approach. Careful evaluation of skin quality and prior incisions.  Principles of skin blood supply should be maintained. Arthrotomy may require a more extensile exposure if knee is stiff.  The Quad Snip or Tibial Tubercle Osteotomy effective mobilize the extensor mechanism to expose the joint.

Mobilization. The knee in extension allows for evaluation of the medial and lateral gutters, which often require fibrous debulking to give sufficient flexion and extension (be careful not to damage the collateral ligaments).  Trying to forcefully flex the knee puts the patellar tendon at risk.  Additionally, try to avoid patellar eversion due to risk of tendon avulsion (if this is a significant concern, place a pin into the medial side of the tendon as a piton).

Implant removal. A good preop plan is necessary to predict difficulties. Know the implants being removed. Predict if implants are fixed or loose, but have removal tools available regardless. There is a sequence of removal that optimizes visualization.

1. Remove poly.  Use narrow curved osteotome to disengage from the tray.

2. Remove femoral component. Removing implants without removing too much bone is the biggest challenge.  Aim the osteotome at the the implant-cement (not cement-bone) interface. If already loose, place impactor over the anterior flange.

3. Remove tibial component. Must be able to hyperflex knee to clear the posterior-lateral tibial condyle.

Re-implantation. Evaluating bone loss and choosing implant that restores stability.  Cones or sleeves are used to compensate for metaphyseal bone loss.  Augments along the distal femur or posterior femur (sizes 5, 10, 15 mm) can be used on the femoral implant to build up the areas of bone loss (cement cannot be used reliably to fill bone defects > 5 mm.  Stems should be used if augments or cones are required.  Constrained implants or hinged implants may be needed if there is ligament insufficiency.  Restoring the joint line is key to obtaining good postoperative motion.


 

Nonoperative Management

Knee arthroplasty is the recommended treatment for end-stage DJD, however knee pain begins long before this and most orthopedic surgeons will manage patients for years before recommending surgery.  There are many treatment modalities with varying levels of clinical-evidence to support or oppose its practice.  The AAOS has developed Clinical Practice Guidelines for knee osteoarthritis to help direct surgeons toward the best evidence-based care.  We will look at some of the common non-arthroplasty treatment options for patients with knee arthritis.

1. REHABILITATION

Rehabilitation involves muscle strengthening, low impact aerobic exercise, coordination training, and stretching. Remember that knee kinematics are determined by a balance of muscle strength, soft tissue tension and articular geometry. Rehab successfully improves knee kinematics and patellar tracking in particular.  5 out of 7 well conducted studies showed benefit for pain and function.  Looking various exercises, both weight-bearing and non-weight bearing exercises, isometric/isotonic/isokinetic exercises all demonstrated superior improvement over control.  The general idea is that aerobic conditioning, muscle-strengthening, regular activity and balance training are all effective treatments for older adults with arthritis [1-7]. 

A cane or walker can also provide significant relief.  It is demonstrated that assisted devices can decrease joint reactive force by 50%.

2. WEIGHT LOSS

Studies have demonstrated statistical improvement in pain and function scores, with some signs of clinically significant improvement as well.  Logically it makes sense as 1 lbs weight loss is about 5 lbs of stress off the knee. However, some studies do not show significant clinical improvement [8] [9] [10].

Diet and exercise combined show improved outcomes in function. Groen et al. suggests bariatric surgery improved knee function and pain possibly delaying the need for TKA [11].

3. GLUCOSAMINE, CHONDROITIN

Not recommended as the AAOS Clinical Practice Guidelines examined 21 studies comparing placebo to glucosamine and failed to show clinically significant improvements (only 11 of 52 outcomes showed improvement with glucosamine).  Meta-analysis did not show any significant differences.

Chondroitin showed some statistical difference but not translated into meaningful clinical difference.  The GAIT trial published in the NEJM examined 1600 patients in a RCT and showed 50% improvement with placebo after 6 months, and an 80% response rate with the Glucosamine-Chondroitin supplement [12].  

4. NSAIDS

NSAIDS are recommended with strong evidence, but make sure it is safely prescribed as chronic use can effect the gastric mucosa, platelet function, and renal function.   This risk profile depends on the type of NSAID.  There are Non-Selective NSAIDS (COX-1,2 inhibitors) and Selective COX-2 NSAIDS.

Non-Selective NSAIDS: Naproxen (Aleve), Ibuprofen (Motrin, Advil), Meloxicam (Mobic).  These have moderate GI risk, lower cardiovascular risk.  Naproxen has the lowest cardiac risk.

Selective COX-2 Inhibitors: Celecoxib (Celebrex), and Vioxx (withdrawn).  These target inflammation and pain with less risk for gastric ulceration.  Large database analysis suggests that chronic use of these increases risk for heart attach and stroke (the theory is that blocking COX-2 in blood vessels decreases prostacyclin, which is important for preventing platelet aggregation and vasoconstriction).

Tylenol.  Not a true NSAID, as it works in the central nervous system, and does not have a direct effect on inflammation.  Tylenol appears to have a better side effect profile than NSAIDS, yet there isn’t good evidence to suggest equivalent (or any) anti-inflammatory properties.  

5. ACETAMINOPHEN

Not a true NSAID, as it works in the central nervous system, and does not have a direct effect on inflammation.  Tylenol appears to have a better side effect profile than NSAIDS, yet there isn’t good evidence to suggest equivalent (or any) anti-inflammatory properties [14]. 

One RCT comparing acetaminophen to placebo found 50% improvement in both groups, although the acetaminophen group was limited to 400 mg daily (the current accepted daily max is 3000 mg).

6. INJECTIONS

Steroid. There are few placebo controlled studies for intraarticular steroid injections [15] [16, 17].  One study found statistical but not clinical improvement at 4 weeks on WOMAC but no difference by 12 weeks [18].  Other studies failed to find significant evidence for improvement compared with needle lavage or viscosupplement. Another important consideration is the non-medical purpose of such injections.  Many surgeons use injections as a means to get to know their patients, a way to offer them a service without suggesting surgery upon their first meeting.  Some epidemiologic data supports the use of injections for such purposes.  There is a wide variation in the rates of injection among Medicare patients, from 1 in 115 (high rate) to 1 in 1000 [19].  The highest rates of injection correlated with the areas with the highest density of doctors (not orthopedic surgeons specifically), with rates of injection increasing 3x over the past 15 years, suggesting there are economic or social forces beyond medical need affecting utilization.  Orthopedic surgeons administer about 59% of injections.  While certainly less expensive then performing a knee arthroplasty, injections do come at a cost: $69 per injection and $80,000,000 per year nationally [20].   

Viscosupplement. The AAOS strong recommends “against” the use of viscosupplements based on a number of RCT.  While some studies demonstrated statistical improvements, no gains were proven to be clinically relevant [15].  Similar to the steroid injections, roughly 50% of patients with mild to moderate OA report satisfaction with the injection (yet no difference from placebo).  Its therefore important to recognize the placebo effect of these medications, patients improve, but arguably the high cost of the viscosupplement is not justified (average injection costs $180, almost 3x more than steroid).  Interestingly a review paper compared placebo trials and studies without a placebo (“open-label” study) and found significantly less impressive results with the placebo trials [16]. 

There is unclear benefit tied directly to the type of injection, although injections overall appear to help many patients, in part from the placebo effect.  So if it helps some people, for whatever reason, is there any harm in giving it, even if it only helps 50% of people, for a few weeks at most?  There are studies that suggest that recent injections can increase the risk for periprosthetic joint infection in TKA if given within 3 months of surgery. Cost is another.  Annual cost of HA injections is about $207 million. 

There are many brands for the viscosupplmentation, but they are all hyaluronic acid.

Orthovisc = hyaluronan

Synvisc = hylan G-F 20.  This somes as either a single shot (Synvisc-One) or three shots given over the course of a month.

Hyalgan = hyaluronate

Fermathron plus - produced from the bacterium Streptococcus equi 

7. ARTHROSCOPY

There is strong evidence against the use of arthroscopy for the treatment of OA.

Moseley et al is a well-known study published in the New England Journal of medicine that randomized patients to scope or no scope and followed them from 2 weeks to 2 years and found no improved pain or function [23].  While the study received public recognition due to its impressive attention to detail: placebo patients received real anesthesia and a skin incision, water was splashed on the floor and a simulated debridement was played on a monitor while no arthroscope was actually inserted...the overall study design was more flawed: 44% of patients declined to participate and those that did participate were likely to be white, younger and have more severe arthritis, suggesting potential bias. 

Similar RCT have also evaluated the efficacy of arthroscopic debridement for OA and failed to identify significant benefit over PT or medical management [24].  A meta-analysis found some signs of pain relieve at 3 or 6 months, however, this change was not clinically significant [25]. 

When arthroscopy was performed within 6 months of TKA, rates of infection (and stiffness) increased, emphasizing the outcomes from knee injections, which suggest that you don’t want to be messing around with the knee before implanting metal [26].  


REFERENCES

1.         Deyle, G.D., et al., Effectiveness of manual physical therapy and exercise in osteoarthritis of the knee. A randomized, controlled trial. Ann Intern Med, 2000. 132(3): p. 173-81.

2.         Jan, M.H., et al., Investigation of clinical effects of high- and low-resistance training for patients with knee osteoarthritis: a randomized controlled trial. Phys Ther, 2008. 88(4): p. 427-36.

3.         Silva, L.E., et al., Hydrotherapy versus conventional land-based exercise for the management of patients with osteoarthritis of the knee: a randomized clinical trial. Phys Ther, 2008. 88(1): p. 12-21.

4.         O'Reilly, S.C., K.R. Muir, and M. Doherty, Effectiveness of home exercise on pain and disability from osteoarthritis of the knee: a randomised controlled trial. Ann Rheum Dis, 1999. 58(1): p. 15-9.

5.         Lin, D.H., et al., Efficacy of 2 non-weight-bearing interventions, proprioception training versus strength training, for patients with knee osteoarthritis: a randomized clinical trial. J Orthop Sports Phys Ther, 2009. 39(6): p. 450-7.

6.         Fitzgerald, G.K., et al., Agility and perturbation training techniques in exercise therapy for reducing pain and improving function in people with knee osteoarthritis: a randomized clinical trial. Phys Ther, 2011. 91(4): p. 452-69.

7.         Kovar, P.A., et al., Supervised fitness walking in patients with osteoarthritis of the knee. A randomized, controlled trial. Ann Intern Med, 1992. 116(7): p. 529-34.

8.         Miller, G.D., et al., Intensive weight loss program improves physical function in older obese adults with knee osteoarthritis. Obesity (Silver Spring), 2006. 14(7): p. 1219-30.

9.         Focht, B.C., et al., Exercise, self-efficacy, and mobility performance in overweight and obese older adults with knee osteoarthritis. Arthritis Rheum, 2005. 53(5): p. 659-65.

10.       Rejeski, W.J., et al., Obese, older adults with knee osteoarthritis: weight loss, exercise, and quality of life. Health Psychol, 2002. 21(5): p. 419-26.

11.       Groen, V.A., et al., Effects of bariatric surgery for knee complaints in (morbidly) obese adult patients: a systematic review. Obes Rev, 2015. 16(2): p. 161-70.

12.       Clegg, D.O., et al., Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. N Engl J Med, 2006. 354(8): p. 795-808.

13.       Scarpignato, C., et al., Safe prescribing of non-steroidal anti-inflammatory drugs in patients with osteoarthritis--an expert consensus addressing benefits as well as gastrointestinal and cardiovascular risks. BMC Med, 2015. 13: p. 55.

14.       Miceli-Richard, C., et al., Paracetamol in osteoarthritis of the knee. Ann Rheum Dis, 2004. 63(8): p. 923-30.

15.       van der Weegen, W., et al., No difference between intra-articular injection of hyaluronic acid and placebo for mild to moderate knee osteoarthritis: a randomized, controlled, double-blind trial. J Arthroplasty, 2015. 30(5): p. 754-7.

16.       Jevsevar, D., et al., Viscosupplementation for Osteoarthritis of the Knee: A Systematic Review of the Evidence. J Bone Joint Surg Am, 2015. 97(24): p. 2047-60.

17.       Raynauld, J.P., et al., Safety and efficacy of long-term intraarticular steroid injections in osteoarthritis of the knee: a randomized, double-blind, placebo-controlled trial. Arthritis Rheum, 2003. 48(2): p. 370-7.

18.       Jones, A. and M. Doherty, Intra-articular corticosteroids are effective in osteoarthritis but there are no clinical predictors of response. Ann Rheum Dis, 1996. 55(11): p. 829-32.

19.       Gaffney, K., J. Ledingham, and J.D. Perry, Intra-articular triamcinolone hexacetonide in knee osteoarthritis: factors influencing the clinical response. Ann Rheum Dis, 1995. 54(5): p. 379-81.

20.       Chao, J., et al., Inflammatory characteristics on ultrasound predict poorer longterm response to intraarticular corticosteroid injections in knee osteoarthritis. J Rheumatol, 2010. 37(3): p. 650-5.

21.       Schmajuk, G., K.J. Bozic, and J. Yazdany, Using Medicare data to understand low-value health care: the case of intra-articular hyaluronic acid injections. JAMA Intern Med, 2014. 174(10): p. 1702-4.

22.       Koenig, K.M., et al., The Use of Hyaluronic Acid and Corticosteroid Injections Among Medicare Patients With Knee Osteoarthritis. J Arthroplasty, 2016. 31(2): p. 351-5.

23.       Moseley, J.B., et al., A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med, 2002. 347(2): p. 81-8.

24.       Kirkley, A., et al., A randomized trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med, 2008. 359(11): p. 1097-107.

25.       Thorlund, J.B., et al., Arthroscopic surgery for degenerative knee: systematic review and meta-analysis of benefits and harms. Br J Sports Med, 2015. 49(19): p. 1229-35.

26.       Werner, B.C., et al., Total Knee Arthroplasty Within Six Months After Knee Arthroscopy Is Associated With Increased Postoperative Complications. J Arthroplasty, 2015. 30(8): p. 1313-6.